Engineering_and_Operations_in_the_Bell_System_2ed_1984 Engineering And Operations In The Bell System 2ed 1984

User Manual: Engineering_and_Operations_in_the_Bell_System_2ed_1984

Open the PDF directly: View PDF PDF.
Page Count: 902

DownloadEngineering_and_Operations_in_the_Bell_System_2ed_1984 Engineering And Operations In The Bell System 2ed 1984
Open PDF In BrowserView PDF
ENGiNEERiNG ANd
OPERATioNS iN
TilE BELL SySTEM

ENGINEERING
and

OPERATIONS

BELt

SYSTEM

ENGINEERING
OPERATIONS

B~nLt

SYSTEM
Second Edition
Reorganized and Rewritten
Telecommunications in the
Bell System in 1982 - 1983

Prepared by Members of the Technical Staff
and the Technical Publication Department
AT&T Bell Laboratories
RJ. Rey, Technical Editor

AT&T Bell Laboratories
Murrav Hill. N.J. '

Copyright© 1977, 1983 by Bell Telephone Laboratories, Incorporated
All rights reserved
This second edition differs significantly from that of 1977. It
contains revised and updated information, new entries and
illustrations, and a different organization and format.

Second Printing 1984

International Standard Book Number 0-932764-04-5
Library of Congress Catalog Card Number 83-72956
AT&T Customer Information Center Select Code 500-478
No part of this publication may be reproduced or transmitted in
any form or by any means, electronic or mechanical - including
photocopy, recording, or any information storage and retrieval
system - without permission in writing from the publisher.

Printed in the United States of America

Foreword

The first edition of this book, published in 1977, presented a
comprehensive view of the Bell System as seen from AT&T Bell Laboratories. Its primary purpose was to provide new members of technical
staff of Bell Laboratories with a basic understanding of the Bell System
from the standpoint of the services provided, the equipment and systems
constituting the nationwide network, the planning and engineering considerations involved in the evolution of the network, and the many
activities required for day-to-day operation. However, the book has been
used more widely than expected as a general reference and as a primer
for others unfamiliar with telecommunications.
This revision was prompted by the significant changes in technology,
services, and the environment that have occurred since 1977. Consequently, material from many sections of the first edition has been
updated, and the organization and presentation of material have been
improved. The level of detail has been reduced in some places and more
emphasis placed on explaining important concepts and defining terminology. In addition, an attempt was made to present the material in a
manner suitable for a mix of academic backgrounds.
The material for this edition was almost complete when agreement
was reached between AT&T and the United States Department of Justice
to settle an antitrust suit by divesting the Bell operating companies from
AT&T. Recognizing that it would be a long time before the massive
change associated with divestiture could be recorded and that most of the
information on services and systems would remain valid, revision of the
first edition continued as planned. Only the first chapter was revised to
indicate the major provisions related to divestiture and to provide an
overview of the postdivestiture corporate units. The rest of the book portrays the Bell System as it was near the end of 1982 and early 1983 and
reflects changes resulting from the Federal Communications
Commission's Computer Inquiry II order that were effective January 1,
1983. Since the existence of the Bell System ends with divestiture, this
second edition of Engineering and Operations in the Bell System is also the

v

vi

Foreword

final edition. It is perhaps a fitting closure to an era in which the Bell
System fulfilled its historic mission of providing universal high-quality,
economical telephone service and also provided much of the technology
for the future.
The book is divided into five parts. Part 1 presents an overview of the
Bell System (and the postdivestiture configuration) in terms of the corporate units and their responsibilities, the services provided, and basic
communications via the network. Part 2 deals with concepts, principles,
and engineering considerations related to the components of telecommunications. In part 3, specific systems and equipment are described
with emphasis on applications, distinguishing characteristics and features,
and major design considerations. Part 4 describes telephone company
operations on a functional level, presents a description of selected
computer-based operations systems, and discusses operations planning
and the evaluation of performance and services. Finally, Part 5 traces the
events that have shaped the telecommunications environment and
discusses factors related to the evolution of products and services.
Authors and coordinators of the material are identified at the end of
each chapter. A committee consisting of R. A. Bruce, L. E. Gallaher, W. S.
Hayward, Jr., J. R. Harris, M. M. Irvine, and J. A. McCarthy reviewed the
plan for the book and each chapter, providing many corrections and comments. J. W. Falk and D. E. Snedeker reviewed the material from a legal
and regulatory viewpoint. J. R. Harris deserves special recognition for
his contributions to the planning and early development of the book and
for his guidance and support throughout the project. The Technical Publication Department of Bell Laboratories provided editorial assistance
throughout the project and prepared the material for publication. Many
others assisted by reviewing portions of the text or serving as information sources. The principal credit, however, belongs to the editor, R. F.
Rey, whose technical and managerial talents were wisely used in this
undertaking.

Tom L. Powers
Executive Director
Network Planning Division
Bell Laboratories

Contents

Foreword

v

PART ONE
INTRODUCTION TO THE BELL SYSTEM
1 Structure and Activities

1

3

1.1 Introduction
3
4
1.2 The Bell System in 1982
4
1.2.1 American Telephone and Telegraph Company
11
1.2.2 Bell Operating Companies
1.2.3 Western Electric
14
20
1.2.4 Bell Laboratories
1.2.5 Relationships with Non-Bell System Companies
27
28
1.2.6 Resources and Volume of Service
29
1.3 The Postdivestiture View
1.3.1 Summary of Modification of Final Judgment Provisions
1.3.2 The New AT&T
32
34
1.3.3 The Divested Operating Companies

2 Services

29

39

2.1 Introduction
39
39
2.1.1 Basic and Vertical Services
40
2.1.2 Ordinary and Special Services
2.2 Terminal Products
41
2.2.1 Telephone Sets and Vertical Services
41
2.2.2 Modular Telephones and Retail Sales
43
43
2.2.3 Data Products
47
2.2.4 Special-Purpose Terminals
2.2.5 Aids for the Handicapped
49
52
2.3 Customer Switching Services
52
2.3.1 Customer Needs
2.3.2 Small-Size Business Customers
53
54
2.3.3 Medium-Size Business Customers

vii

viii

Contents

2.4

2.5

2.6
2.7

2.3.4 Large-Size Business Customers
55
2.3.5 Special Customers
55
56
Exchange Services
56
2.4.1 Exchange Lines and Local Calling
2.4.2 Custom Calling Services I
57
2.4.3 TOUCH-TONE@ Service
58
2.4.4 Exchange Business Services
58
60
2.4.5 911 Emergency Service
Network Services
60
2.5.1 Public Switched Telephone Network Services
2.5.2 Private-Line Services
66
2.5.3 Private N~twork Services
66
2.5.4 Data Services
70
2.5.5 Mobile Telephone Services
73
2.5.6 Video Teleconferencing Service
75
Public Communications Services
76
Customer Support Services
78
79
2.7.1 Retail Sales and Service
79
2.7.2 Business Office Services
2.7.3 Installation and Maintenance Services
79
2.7.4 Directory Services
80

3 Introduction to the Network

61

81

3.1 What is a Telecommunications Network?
81
3.2 The Facilities Network
84
3.2.1 Station Equipment
85
3.2.2 Transmission Facilities
85
87
3.2.3 Switching Systems
3.3 Traffic Networks
88
3.3.1 Public Switched Telephone Network
88
3.3.2 Private-Line Voice Networks
89
3.3.3 Private-Line Data Networks
91
3.3.4 Program Networks
92
3.4 A Typical Telephone Call
92
3.4.1 Setting the Stage
92
3.4.2 Initiating the Call
92
3.4.3 Call Processing at the Originating Central Office
94
3.4.4 Call Advancement to the Terminating Central Office
94
97
3.4.5 Call Completion

PART TWO
NETWORK AND SYSTEMS CONSIDERATIONS
4 Network Structures and Planning

103

4.1 Introduction
103
4.2 Structure of Traffic Networks
103
4.2.1 The Public Switched Telephone Network
4.2.2 Operator Functions
111

103

101

Contents

ix

4.2.3 Structure of Private Switched Networks
113
114
4.3 Numbering Plan
115
4.3.1 Nomenclature
117
4.3.2 Input Devices and Dialing Procedures
4.3.3 History and Evolution
117
120
4.3.4 The Principal City Concept
120
4.3.5 International Numbering
4.3.6 Other Special Numbering
121
4.4 Structure of the Facilities Network
122
122
4.4.1 Local Facilities Network
125
4.4.2 Interoffice Facilities Network
132
4.4.3 Evolution of the Digital Facilities Network
134
4.5 Network Configuration Planning
136
4.5.1 Alternative Plans
4.5.2 Economic Analysis
140
141
4.5.3 Other Network Planning Considerations
5 Traffic

147

5.1 Introduction
147
148
5.2 Traffic Theory Background
148
5.2.1 Measure of Traffic Demand: Offered Load
149
5.2.2 Grade of Service
151
5.2.3 Engineering Periods
5.2.4 Traffic Theory Techniques
155
156
5.2.5 Erlang's Blocked-CalIs-Delayed Model
158
5.2.6 Erlang's Blocked-CalIs-Cleared Model
5.3 Engineering Switching Systems
160
160
5.3.1 Digit Receivers
5.3.2 Operator Force
161
162
5.3.3 Stored-Program Control Systems
163
5.3.4 Capacity Considerations - Load Balancing
165
5.4 Engineering Trunk Groups
166
5.4.1 Applicability of the Erlang B Model
5.4.2 Day-to-Day Variation
166
5.4.3 Reattempts
167
168
5.4.4 Nonrandom Load
5.5 Traffic Network Design
169
169
5.5.1 The Economics of Alternate Routing
5.5.2 Modifications to the ECCS Engineering Technique
5.5.3 Dynamics of Network Design 173
5.6 Network Management
176
177
5.6.1 Trunk Congestion
177
5.6.2 Switching Congestion
177
5.6.3 Analysis of Congestion
179
5.6.4 Detection of Congestion
180
5.6.5 Control of Congestion
180
5.7 Traffic Measurements and Service Objectives
182
5.7.1 Traffic Measurement Procedures
5.7.2 Service Objectives
183

172

Contents

x

5.8 Traffic Considerations in Data Networks
184
5.8.1 The Nature of Data Traffic 184
5.8.2 Data Traffic Models
186
188
5.8.3 Data Performance Concerns
190
5.8.4 Traffic-Engineering Concepts

6 Transmission

193

6.1 Signal Types and Characteristics
193
194
6.1.1 Speech Signals
194
6.1.2 Program Signals
195
6.1.3 Video Signals
196
6.1.4 Data Signals
198
6.1.5 Overview
198
6.2 Channels
198
6.2.1 Basic Concept
199
6.2.2 Voice-Frequency and Carrier-Derived Channels
201
6.3 Transmission Media
6.3.1 Open-Wire Lines
201
6.3.2 Paired Cable
201
6.3.3 Coaxial Cable
203
6.3.4 Waveguides
204
6.3.5 Lightguide Cable
205
6.3.6 Terrestrial Microwave Radio
207
6.3.7 Satellite
209
6.4 Modulation
210
6.4.1 Double-Sideband Amplitude Modulation
211
6.4.2 Single-Sideband Amplitude Modulation
213
214
6.4.3 Pulse-Code Modulation
217
6.4.4 Modulating Digital Data Signals for Analog Facilities
218
6.5 Multiplexing
219
6.5.1 Frequency-Division and Time-Division Multiplexing
220
6.5.2 Multiplex Hierarchies
6.5.3 Multiplex Synchronization
221
223
6.6 Transmission Impairments and Objectives
224
6.6.1 Analog Speech Signals
236
6.6.2 Digital Data Signals

7 Switching

241

7.1 Introduction
241
7.2 Basic Switching Functions
242
242
7.2.1 Control
242
7.2.2 Signaling
243
7.2.3 Administration and Maintenance
243
7.2.4 Customer Services
243
7.3 Switching Networks
7.3.1 Circuit-Switching Network Types, Applications,
and Technologies
244
245
7.3.2 Space-Division Networks
250
7.3.3 Time-Division Networks

Contents

xi

7.4 Control Mechanisms
258
258
7.4.1 Direct Progressive Control
7.4.2 Registers, Translators, and Markers - Common Control
261
7.4.3 Stored-Program Control
8 Signaling and Interfaces

259

265

8.1 Introduction
265
265
8.1.1 Signaling
8.1.2 Interfaces
266
266
8.2 Signaling Functions
269
8.3 Fundamental Considerations
8.4 Signaling System Applications and Interfaces
270
8.4.1 Customer-Line Signaling
272
276
8.4.2 Interoffice Trunk Signaling
284
8.4.3 Special-Services Signaling
286
8.5 Signaling Techniques
286
8.5.1 Dc Signaling
289
8.5.2 Inband Signaling
8.5.3 Out-of-Band Signaling
290
8.5.4 Signaling Over Digital Facilities
291
8.5.5 Signaling Over Separate Facilities Common-Channel Interoffice Signaling
292
294
8.6 Internal Interfaces
295
8.6.1 The 4-Wire Analog Carrier Interface
299
8.6.2 The DSX-1 Digital System Cross-Connect Interface
8.6.3 The Digital Carrier Trunk Interface
301
8.7 Interfaces for Interconnection
305
8.7.1 Interconnection
305
8.7.2 Interfaces
306
8.7.3 Interconnection Environment
307
8.7.4 Evolution of Interfaces for Interconnection
310
8.7.5 Network Protection
311
313
8.7.6 Typical Network Interfaces for Interconnection
316
8.8 Data Communications Interfaces and Protocols
317
8.8.1 Layers of the Reference Model
8.8.2 Interfaces
319

PART THREE
NETWORK AND CUSTOMER-SERVICES SYSTEMS
9 Transmission Systems

327

9.1 Introduction
327
9.1.1 Areas of Application
327
332
9.1.2 Investment by Area
333
9.1.3 General Transmission System Types
9.2 Voice-Frequency Transmission
337
9.2.1 Loop Area Applications
337
339
9.2.2 Metropolitan Interoffice Applications
340
9.2.3 Outstate Interoffice Applications

325

Contents

xii

9.3 Analog Carrier Transmission
340
9.3.1 Loop Area Applications
340
9.3.2 Metropolitan Interoffice Applications
341
343
9.3.3 Outstate Interoffice Applications
347
9.3.4 Long-Haul Interoffice Applications
9.3.5 Analog Frequency-Division Multiplex Terminals
369
9.4 Digital Carrier Transmission
370
9.4.1 Loop Area Applications
9.4.2 Interoffice Area Applications
373
386
9.4.3 Digital Multiplex Equipment
393
9.5 Transmission Application Overview
393
9.5.1 Loop Applications
394
9.5.2 Interoffice Applications
10 Network Switching Systems

362

397

10.1 Introduction
397
398
10.2 Electromechanical Switching Systems
10.2.1 Evolution
398
10.2.2 Step-by-Step Systems
400
401
10.2.3 No.1 Crossbar System
10.2.4 No.5 Crossbar System
406
408
10.2.5 No. 4A Crossbar System
409
10.3 Electronic Switching Systems
409
10.3.1 Concepts
413
10.3.2 Evolution of Electronic Switching
10.3.3 Space-Division Electronic Switching Systems
415
425
10.3.4 Time-Division Switching Systems
10.4 Operator Systems
438
10.4.1 Toll Service
438
10.4.2 Number Services
443
10.5 Billing Equipment and Systems
445
10.5.1 Introduction
445
10.5.2 Early Billing-Data Collection Systems
447
10.5.3 Local Automatic Message Accounting
448
10.5.4 Centralized Automatic Message Accounting
451
10.5.5 Remote Recording
452
10.5.6 Remote Recording Using Store and Forward
456
10.5.7 AMA Summary and Future Trends
457
10.6 Switching Application Overview
457
10.6.1 Number of Lines and Local Switching Systems
461
10.6.2 Number of Toll Terminations and Toll Switching Systems
10.6.3 Future Trends
464
11 Customer-Services Equipment and Systems

11.1 Station Equipment
465
465
11.1.1 Telephone Sets
11.1.2 Data Sets
476
11.1.3 Data and Graphics Terminals

465

480

462

Contents

11.2

11.3

11.4

11.5

11.6

11.1.4 General Design Considerations
485
11.1.5 Current Trends
489
489
Customer Switching Systems
11.2.1 Key Telephone Systems
490
11.2.2 Modern Small Communications Systems
491
11.2.3 Vintage PBXs
493
11.2.4 Recent Developments in the Design of PBXs
499
11.2.5 Modern PBX Equipment
499
503
11.2.6 Automatic Call Distributors
11.2.7 Telephone Answering Systems
504
Public Switched Network Service Systems
507
11.3.1 The Stored-Program Control Network
507
11.3.2 Mass Announcement System
513
516
Mobile Telephone Systems
11.4.1 Land Mobile Telephone Systems
518
11.4.2 Paging
524
Visual Systems
526
11.5.1 Early Television
526
11.5.2 PICTUREPHONE® Visual Telephone Service
526
11.5.3 Video Teleconferencing
527
11.5.4 Research Activities
528
Data Communications Systems
529
11.6.1 Digital Data System
530
11.6.2 Packet-Switching Systems
535
11.6.3 DATAPHONE® Select-a-Station Service Implementation

12 Common Systems

xiii

538

541

12.1 Introduction
541
12.2 Power Systems
541
12.2.1 Energy Sources
543
12.2.2 Energy Storage
543
544
12.2.3 Power System Operation
12.2.4 Rectifiers
547
12.2.5 Dc-to-Dc Converters
548
12.2.6 Inverters
548
12.2.7 Power for Customer-Premises Equipment
549
549
12.3 Distributing Frames
12.3.1 General Description
549
550
12.3.2 Distributing Frame Functions
12.3.3 Distributing Frame Hardware and Application
551
12.3.4 Distributing Frame Administration and Engineering
559
12.4 Equipment Building Systems
12.4.1 Telephone Equipment Areas
559
12.4.2 Building Electrical Systems
561
12.4.3 Building Mechanical Systems
561
12.4.4 Special Construction
563
12.4.5 Equipment Building System Standards
564

555

xiv

Contents

12.5 Other
12.5.1
12.5.2
12.5.3

Common Systems
565
Cable Entrance Facility
565
565
Cable Distribution Systems
Alarm Systems
567

PART FOUR
OPERATIONS

569

13 Overview of Telephone Company Operations

571

13.1 Introduction
571
573
13.2 Customer-Related Operations
13.2.1 Provision of Service to the Customer
573
13.2.2 Service Administration
577
579
13.2.3 Customer-Service Maintenance Operations
13.2.4 Operations Related to Public Telephone Service
582
13.2.5 Directory Service
584
585
13.3 Network-Related Operations
13.3.1 Provisioning
585
592
13.3.2 Network Administration
13.3.3 Maintenance of the Network 597
14 Computer-Based Systems for Operations

603

14.1 Introduction
603
14.2 Recordkeeping and Order Processing
605
14.2.1 Trunks Integrated Records Keeping System (TIRKS)
605
14.2.2 Plug-In Inventory Control System (PICS)
611
14.2.3 Premises Information System (PREMIS)
616
14.3 Equipment Maintenance, Administration, and Control
621
14.3.1 Total Network Data System (TNDS)
622
632
14.3.2 Switching Control Center System (SCCS)
14.4 Planning and Engineering
638
14.4.1 Central Office Equipment Engineering System (COEES)
639
14.4.2 Facility Network Planning Programs
641
14.5 Development and Support
642
15 Operations Planning

645

15.1 Introduction
645
15.1.1 Initial Operations-Planning Efforts
646
15.1.2 Plans for Bell System Operations
647
15.2 Elements of Bell System Operations Plans
648
15.2.1 Operations Centers
650
15.2.2 Operations Systems
651
652
15.2.3 Operations Processes
15.2.4 Model Areas and Model Companies
652
15.3 Impact of Operations Plans on Bell Operating Companies
15.3.1 Improving Operations
654
15.3.2 Operating New Network Services
654
15.3.3 Operations Planning in the BOCs
656

654

xv

Contents

15.4 Impact of Operations Plans on the Development of
Operations Systems
659
660
15.5 Operations Systems Network Planning
15.6 Summary
662
16 Evaluation of Service and Performance

663

16.1 The Service Evaluation Concept
663
665
16.2 Network Characterization
665
16.2.1 The Need for Network Characterization Studies
666
16.2.2 The Results of Modern Characterization Studies
668
16.2.3 Mechanizing Performance Characterization Studies
669
16.3 Setting Performance Objectives
669
16.3.1 Overview
670
16.3.2 Creating Performance Models
16.3.3 Assembling Customer Opinion Models
671
16.3.4 Determining Grade-of-Service Ratings
674
675
16.3.5 Formulating Objectives
16.4 Measurement and Control of Service and Performance
676
16.4.1 Measurements in the Bell System
676
678
16.4.2 The Measurement of Service and Performance
678
16.4.3 The Control of Service and Performance
679
16.4.4 The Administration of Measurement Results
680
16.4.5 Typical Measurement Plans
16.4.6 Future Trends
683

PART FIVE
ENVIRONMENT AND EVOLUTION

685

17 The Environment
687
687
17.1 Introduction
17.2 Regulation
687
688
17.2.1 The Interstate Commerce Act
17.2.2 The Sherman and Clayton Antitrust Acts
688
17.2.3 The Kingsbury Commitment and the Graham-Willis Act
17.2.4 The Federal Communications Commission
692
692
17.2.5 State Regulatory Commissions
693
17.2.6 The "Above-890" Ruling
17.2.7 FCC Interconnection Rulings
693
17.2.8 The 1956 Consent Decree
694
695
17.3 Tariffs and Rate Settings
695
17.3.1 The Elements of a Tariff
17.3.2 The Development of a Tariff
697
17.3.3 Rate Setting
697
17.4 Competition
699
700
17.4.1 Early Years
17.4.2 Interconnection
701
17.4.3 Merging Data Processing with Telecommunications 702
FCC Inquiries
17.4.4 The 1982 Modification of Final Judgment
703

691

Contents

xvi

18 Evolution of Products and Services

705

1B.1 Overview of Product and Service Evolution
705
1B.1.1 Concept Formation
705
1B.1.2 Development
707
18.1.3 Introduction into the Market
70B
70B
1B.1.4 Mature Product Management
709
1B.2 Marketing
1B.2.1 Marketing Concepts
709
1B.2.2 Other Marketing Considerations
711
1B.2.3 Bell System Marketing Organizations
712
1B.3 Economic Evaluation
713
1B.3.1 Overview
713

1B.4

1B.5

1B.6

1B.7

18.3.2 Preparing Input Data
717
18.3.3 Calculating Economic Measures
718
18.3.4 Presenting and Interpreting Results
727
Application of New Technology
729
729
18.4.1 Matching Technology to Bell System Needs
18.4.2 Examples of Applying New Technologies
730
Integration of New and Old Systems
734
734
18.5.1 The Nature of the Problem
18.5.2 Some Problems in Interfacing TOUCH-TONE Service
with Older Equipment
736
Human Factors in the Bell System
737
737
18.6.1 The Human Factors Discipline
18.6.2 Designing for Customer Services
739
18.6.3 Designing for Employees
740
Quality Assurance
742
18.7.1 The Role of Quality Assurance
742
745
18.7.2 Quality Assurance Auditing
18.7.3 Quality Assurance Monitoring
756

References
Glossary

761
779

Acronyms and Abbreviations
Index

829

823

PART ONE
INTRODOCTION TO
THE BELL SYSTEM

The first three chapters of this book provide an overview of the Bell System and prepare the reader for the more detailed treatment of topics to
follow. Each chapter views the Bell System from a different perspective.
Chapter 1 discusses the overall corporate structure, activities, and responsibilities of the Bell System and its constituent companies and suggests
the size and complexity of the business. Material added to the chapter
late in 1983 describes structural changes associated with the divestiture of
the Bell System on January 1, 1984, and summarizes major provisions of
the Modification of Final Judgment. Chapter 2 describes the services the
Bell System makes available to customers in terms of customers' needs
and uses. Although this chapter includes a discussion of terminal products, it describes the service provided by the product rather than the
product itself. (Chapter 11 addresses equipment characteristics and
design considerations for customer products and services.) Chapter 3
completes the overview of the Bell System: It introduces the telecommunications network and a set of terms and concepts related to network
components and functions. These concepts are then applied in a discussion of the procedures involved in a typical telephone call.

1

1
Structure and Activities

1.1 INTRODUCTION
The Bell System led the development and use of communications equipment and techniques in the United States throughout most of this
century. It became the nation's major supplier of telecommunications
products and services ranging from basic residence telephones to increasingly sophisticated information services.
From its beginnings, the Bell System matched its organizational structure to the environment in which it operated. Early in the century,
universal service---..,.providing basic telephone service at an affordable
price anywhere in the nation-became the Bell System goal. The Bell
System approached this goal by creating a functional organization: Each
of the local telephone companies and the American Telephone and Telegraph Company (AT&T) itself were organized along the lines of the job
that had to be done. Tasks in each functional area were performed by
specialists in that area to maximize efficiency. The local companies were
responsible for responding to the particular needs of the communities
they served, but they all used standard technology and operating
methods. Thus, AT&T and the telephone companies achieved coordination on a national scale, while responding to local needs. As a result, the
goal of universal service has been met-nearly everyone in the United
States has a telephone that is connected to a single nationwide network.
This public switched telephone network! is available to the general public and to other carriers and networks.
The functional organization that made providing universal service
efficient and practical was pcssible because, for most of its history, the

1 Sections 3.3.1 and 4.2.1 discuss the public switched telephone network in detail.

3

4

Introduction to the Bell System

Part 1

Bell System was almost the sole source of telecommunications servicealthough under terms and conditions approved by federal and state regulators. AT&T and the telephone companies have changed their organizational structure to match environmental changes such as new and diverse
customer needs and, more recently, new markets.
Now, in the 1980s, the way in which telecommunications services are
provided is changing entirely. The 1982 Modification of Final Judgment
(MFJ), which terminated a 1974 Department of Justice antitrust suit
against AT&T, ordered the breakup (divestiture) of the integrated Bell
System. AT&T set January 1, 1984, as the target date for completion of
the massive job of restructuring a business involving about 1 million
employees and about $160 billion in assets.
Subsequent sections of Chapter 1 describe the organization of the Bell
System in 1982, the major provisions of the MFJ, and the postdivestiture
structure. The chapter was revised late in 1983 to provide an introductory account of the impact of divestiture. The rest of the book describes
the engineering and operations of the Bell System at the end of 1982 and
in early 1983 and does not reflect divestiture because much of the
material was prepared before the antitrust suit was settled in 1982.2 However, the primary effects of divestiture are on the structure of the corporate units and the allocation of the roles in providing telecommunications services. The technology in the network, the considerations
involved in its design, and the nature of functions required to operate
the network and to provide service to customers remain essentially
unchanged. Thus, this book constitutes a valid description of telecommunications engineering and operations and meets a growing need for
an update to the previous book (Bell Laboratories 1977).

1.2 THE BELL SYSTEM IN 1982
1.2.1 AMERICAN TELEPHONE AND TELEGRAPH COMPANY
In 1982, the Bell System, serving more than 80 percent of the nation's
telephones, had long been the largest of hundreds of companies providing communications services in the United States. Before divestiture, the
Bell System operated as a partnership among AT&T; a number of Bellowned telephone companies (known as operating companies); the Western
Electric Company, Incorporated; and Bell Telephone Laboratories, Incorporated. The product of this partnership was service, provided through a
dynamic and dependable communications network designed, built, and
operated as a single system.

2 Because the text describes the Bell System in 1982 and early 1983 and was written in that
time frame and earlier, the reader will find the Bell System referred to extensively in the
present tense.

5

Structure and Activities

Chap. 1

Figure 1-1 shows the structure of the Bell System as it was in 1982.
AT&T, the parent company, was publicly owned by 3.1 million stockholders. In turn, AT&T owned Western Electric and-totally or partiallyeach of the Bell operating companies. 3 AT&T and Western Electric jointly
owned Bell Laboratories. Both AT&T and Western Electric also had
subsidiary companies (shown on the figure); some of which supported
Bell System operations, others that provided domestic and international
communications services.

AT&T

GENERAL
DEPARTMENTS

195
BROADWAY
CORPORATION

LONG LINES
DEPARTMENT

IBELL
OPERATING
COMPANIES

WESTERN
ELECTRIC
COMPANY

I--

AMERICAN
BELL

I--

AT&T
INTERNATIONAL

L.....-

ADVANCED
MOBILE
PHONE
SERVICE

BELL
LABORA TORIES

I

I
TELETYPE
CORPORATION

NASSAU
RECYCLE
CORPORATION

SANDIA
CORPORATION

Figure 1-1. Structure of the Bell System (1982).

3 AT&T was sole stockholder in twenty-one operating companies and a minority
stockholder in two: the Southern New England Telephone Company and Cincinnati Bell,
Inc. Bell Telephone Company of Nevada is wholly owned by the Pacific Telephone and
Telegraph Company. Four Chesapeake and Potomac Telephone Companies offer service
in Washington, D.C.; Maryland; Virginia; and West Virginia.

6

Introduction to the Bell System

Part 1

AT&T, the parent company of the Bell System, had its headquarters in
New York City.4

Corporate Functions
General Departments. The General Departments of AT&T set the major
goals and large-scale programs for the Bell System. They advised and
assisted the Bell operating companies on such matters as finance, operations, personnel, legal, accounting, marketing, planning, public relations,
and employee information, thereby providing continuity of direction and
consolidating the particular specialities of each Bell System partner.
AT&T, through its General Departments, coordinated pricing activity and
represented the Bell System before federal regulatory agencies. It determined price structures and recommended costing and pricing matters
through federal agencies.
The General Departments ensured that new developments, solutions
to existing problems, and provisions for the future needs of customers
became part of the entire Bell System. This involved directing the work
of Bell Laboratories and Western Electric and coordinating the integration
of new technology into the network.
. The General Departments established and maintained standards and
procedures for the Bell System and for the interconnection of non-Bell
System equipment and facilities with the Bell System network. They
served as an information clearinghouse for associations of independent
telephone companies (such as the United States Independent Telephone
Association) and for general-trade (that is, other than Western Electric)
suppliers.
Long Lines Department. The Long Lines Department, with headquarters
in Bedminster, New Jersey, operated the long-distance network. Many of
its activities were similar to those of the Bell operating telephone companies. Long Lines built, operated, and maintained most of the interstate
network of long-distance lines, thereby providing interstate and international communications services for people throughout the United States.
It directed the overall design and management of the network and coordinated the teamwork among the various Bell and independent companies who jointly own and operate this complex, widespread system of
microwave radio, coaxial cable, optical fibers, satellites, and intricate
switching systems.
To handle the network efficiently, Long Lines was divided into six territorial regions (see Fig1.l.re 1-2). Each region took care of engineering,
sales, and network operations in its territory.
4 A new headquarters building at 550 Madison Avenue has replaced the building at 195
Broadway, which was the headquarters location for many years.

,

,0

~\/

1 I

/

---~-i
,

\

/

),
\l_

r- _ _ _

'~'7' WESTERN REGION
I

"

-

"'--""71---';I

SA~ FRANCISCO. CA
•

L - r- - _

•

I

\

I
\

I

I

\ J -----1'{
I

"

: --- --

I
I
I
I

Figure 1-2. Long Lines regions and regional headquarters locations.

Corporate Structure
Figure 1-3 is a block diagram of the AT&T organization as it existed in
1982. The office of the assistant to the president reviewed all aspects of
the organization and ensured that each unit's plans, budgets, and operations were consistent with system requirements. It also maintained
liaison with Bell Laboratories and with Western Electric. The figure
shows the network function and how AT&T's customer-related operations
and marketing functions were structured to reflect the Bell System's
market segments: business, residence, directory, and public services.
• Business organizations coordinated the response of the Bell System to
the needs of business customers; assisted telephone companies in the
areas of marketing, pricing, costing, forecasting, training, and budget
matters related to serving business customers; and supported telephone companies with installations, repairs, maintenance, customer
contacts, engineering, and measurements required for customer
services.

Business Marketing provided leadership and support for Bell System
business marketing efforts.
Business Services combined under common management all the
closely related delivery functions that flow from marketing and
sales.
7

CHAIRMAN
AND CHIEF
EXECUTIVE
OFFICER

PRESIDENT
AND CHIEF
OPERATING
OFFICER

ASSISTANT
TO THE
PRESIDENT

I
-

BUSINESS

-

BUSINESS
MARKETING

-

BUSINESS
SERVICES

I
--,

L. __

-

,.--RESIDENCE

....

RESIDENCE
MARKETING
SALES AND
SERVICE

~

STAFF

-

DIRECTORY
AND PUBLIC
SERVICES

-

INFORMATION
SYSTEMS

-

~-.I

NETWORK

-

NETWORK
PLANNING
AND
DESIGN

-

NETWORK
SERVICES

-

LONG LINES
DEPARTMENT

Figure 1-3. Corporate structure of AT&T (1982).

Structure and Activities

Chap. 1

9

· Residence organizations coordinated the response of the Bell System
to the needs of residential customers for telecommunications products
and services.

Residence Marketing Sales and Service offered telephone companies
help in marketing, pricing, costing, forecasting, training, and
budget matters related to serving residence customers. It also supported telephone companies with installations, repairs, maintenance, customer contacts, engineering, and measurements required
for customer service.
In addition, for organizational purposes, several other units reported
to the executive vice-president-residence.

Staff supplied support services within AT&T itself and coordinated
support services such as inventory management, automotive operations, building planning, real-estate management, energy conservation, and environmental protection offered by other Bell System
companies.
The Directory organization assisted telephone companies with
marketing, costing, pricing, forecasting, training, and budget
matters for both white pages and Yellow Pages as well as with
producing and distributing directories.

Public services coordinated Bell System activities involved in providing communications services for users who are away from home or
office. Public services comprised public and semipublic telephone
service, including Charge-a-Call and DIAL-Irs network communications services such as Public Announcement Service and Media
Stimulated Calling (see Section 2.5.1).
Information Systems provided planning, design, and development of
functional accounting and information systems for use by the
operating companies.
• Network organizations supported the business, residence, directory,
and public-service markets by guiding and coordinating the operation
of the network and the activities that provide telecommunications services between customers' locations.

Network Planning and Design oversaw the provision of reliable and
innovative interpremises communications services, ensured that
existing Bell System services were continually improved, coordinated the development of the national and international network,
guided the efforts of Bell Laboratories in the area of interpremises
5 Service mark of AT&T Co.

10

Introduction to the Bell System

Part 1

services, guided the technical planning of the operating companies,
and maintained technical liaison with both independent telephone
companies and international and other domestic carriers.

Network Services provided methods and guidance to operating telephone company network organizations. Supported functions
included the administration and maintenance of network switching
systems and transmission facilities; operator services; and the
engineering, construction, maintenance, and administration of distribution facilities and services.
The Long Lines Department was included in the network segment.
This integration made it easier to combine the planning, design,
and management of the interstate network with the intrastate networks for improved overall network service.

Subsidiary Companies
AT&T owned several subsidiary companies that supported Bell System
operations or provided domestic and international services. The primary
subsidiaries were:
• 195 Broadway Corporation, which provided real-estate management
services for AT&T corporate locations. These services included constructing, owning, and leasing buildings; administering office space,
facilities, and equipment; and providing related building and housekeeping support services such as transportation, maintenance, and
cafeterias for corporate buildings.
• AT&T International Inc., which was formed in August 1980 to sell
Bell System products worldwide and apply Bell System technology,
products, and experience to the needs of telephone administrations
overseas and international business customers. It also provides technical and advisory services and directory and informations systems.
• American Bell Inc., which was formed in June 1982 in response to
Computer Inquiry II (see Section 17.4.3). As a "separate" subsidiary,
American Bell could sell its products and services to customers
without government approval and had certain limitations in the way
it dealt with other Bell System companies.
• Advanced Mobile Phone Service, Inc., which was responsible for
planning and developing a nationwide cellular radio system to provide communications from moving customers to the land-line telecommunications system. Section 11.4.1 discusses cellular radio.

Chap. 1

Structure and Activities

11

1.2.2 BELL OPERATING COMPANIES
Before divestiture, the Bell operating companies built, operated, and
maintained the local and intrastate networks and provided most of the
day-to-day service for customers in their individual communities.
Chapter 13 discusses the many functional activities performed by the
operating companies. Long-distance calls also traveled over individual
company facilities, but those that went from the territory of one company
to that of another were carried by Long Lines or another common carrier.
(Figure 1-4 shows the Bell operating companies as they existed in 1982
and the territories they served.) The operating companies also joined
with Long Lines to furnish certain interstate services such as carrying
radio and television network programs to broadcasting stations
throughout the country. They also cooperated with the hundreds of
independent telephone companies so that the public had access to a
unified national telephone network.
The operating companies differed from one another in size and
organization. Geographically, the smallest was that part of the Chesapeake and Potomac Telephone Companies that offered service in the 61
square miles of the District of Columbia. The largest was the Mountain
States Telephone and Telegraph Company, which operated in seven states
and had a service area of more than 300,000 square miles.
The difference in size was one reason for differences in organization.
For example, a function that might have been centralized in a single-state
company might have had separate organizations for each state in a multistate company. There were other reasons for differences as well. For
example, the operating problems and priorities of rural areas differ from
those of urban areas. Traditionally, each company worked out operational methods most suited to its own needs, within guidelines and standards provided by AT&T.
As sole or part owner of the operating companies, AT&T derived a
large portion of its earnings from those companies. The relationship
between AT&T and an operating company was traditionally governed by
an agreement called the license contract (which terminated with divestiture). Each license contract described the reciprocal services, licenses,'
and privileges that existed between the parties. The operating company
was charged a fee for the services supplied by the AT&T General Departments. The fee was based on services the company received, but it could
not exceed 2.5 percent of the company's annual revenues. The licensed
company agreed to certain policies and procedures defined by the parent
company.
The term license contract goes back to the early days of the business
when local companies were first licensed to use Bell telephones. For
many years, the contract guaranteed that the operating companies would

---

---,--" -'--

MICHIGAN BELL

,

\

\
'... " r- - __

"

\

-----t
NORTHWESTERN BELL

I

____

I

- -J

l _,_

,

,L ___ _

,

~-",

---

,--,,

DIAMOND STATE
TELEPHONE
\.

I
I

---.L..

I

I

,I

SOUTHERN
NEW ENGLAND
- TELEPHONE
NEW JERSEY BELL

MOUNTAIN BELL

I
I

NEW ENGLAND
TELEPHONE

--,,.-----7
I

I

•

SOUTHWESTERN BELL

C&P TELEPHONE (VA.)

......

CINCINNATI BELL

Figure 1-4. Bell operating companies and their territories (1982).

13

Structure and Activities

Chap. 1

benefit from research, financing, engineering, and other important services offered by the parent company. It assured the manufacture of telephones and other devices and apparatus needed by the licensees for their
business. AT&T accomplished this through Western Electric.
Corporate Structure
As with AT&T, the'original organizational structure of the operating companies was defined by the jobs that needed to be done. This functional
organization, shown in Figure 1-5,6 later evolved into a network orientation. Still later, marketing became the driving force in shaping the Bell
System and its operations. When AT&T reorganized around market segments, it recommended that the operating companies do the same by

EXECUTIVE

I

I

TRAFFIC
DEPARTMENT

PLANT
DEPARTMENT

ENGINEERING
DEPARTMENT

OPERATORS

TEST & MAINTENANCE

FACILITIES ENGINEERING

TRAFFIC ADMINISTRATORS

INSTALLERS

TRAFFIC ENGINEERING

TERMINAL & SWITCHING
EQUIPMENT ENGINEERING

OUTSIDE CONSTRUCTION

MESSAGE TRUNK
FORECASTING

ASSIGNMENT

LONG-RANGE PLANNING

COMMERCIAL
DEPARTMENT
CUSTOMER INTERFACEBASIC SERVICE

MARKETING
DEPARTMENT
CUSTOMER INTERFACESPECIAL SERVICES
SPECIAL SERVICES
FORECASTING

Figure 1-5_ Traditional functional organization
of the Bell operating companies.

6 Chapter 13 describes functional activities such as those shown in Figure 1-5.

14

Introduction to the Bell System

Part 1

forming business, residence, directory, public service, and network
organizations.
The business, residence, directory, and public services organizations
were to be responsible for marketing, sales, and delivery of products
directly to customers. The network organization was to provide services
between customer locations. It did the planning and engineering, provided the facilities and equipment, and operated the network. This restructuring of traditional lines of managerial authority did not reflect a
difference in overall goals, however. Rather, it was done to keep pace
with the emerging technological, business, and regulatory environment.
1.2.3 WESTERN ELECTRIC
Western Electric, with headquarters at 222 Broadway in New York City,7
was the manufacturing and supply unit of the Bell System. In 1982, with
about $12.6 billion in sales, Western Electric ranked 22nd on Fortune
magazine'S list of the nation's 500 largest industrial corporations. The
company's almost 150,000 employees worked in nearly every state.
Before divestiture, Western Electric made a variety of customerpremises equipment, including millions of telephones each year. It also
manufactured much of the other equipment that made up the telephone
network. These products were designed by Bell Laboratories, manufactured by one of Western Electric's manufacturing divisions, and distributed to the telephone companies by Western Electric's Bell Sales division. Table 1-1 lists Western Electric's manufacturing divisions and their
locations. Figure 1-6 shows the twenty-two manufacturing plants. The
structure of Western Electric included a number of divisions responsible
for major functional areas.
The Corporate Engineering Division coordinated the work of the
manufacturing divisions to ensure that the products were compatible
with the network. The· division also provided research and development
support for all Western Electric's engineering activities including
manufacturing, equipment engineering, distribution, installation, and
repair of products. In addition, it coordin(ited the company's quality
assurance program, which required that inspectors check products to
ensure that they met Bell Laboratories standards. Engineering also
assisted in planning for the acquisition, leasing, and development of company facilities. It evaluated and verified company-wide cost reductions
and monitored energy use at all company locations.
Western Electric's Engineering Research Center near Princeton, New
Jersey, developed and improved manufacturing processes. Examples of

7 In October 1983, plans were announced for the sale of 222 Broadway and the
establishment of a new headquarters facility in Berkeley Heights, New Jersey.

Chap. 1

Structure and Activities

15

innovations that emerged from the Research Center include industrial
applications of the laser and a technique for sensing minute abnormalities in ceramics.

TABLE 1-1
WESTERN ELECTRIC MANUFACTURING (1982)
Products by
Manufacturing Division

Locations

Cable and Wire Products

Atlanta Works
(Norcross, GA)
Baltimore Works
Omaha Works
Phoenix Works

Electronic Components

Allentown Works
Kansas City Works
(Lee's Summit, MO)
Reading Works
New River Valley Plant
(Fairlawn, VA)

Business and Residence
Products

Denver Works
Indianapolis Works
Kearney Works
Montgomery Works
(Aurora, IL)
Shreveport Works

Switching EqUipment

Columbus Works
Dallas Works
(Mesquite, TX)
Hawthorne Works
(Chicago)
Lisle Plant
(Lisle, IL)
Oklahoma City Works

Transmission Equipment

Merrimack Valley Works
(North Andover, MA)
North Carolina Works
(Winston-Salem)
Burlington Works
Richmond Works

16

Introduction to the Bell System

Part 1

The Network Systems and Produ~t Planning Division ensured that
products from the different manufacturing divisions were compatible
with the network and met the needs of customers.
The principal points of contact between Western Electric and its Bell
System customers were the Material and Account Management Division
and two Bell Sales divisions. These organizations were responsible for
the delivery of products and services to the customer. To facilitate this,
seven Bell Sales regions were established (see Figure 1-6). Regional
account management teams assisted operating companies in planning
applications for Western Electric products and services and helped
Western Electric identify emerging telephone company needs and
develop marketing strategies to meet those needs.
• The Material and Account Management Division developed plans
. and support for the regional account management teams. It also forecast the demand for products, placed orders with the factories, and
controlled stock supplies in all seven regions. The division established prices and administered the standard supply contracts (see below).
It prepared brochures, handbooks, and customer instruction booklets
on products.
• Two Bell Sales Divisions (East and West) provided regional account
management. Their responsibilities also included systems equipment
engineering, installation and repair of switching and transmission
equipment, warehousing, and distribution for the Bell System. Systems equipment engineers tailored complicated Western Electric
equipment to the exact needs of the customer and ensured its compatibility with existing equipment. The Bell Sales divisions operated a
distribution network consisting of material management centers (huge
warehouses) in each of the seven regions; thirty-one smaller service
centers, which usually combined stocks of the most frequently needed
items and repair facilities; and strategically located local distribution
centers (see Figure 1-6).
Through its Purchasing and Transportation Division, Western Electric coordinated the purchase of over $4.5 billion in supplies and services
from other manufacturers both for its own m,anufacturing needs and for
resale to Bell System companies. Western Electric used more than 47,000
suppliers and transportation carriers and delivered raw materials, parts,
and finished products to more than 100 company locations and to Bell
customers around the country. Purchases included telecommunications
equipment, computers, power equipment, telephone booths, telephone
poles, office machines, maintenance items, and stationery supplies. An
important part of this work, the engineering and inspection of purchased
products to ensure their compatibility and quality, was the responsibility

l
PACIFIC !,EGION

o SUNS~.WHITNEY

* LlVERMO~ \

!

SOUTHW;~'teRN~EGION

o

GENERAL HEADQUARTERS

~

SOUTHGATE OFFICE BUILDING
OPERATING VICE PRESIDENTS

DALLAS eljl
MESQUITI!

e

GUILFORD CENTER

a COMMERCIAL SALES

0

RE'310NAL HEADQUARTERS

m

MANUFACTURING LOCATIONS

0

MATERIAL MANAGEMENT CENTERS

~

SERVICE CENTERS

•

DISTRIBUTION CENTERS

I:l

•
,

CORPORATE ADMINISTRATION
GOVERNMENT

i

m
SHREVEP~T

JACK SO,
MONTGOMERY

t

0

t

PURCHASED PRODUCTS ENGINEERING AND INSPECTION

e

INSTALLATION AREA OFFICES

<>

ENGINEERING RESEARCH CENTER

•

BELL SALES DIVISION TRAINING CENTER

...

NETWORK SOFTWARE CENTER

CORPORATE EDUCATION CENTER

*

SUBSIDIARIES

Figure 1-6. Western Electric's principal locations and regions (1982).

Chap. 1

Structure and Activities

19

of Purchased Products Engineering and Purchased Products Inspection,
which were located in Springfield, New Jersey.
Western Electric's responsibilities in this area were defined by the
standard supply contract, an agreement it had with the Bell operating companies. The supply contract, which terminated at divestiture, required
Western Electric either to manufacture or to purchase materials that the
operating companies might reasonably require, which they then might
order from "Western Electric. However, the supply contract did not obligate the operating companies to purchase these materials from Western
Electric. They were free to buy from anyone.
The Government and Commercial Sales Division was responsible for
the sale of Western Electric products and services to the United States
government and other non-Bell System customers.
In addition to its role as the Bell System manufacturing and supply
unit, Western Electric responded to the government's needs for both
specific design projects and telecommunications systems. During World
War II, Western Electric provided communications and radar equipment
to the armed forces. After the war, the company did pioneering work in
early-warning defense systems such as the Distant Early Warning (DEW)
Line, extending from Iceland to the Aleutians. Later, Western Electric
and Bell Laboratories developed the Nike-Ajax and Nike-Hercules
ground-to-air missile defense systems. More recently, in the early 1970s,
Western Electric was prime contractor for the Safeguard antiballistic missile system.
Western Electric has been a major technological contributor to the
space program. The company provided the tracking and communications
system for the United States' first manned space flight, Project Mercury,
and headed the industrial team that designed and built tracking and communications systems for the Gemini and Apollo programs. Bellcomm, a
subsidiary of Western Electric, was formed to carry out the systems
engineering work on these programs under contract to the National
Aeronautics and Space Administration. The United States Air Force and
the National Aeronautics and Space Administration use a Western Electric
command guidance system and missile-borne guidance equipment to support satellite launches. 8
Western Electric has also provided complete telecommunications facilities for various government agencies both in the United States and at
military bases and embassies abroad. Western Electric recently modified
the Nike-Hercules air defense system for the North Atlantic Treaty
Organization. The company has been engaged, with Bell Laboratories, in
United States Navy submarine sonar and underwater surveillance projects
8 For further details on Bell System contributions to military and space programs, see Fagen

1975/1978, vol. 2.

20

Introduction to the Bell System

Part 1

involving the application of acoustic technology and oceanography.
These designs include underwater sensor components, cable systems,
associated data-processing equipment, displays, data transmission, and
communications links.

Subsidiary Companies
Western Electric owns several subsidiary companies that have supported
Bell System operations or provided services. The subsidiaries include:
• Teletype Corporation, which develops and manufactures data terminals for the Bell operating companies, other companies, and the
United States government. This equipment is used in news and wireservice operations, in data communications, and in computer systems.
Teletype Corporation, with headquarters and engineering operations
in Skokie, Illinois, maintains two manufacturing plants and a nationwide network of service centers.
• Nassau Recycle Corporation, which reclaims and recycles nonferrous
metals such as copper and zinc from scrap equipment and cable.
About one-third of the copper Western Electric uses in manufacturing
is provided by Nassau Recycle. The company has plants in South
Carolina and New York.
• Sandia Corporation, which is managed by Western Electric for the
United States Department of Energy under a no-profit, no-fee contract.
Sandia's principal functions are research and development of nuclear
ordnance, research on energy projects, and various other programs in
the national interest. Sandia has laboratories in Albuquerque, New
Mexico, and Livermore, California.

1.2.4 BELL LABORATORIES
Before divestiture, Bell Laboratories was the Bell System's research and
development organization. Recognized worldwide as a prestigious
scientific and technical institution, it was the driving influence behind
the Bell System's contributions to telecommunications science and technology. The broad scope of these contributions is reflected in Table 1-2.
In 1982, engineers and scientists at Bell Laboratories received 310 patents,
bringing the total number of patents issued to the company since its
founding in 1925 to 19,833. In 1982, they also originated 3823 technical
talks to outside organizations and 2087 papers and received more than 87
scientific and engineering awards. Seven scientists from Bell Laboratories
have been awarded the Nobel Prize in physics (see Table 1-3).

TABLE 1-2
A SAMPLING OF BELL LABORATORIES
CONTRIBUTIONS TO TELECOMMUNICATIONS
SCIENCE AND TECHNOLOGY*

Microelectronics

Photonics

Transistor effectt
Silicon gate technology
Molecular beam epitaxy
Charge-coupled devices
Microprocessors and
microcom puters

Lasers
Lightwave
communications
systems
Ultra-transparent glass
fibers
Light-emitting diode

Software Systems

General Science and
Engineering

Error-correcting code
Computer languages
Computer graphics
Computer operating
systems
Operations systems
Centralized
maintenance
systems
Stored-program control
network

Digital Technology
Electrical digital
computer
Digital switching system
Digital transmission
Packet switching
Echo canceler chip

Single-side band
transmission
Network theory
Quality control
Systems engineering
concept
Negative feedback
Wave nature of mattert
Thermal noise
Speech synthesis
Radio astronomy
Traveling-wave tube
Microwave technology
Information theory
Solar cell
Cellular radio concept
Communications
satellite
Supercurrent junctionst
Cosmic background
noiset

... For a more complete list and discussion of Bell Laboratories' contributions,
see Bell Laboratories 1982, Lustig 1981, and Mueser 1979.
tNobel Prize.

22

Introduction to the Bell System

Part 1

TABLE 1-3
BELL LABORATORIES
NOBEL LAUREATES IN PHYSICS

1937

C. J. Davisson

Demonstration of wave
nature of matter

1956

John Bardeen
Walter Brattain
William Shockley

Discovery of transistor effect

1977

Philip Anderson

Study of electronic structure
of magnetic and disordered
materials

1978

Arno Penzias
Robert Wilson

Detection of cosmic
microwave background
radiation

NOTE: Arthur Schawlow, co-inventor (with C. H. Townes) of the laser while at Bell
Laboratories from 1951 to 1962, shared the 1981 prize for work done later at Stanford on
Doppler-free spectroscopy.

Purpose
Before divestiture, the purpose of Bell Laboratories was to provide the
knowledge and technology essential to meeting the current and future
communications needs of Bell System customers. Its activities were
divided into two categories: Research and Systems Engineering (R&SE)
and Specific Development and Design (SD&D).
In undertaking research, Bell Laboratories sought new knowledge
relevant to communications, explored the potential usefulness of that
knowledge, and looked for new modes of communication based on that
knowledge. The aim of Bell Laboratories research was to improve the
services provided by the Bell operating companies and to reduce the cost
of providing those services. Fields of research included the physical and
mathematical sciences, computer science, economics, communications
principles, communications technology, engineering, and the behavioral
sciences. To maintain its leadership in telecommunications, Bell Laboratories has, in recent years, devoted ever more of its efforts and resources
to certain fundamental information.;.age technologies, especially
microelectronics, software systems, digital systems, and photonics.
Systems engineers planned the nationwide telephone network and its
operations. They considered the entire network rather than just one part

Chap. 1

Structure and Activities

23

of the pZani 9 or one phase of operations. This included studying performance objectives, evaluating service quality, planning network
configurations, generating operations plans and methods, forming
requirements for equipment to provide service, and defining plans and
procedures for introducing new equipment and services into the network.
Systems engineering at Bell Laboratories ensured that the entire telecommunications network worked efficiently to provide continuous service
while new technologies, operations, equipment, and services were introduced as they became available. Research and Systems Engineering at
Bell Laboratories was funded by AT&T, primarily as a part of the service
the parent company provided to the operating companies under the
license contracts.
Specific Development and Design was funded by Western Electric. It
was directed toward components, devices, and specific products (often
involving both hardware and software) to be manufactured and furnished
by Western Electric. It was concerned with designing new telecommunications applications using existing types of devices, designing completely
new circuits and equipment arrangements, and preparing manufacturing
information and test specifications. It involved building and testing
equipment designs both in the laboratory and under field conditions,
dealing with the problems of early manufacture and use of a product, and
making changes as indicated by experience.
The operating companies directly funded certain other work that Bell
Laboratories undertook at their specific request, for example, developing
computerized operations systems (see Chapter 14) for use in telephone
company business operations.

Corporate Structure
Figure 1-7 shows the corporate structure of Bell Laboratories as it existed
in January 1983 following the transfer of certain customer products and
services organizations to American Bell Inc.

· Research included divisions devoted to physics, the information sciences, the communications sciences, and materials science and
engineering.
• Legal comprised general law, patents, regulatory matters, and corporate studies.
• Research and Development Planning was concerned with the organizational structure of Bell Laboratories as it was affected by the changing regulatory and business environment of the Bell System.
9 All of the facilities (such as land, buildings, machinery, apparatus, instruments, and
fixtures) needed to provide telecommunications services. Plant is usually divided into
outside plant and inside plant.

PRESIDENT

I
~

CUSTOMER
SYSTEMS

I....

COMPUTER
TECHNOLOGIES
AND MILITARY
SYSTEMS

I

I

I
NETWORK
SYSTEMS

,....

STAFF

I-

ELECTRONICS
TECHNOLOGY

-

FINANCE AND
GENERAL
SERVICES

.....

PERSONNEL
AND
PUBLIC
RELATIONS

""'"

I-

RESEARCH

I-

LEGAL

I-

TRANSMISSION
SYSTEMS

-

RESEARCH AND
DEVELOPMENT
PLANNING

~

SWITCHING
SYSTEMS

I....

OPERATIONS
SYSTEMS AND
NETWORK
PLANNING

Figure 1-7. Corporate structure of Bell Laboratories (January 1983).

• Computer Technologies and Military Systems concentrated on
research and development of computer software and hardware and on
military systems work. Software development recently became an
ever increasing part of the work at Bell Laboratories, and varying
24

Chap. 1

Structure and Activities

25

amounts were done in other organizational areas as well. The efforts
of the Military Systems Division were devoted exclusively to Western
Electric's government contracts.
• Electronics Technology developed electronic components for systems
of all kinds. It was involved in broad areas of research and development ranging from the design and processing of integrated circuits to
lightwave communications subsystems to interconnection hardware
and power-conversion systems.
• Transmission Systems provided the systems engineering, design, and
development of systems to meet needs ranging from loop transmission
to long-haul land, satellite, and undersea transmission. Research and
development of digital transmission was an important aspect of work
in this area.
• Switching Systems performed research, planning, and development to
maintain, improve, and offer new services on existing switching systems and for planning and developing future switching systems.
• Operations Systems and Network Planning developed and designed
computer-based systems to support telephone company operations and
formulated plans for the effective integration of these systems with
each other and with the people these systems support. Network planning encompassed functions involved in the evolution and implementation of the network, determining its configuration to best meet
future service needs. The Bell Laboratories' Quality Assurance Center
was part of this area. Bell Laboratories set the quality standards for
products, and it worked with Western Electric quality assurance organizations to monitor manufacturing operations. 10
• Finance and General Services had a wide range of company responsibilities including corporate auditing and finance, internal communications services, security, and managing buildings and grounds.
• Personnel and Public Relations included salary administration, personnel matters, affirmative action, public and employee communications, education, and medicine and environmental health.
In 1982, Bell Laboratories' main facilities for research and development were in New Jersey and Naperville (Indian Hill), Illinois (see Table
1-4). In addition, about 15 percent of the staff of nearly 26,000 people
was located at seven Western Electric manufacturing plants. These branch
laboratories (see Table 1-5) helped coordinate and implement Bell Laboratories' specific development and design functions that resulted in drawings and specifications for the telecommunications and software products

10 Section 18.7 discusses quality assurance.

TABLE 1-4
BELL LABORATORIES LOCATIONS (1982)
Activities

Location

Murray Hill, NJ

Administrative headquarters, electronics
technology, basic research in various fields

Holmdel, NJ

Systems planning, network planning,
operations systems planning, operations
research, quality assurance, switching,
transmission, customer equipment, research in
communications sciences

Whippany, NJ

Loop transmission, mobile radio systems,
interconnection, computing technology,
engineering information, electronic power
systems, military systems

Indian Hill, IL

Electronic switching, computer technology

Piscataway, NJ

Operations and network systems

Chester, NJ

Field laboratory for outside plant equipment
and materials and constructed equipment

Crawford Hill, NJ

Radio and guided wave research

Freehold, NJ*

Business services operations and
communications systems engineering

Neptune, NJ*

Engineering for facility networking operations
and residence systems, DATAPHONEt digital
service field support and exploratory
development

Short Hills, NJ

Personnel, public relations, legal, finance and
general services

South Plainfield, NJ

Computer program development, new services
planning, quality assurance, network
performance planning, education

Warren, NJ

Service center for New Jersey locations, stock
supply center for all Bell Laboratories
locations

West Long
Branch, NJ

Switching and transmission engineering,
network planning, operations research, quality
assurance administration

* Location became part of American Bell Inc. in 1983.

t Registered service mark of AT&T Co.

Structure and Activities

Chap. 1

27

TABLE 1-5
BELL LABORATORIES
BRANCH LABORATORIES (1982)
Location

Activities

Allentown, PA

Electronic devices, integrated circuits

Atlanta, GA

Wire, cable, glass lightguides, systems for
joining media

Columbus, OH

Switching systems

Denver, CO*

Customer switching systems for PBX services

Indianapolis, IN

Telephones, residential terminals and home
communications systems

Merrimack Valley, MA

Microwave radio, carrier transmission

Reading, PA

Electronic devices, integrated circuits

*Location became part of American Bell Inc. in 1983.

that Western Electric manufactures. Bell Laboratories also had field
representatives at the headquarters locations of the Bell operating companies. They provided designers with rapid feedback on the quality and
performance of new and existing telecommunications equipment.
1.2.5 RELATIONSHIPS WITH NON-BELL SYSTEM COMPANIES
Before divestiture, the Bell System served over 80 percent of the 180 million telephones in the United States, encompassing 30 percent of its geographical area. The remaining 36 million telephones were served by
more than 1400 telephone companies that were not part of the Bell System. These independents, as they are called, worked with each other and
interfaced with the Bell System through the United States Independent
Telephone Association (USITA).
Through committees representing different aspects of the telecommunications business, USITA served as a focal point for agreements with
the Bell System on issues such as routing the long-distance network and
sharing revenues. On technical issues, the Bell System prepared network
planning information and equipment compatibility specifications and
released them to the independent telephone companies through USITA's
Equipment Compatibility Committee and Subcommittee on Network

28

Introduction to the Bell System

Part 1

Planning. These releases, called technical advisories, ensured that the
equipment and systems installed by the independents were compatible
with the Bell System network.
Overseas and other international services use high-frequency radio,
undersea cable, and satellite links. AT&T, other United States carriers,
and foreign carriers share the ownership of some transmission facilities
such as undersea cable; and they lease overseas voice circuits from satellite channels provided by the Communications Satellite Corporation
(COMSAT). Service agreements with each foreign agency set up the type
and extent of service and procedures for dividing revenues, and they
establish the criteria for operations such as circuit engineering and quality of service.
The Bell System has participated in technical planning for international coordination through the Comite Consultatif International
Telegraphique et Telephonique (CCITT) and the Comite Consultatif International des Ra4io-communications (CCIR). These are the technical
organs of the United Nations' specialized agency for telecommunications,
the International Telecommunication Union (ITU). They function
through international committees of telephone administrations and
private operating agencies. Their recommendations, although not carryingthe force of regulations, are generally observed, and more and more,
they are becoming a consideration in system design, particularly for digital transmission and switching.
The Bell System has also worked with manufacturers other than
Western Electric (general trade) who sell their products to Bell System
companies. Information essential to general-trade manufacturers, specialized carriers, and other communications companies was made available in
various documents. These include technical descriptions, technical manuals for Western Electric products, technical references containing interface
information and technical standards for all aspects of the network and its
operation, and textbooks and manuals used by Bell System designers.
The public has also been able to subscribe to periodicals such as The Bell
System Technical Journal, the Bell Laboratories Record, and the Bell Journal of

Economics.
1.2.6 RESOURCES AND VOLUME OF SERVICE
After decades of growth, the goal of universal service has been achieved,
and telecommunications services have become an increasingly important
part of personal and business activities. Almost 1-~ billion miles of wire
and radio paths interconnect almost every home and office in the United
States. Over 180 million telephones have immediate, real-time access to
each other and to 98 percent of another 315 million telephones in other
countries.
As a result of such growth, the Bell System became a very large enterprise. Its size can be viewed from several perspectives: financial measures

Chap. 1

Structure and Activities

29

such as its revenues and the amount of capital investment or plant, service measures such as the number of telephones and number of calls handled, and a measure of the vast amount of effort required to build and
operate the network and deliver its services-the number of employees.
Table 1-6 summarizes Bell System resources and volume of service at the
end of 1982. To complete the picture, the corresponding figures for the
independent telephone companies are also shown.

TABLE 1-6
RESOURCES AND VOLUME OF SERVICE (1982)

Operating companies
Employees
PIC:lnt ($ billions)
Construction ($ billions)

Bell System

Independents

Total

22*
1,009,817
158.0
16.8

1,432
192,100
41.5
4.7

1,454
1,201,917
199.5
21.5

84.7

21.7

106.4

Access lines (millions)
Central office codes

19,660

Local calls (billions)
Long-distance calls
(billions)
Average calls / day
(millions)
Revenues (billions)

11,742

31,402

178.9

67.1

246.0

25.9

6.6

32.5

561.1

201.8

762.9

65.1

13.9

79.0

SOURCES: AT&T 1983 and USITA 1983.
"Excluding Southern New England Telephone Company and Cincinnati Bell, Inc.

1.3 THE POSTDIVESTITURE VIEW
1.3.1 SUMMARY OF MODIFICATION OF FINAL JUDGMENT
PROVISIONS
The 1982 Modification of Final Judgment (MFJ) requires AT&T to divest
itself of the twenty-two Bell operating companies (BOCs). The major provisions of the MFJ are summarized in Table 1-7. Those provisions that
significantly affect the conduct of engineering and operations are
explained in more detail in the following paragraphs.
The nationwide Bell System network, which was designed, built, and
operated as a single unit prior to divestiture, is now divided into two

TABLE 1-7
MAJOR PROVISIONS OF THE
MODIFICATION OF FINAL JUDGMENT
1.

Sufficient facilities, personnel, systems, and rights to technical
information must be transferred from AT&T to the BOCs, or to a new
entity owned by the BOCs, to allow the BOCs to provide exchange
and exchange access services independent of AT&T.

2.

Facilities, personnel, and accounts used to provide interexchange
services or customer-premises equipment (CPE) must be transferred
from the BOCs to AT&T.

3.

License contracts between AT&T and the BOCs and the standard
supply contracts between the BOCs and Western Electric must be
terminated.

4.

BOCs may create and support a centralized organization for the
provision of those services that can be most efficiently provided on a
centralized basis. The BOCs shall provide, through a centralized
organization, a single point of contact for coordination of the BOCs
for national security and emergency preparedness.

5.

BOCs must provide all interexchange carriers with exchange access
services equal in type, quality, and price to those provided to AT&T.
This "equal access" must be provided on a gradual basis over a 2-year
period beginning September 1, 1984. By September 1, 1986, all BOC
switching systexp.s must provide equal access, although exceptions
may be made for electromechanical switches or switches serving
fewer than 10,000 access lines where costs of providing equal access
are prohibitive. Any such exceptions shall be for the minimum
divergence in access and minimum time necessary.

6.

BOC procedures for procurement of products and services,
dissemination
of
technical
information
and
standards,
interconnection and use of BOC facilities and services, and planning
and implementation of new services or facilities must not
discriminate between AT&T and its affiliates and their competitors.

7.

BOCs may provide, but not manufacture, CPE after divestiture.

8.

BOCs may
divestiture.

9.

With the permission of the court, the BOCs may provide products or
services in addition to exchange and exchange access services upon
showing that there is no substantial possibility they could use their
monopoly power to impede competition in the additional markets.

produce

and

distribute

printed

directories

after

Chap. 1

Structure and Activities

31

components: an exchange and exchange access portion provided by the
divested BOCs and an interexchange portion provided by AT&T. This
division does not correspond to the predivestiture distinctions between
AT&T Long Lines and BOC operations, between intrastate and interstate
jurisdictions, or between toll and local services. It is based instead on a
definition of an exchange used in the MFJ.
Prior to divestiture, the term exchange area was used to describe an
area in which there was a single, uniform set of charges for telephone
service. Calls between points in an exchange area were local calls. The
MFJ defines an exchange area or exchange to be generally equivalent to a
Standard Metropolitan Statistical Area (SMSA), which is a geographic
area defined by the United States government for statistical purposes.
The MFJ concept is to group large segments of population with common
social and economic interests within an exchange.
The territory served by the Bell System has been divided into approximately 160 of these exchanges, which are also referred to as local access
and transport areas (LATAs). Depending on population densities and other
factors, most LATAs serve territories ranging from major metropolitan
areas to entire states. Accordingly, LATAs generally contain a number of
predivestiture exchange areas.
The predivestiture BOCs performed functions that now represent both
inter-LATA and intra-LATA functions. The MFJ specifies that, after
divestiture, BOCs offer regulated telecommunications services within
LATAs, while AT&T and other interexchange carriers (ICs) offer services
between LATAs. Some examples of exchange services offered by BOCs
are basic local telephone service for residence and business customers,
public telephone services, and intra-LATA operator services. In addition,
BOCs offer exchange access services that allow inter-LATA networks provided by ICs to access customers within a LATA and allow end-users to
access inter-LATA services. Examples of IC services include inter-LATA
and international telephone service and inter-LATA operator services.
In addition to reconfiguring their operations to accommodate the
transfer of inter-LATA functions to AT&T, the BOCs must modify the
intra-LATA networks to provide all other ICs, at their option, with
exchange access equal in type, quality, and price to that provided to
AT&T. The quality of exchange access is measured in terms of traffic
blocking criteria (see Chapter 5) and transmission performance (see
Chapter 6). In addition, BOCs must implement a new national numbering plan that provides exchange access to every IC through a uniform
number of digits.
The MFJ prohibits joint ownership of switching systems, transmission
facilities, and operations systems (see Chapter 14) by the BOCs and
AT&T. All Bell System assets are assigned to one or the other. The MFJ
does, however, allow sharing, "through leasing or otherwise," of facilities
that support both BOC and AT&T services. Sharing of such multifunctional facilities for a period after divestiture is necessary because of the

32

Introduction to the Bell System

Part 1

impracticality and enormous cost associated with the immediate
reconfiguration and separation of the predivestiture Bell System network.
After this transition period, BOC and AT&T facilities will be completely
separated.
In its Computer Inquiry II decision (see Section 17.4.3), the Federal
Communications Commission (FCC) required that all new customerpremises equipment (CPE) be provided by a separate subsidiary on a
detariffed basis effective January 1, 1983. Installed CPE and remaining
BOC inventories of CPE as of that date were sold or leased by the BOCs
during 1983. The MFJ requires that leased CPE be transferred to AT&T at
divestiture. After divestiture, BOCs are allowed to provide, but not
manufacture, new CPE.
The provisions of Computer Inquiry II and the MFJ have a major
impact on the BOC organizations that directly interface with customers
for the provision of service and equipment, billing and collections,
trouble referral, and other matters. Prior to divestiture, the BOCs provided a single point of customer contact for local and toll services as well
as CPE. Under the MFJ, the BOC personnel and associated customer support responsibilities for these services are divided between the AT&T and
BOC units responsible for regulated network services or CPE.
Figure 1-1, presented earlier, depicts the predivestiture relationship of
the components of the Bell System. Figure 1-8 shows the new corporate
structures resulting from divestiture. Sections 1.3.2 and 1.3.3 provide
additional information on the organization and functions of the postdivestiture AT&T and BOCs, respectively.

1.3.2 THE NEW AT&T
The structure of the new AT&T is shown in Figure 1-8. AT&T Corporate
Headquarters sets overall corporate policy and strategy for the other six
entities shown. Five of these entities are divided into two sectors: AT&T
Communications and AT&T Technologies, which are responsible for
essentially regulated and unregulated activities, respectively. As AT&T
gains experience in the new telecommunications environment, organizational structures and activities will be evolving to improve operating
effectiveness. Therefore, only a brief summary of each entity is provided
in this section.
AT&T Communications
The business of AT&T Communications is moving information electronically, from customer premises to customer premises, domestically and
internationally. Initially, employees were drawn from BOCs, the AT&T
General Departments, and Long Lines. At its inception, the company
served sixty million residence customers and nearly six million

I

RBOC 1

...

I

I

I

BOCS

OTHER
SUBSIDIARIES

...

I

I

RBOC 7

I

I

BOCS

OTHER
SUBSIDIARIES

I
CSO

AT&T
CORPORATE
HEADQUARTERS
AT&T
COMMUNICATIONS
SECTOR

I

I

AT&T
COMMUNICATIONS

AT&T
WESTERN
ELECTRIC

J

I
AT&T
BELL
LABORATORIES

I

I
AT&T TECHNOLOGIES SECTOR

I

I

AT&T
INFORMATION
SYSTEMS

AT&T
INTERNATIONAL

I
AMERICAN
TRANSTECH

Figure 1-8. Corporate structures after divestiture. Top, structure of the
divested operating companies (the RBOCs and their associated BOCs are
identified in Figure 1-9); bottom, structures of AT&T.

businesses. As a result of the new regulatory environment, AT&T Communications will provide inter-LATA long-distance services, which
include the interstate services previously provided by AT&T Long Lines
and intrastate, inter-LATA services. Its goals, highlighted at the beginning of its mission statement, are: "To provide high-quality, innovative,
widely available communications services that satisfy customers' needs to
move information electronically throughout the United States and the
world."
AT&T Western Electric

AT&T Western Electric continues in its leadership role as a provider of
technologically advanced, high-quality products and services in the
telecommunications and information systems markets. These include
equipment and systems for telephone companies, consumer products,
electronic components, and processors. Further, according to its mission
statement, "within this role, AT&T Western Electric has a responsibility
to expand its telecommunications business and address new market
opportunities .... "
33

34

Introduction to the Bell System

Part 1

AT&T Information Systems
AT&T Information Systems, derived initially from American Bell Inc.,
develops, sells, and services leading-edge communications products,
information management systems, and enhanced services to business customers. It also distributes products for residential and small business customers. Its products, services, and systems reflect the rapid convergence
of what were three distinct industries: telecommunications, office equipment, and data processing.
AT&T International
AT&T International, as before divestiture, markets, sells, and services
current and future products and services of the AT&T Technologies sector
outside the United States.
AT&T Bell Laboratories
AT&T Bell Laboratories provides the technology base for AT&T's future
and designs and develops the systems and services needed by AT&T
enterprises. This includes basic research and the engineering and design
of components, devices, and information and operations systems and services. It also conducts systems engineering work to help identify the best
solution to customers' needs. It aids the national defense by making its
special capabilities and expertise available to the government.
American Transtech
American Transtech provides and/or packages quality stock transfer and
related services for AT&T and regional company shareowners and customers at the lowest reasonable cost. Its mission statement further
states: "American Transtech will enter new business opportunities that
maximize existing functions and capacity to produce an attractive rate of
return and growth."
1.3.3 THE DIVESTED OPERATING COMPANIES
The MFJ allows the BOCs considerable latitude regarding their choice of
corporate structure and organization after divestiture. It explicitly states
that "nothing in this Modification of Final Judgment shall require or
prohibit the consolidation of ownership of the BOCs into any particular
number of entities."
After study by a committee of AT&T and BOC officers, the structure
adopted (see Figure 1-8) organizes the postdivestiture BOCs into seven
regional BOCs (RBOCs). These seven independent corporations wholly
own and are supported by a separate central services organization (CSO).
These organizational units are described in the following paragraphs.
They have no corporate connection with the new AT&T or its affiliates.

Chap. 1

Structure and Activities

35

Regional Bell Operating Companies
RBOCs were designed to form corporate units with roughly equivalent
assets and financial strength. Each RBOC contains from one to seven
BOCs that serve the same general region of the country. Figure 1-9 is a
map showing postdivestiture RBOC boundaries. The table below the map
presents statistics that demonstrate the relative equivalence in size of the
seven new corporate units.
The RBOCs operate as holding companies that hold the stock of the
BOCs in their respective regions. The RBOCs are free to enter other,
unregulated lines of business through the creation of separate subsidiaries. For example, each region has already formed a subsidiary to provide cellular mobile telephone services. Finally, the RBOCs jointly direct
the work of the CSO.

Bell Operating Companies
The BOCs offer regulated intra-LATA telecommunications services and
exchange access services within their predivestiture operating territories.
While the BOCs within each region remain as separate corporations with
their own boards of directors and officers, they cooperate to achieve
economies of scale possible within the larger, regional framework.
The BOCs in each region have already consolidated certain functions,
such as procurement and staff services. In most cases, this has been
accomplished through the formation of a Regional Services Company,
staffed and managed by the BOCs in the region. The BOCs within a
region may also cooperate in financial, strategic, and network planning to
meet their obligations to the RBOC that is their parent company.
BOC operations are modified considerably as a result of divestiture.
Personnel and organizations that support the provision and maintenance
of CPE, inter-LATA telecommunications services, and directory services
move to AT&T entities or to other subsidiaries of the RBOCs. While most
of the operations functions and organizations described in Chapters 13
through 16 are still present in the BOCs, the details of operations change
considerably. The creation of LATAs defines a new set of boundaries for
the operation and engineering of the network. Similarly, management
and administration of the network are no longer centrally directed by
AT&T. Massive transfers of assets and personnel, changes to records and
support systems, and modifications to operating procedures were required
to accommodate the physical rearrangement of the Bell System and the
breakup of its corporate structure.

Central Services Organization
While the MFJ ends the centralized ownership and management of the
Bell System and breaks up the vertically integrated structure that combined operations, research, and manufacturing, it recognizes the possibility that the BOCs might still choose to provide certain support functions

NYNEX
NEW ENGLAND TELEPHONE
NEW YORK TELEPHONE

AMERITECH

D

BELL ATLANTIC
BELL OF PENNSYLVANIA
DIAMOND STATE TELEPHONE
C&P TELEPHONE
NEW JERSEY BELL

PACIFIC BELL
NEVADA BELL

Estimated
Number of
Employees
(Thousands)

Total
Assets
($ Billions)

Estimated
Total
Access
Lines
(Millions)

1/1/84

6/30/83

1/1184

Total
Number of
Telephones
(Millions)

NYNEX
Corporation

25

98.2

17.4

12.8

17.4

Bell Atlantic
Corporation

27

80.0

16.3

14.2

23.2

BellSouth
Corporation

30

99.1

20.8

13.6

23.1

American
Information
Technologies
Corporation
(Ameritech)

30

79.0

16.3

14.0

23.6

Southwestern
Bell
Corporation

21

74.7

15.5

10.3

16.9

US WEST,
Inc.

22

75.0

15.1

10.6

16.7

Pacific Telesis
Group

20

82.0

16.2

10.9

15.1

Estimated
Population
Served
(Millions)

NOTE: Based on Bell System figures for December 1982 except as noted.

Figure 1-9. Regional Bell operating companies.

Chap. 1

Structure and Activities

37

on a centralized basis. The CSO provides this support. The economies
realized by not duplicating certain technical and support functions in
each RBOC enhance the financial position of the new companies and
reduce the requirements for staffing these highly technical activities.
Also, the centralization of certain activities in support of the BOC networks promotes technical compatibility and supports high-quality,
nationwide telecommunications service. These activities also facilitate the
introduction of new technology into a national network managed by a
number of totally independent corporations.
Consistent with the intent of the MFJ, the work of the CSO is directed
by the RBOCs. The CSO is owned equally by each of the seven RBOCs,
and its board of directors is composed of BOC and RBOC officers. CSO
work plans and budgets are determined by a committee structure that
includes RBOC representatives at all levels of management.
During 1982, the BOCs determined the initial set of functions to be
performed by the CSO. The vast majority of these functions are technical
in nature. These technical support functions include:

• network planning, which includes analysis and planning for new technologies and services, participation in the development of network
standards, and operations systems planning
• engineering and operations support related to procedures and standards
for network operations and quality assurance implementation
• information systems development for the many current operations systems
whose development is assigned to the CSO and for future systems
• technology systems, which includes generation of generic requirements
for equipment, technical analysis of products for the network, and
quality assurance methods
• applied research in the physical sciences, mathematics, computer science, network technology, and new services.
The CSO also provides support for the BOCs in areas such as market
research, regulatory matters, and other financial and administrative
matters. The initial staff of the CSO is drawn primarily from Bell Laboratories, AT&T, Western Electric, and the BOCs.

AUTHORS

J. A. Civarra
B. R. Eichenbaum

J. A. McCarthy
J. A. Schelke

2
Services

2.1 INTRODUCTION
Like other aspects of the Bell System, services are dynamic: New services
are introduced frequently in response to evolving customer needs and
the capabilities of new technology. Consequently, at any given time,
there are many new services in various stages of planning and development. This chapter covers only services currently available and some services with an announced availability date. The next few paragraphs discuss some general concepts and terminology related to services, followed
by descriptions of specific services.

2.1.1 BASIC AND VERTICAL SERVICES
The concept of basic and vertical services has been central to the Bell
System's role as a regulated monopoly providing telecommunications services. For several decades, basic service has been coupled to the Bell
System's goal of universal service (see Section 17.4.1), that is, that telephone services should be available to every home in America at an
affordable price. Basic service represents that universally available and
affordable service.
Basic residence service, for example, generally implies that, for a fixed
monthly charge, a customer receives the following:
• a standard, rotary-dial telephone set
• on-premises wiring
• a network access line-the connection to a local switching system for
local calling and for access to the network
• a listing in the white pages of a directory.

39

40

Introduction to the Bell System

Part 1

All other services have been classified as vertical services for which customers pay a charge that is additional to the cost of basic service. Vertical
services may provide greater convenience, more attractive telephone sets,
or additional functions or features beyond basic service. Traditionally,
revenues from vertical services have helped to maintain an affordable
price for basic service. However, as competition replaces regulation in
the telecommunications business, the price of a particular service will
become more closely related to the cost of providing that service.
Section 2.2 describes vertical services related to station equipment.
Section 2.3 describes customer switching services, vertical services used
almost entirely by business customers. Sections 2.4 and 2.5 describe vertical services available through access to the network.

2.1.2 ORDINARY AND SPECIAL SERVICES
From the telephone company viewpoint, services are classified as ordinary or special. Ordinary services usually include residence service, public
telephone service, mobile telephone service, and basic individual-line
business services. All other services are considered special services (often
called specials). Special services require special treatment with respect to
transmission, signaling, switching, billing, or customer use and are used
mostly by business customers. The overall demand for special services is
growing twice as fast as the demand for ordinary telephone service.
Examples of special services described in this chapter include foreign
exchange service, Wide Area Telecommunications Services, private
branch exchange, and centrex services, and private-line and private network services. There are about twenty-five major categories of special
services.
Facilities provided by the Bell System for other communications firms
form an important and rapidly growing class of special services. These
offerings include local distribution capability, interoffice facility sections,
and access to the Bell System network for resale carriers and other common carriers (OCCs), including domestic satellite carriers, international
record carriers} and local radio common carriers. To provide communication services to their customers, these competing firms use Bell System
facilities in conjunction with their own facilities or with facilities rented
from independent telephone companies. Facilities offered to other carriers parallel the wide variety of general special services that the Bell System provides to its own customers: voice, data, telegraph, and television.
1 International record carriers, such as International Telephone and Telegraph, RCA Global
Communications, and Western Union, offer data and message services (like telex)
internationally.

Chap. 2

Services

41

2.2 TERMINAL PRODUCTS
Terminal equipment is the principal interface between the customer and
the nationwide telephone network. It ranges from the basic telephone
set, which provides voice services, to the more versatile and specialized
equipment that can link with computers and provide additional services
such as data and graphics transmission.
This section describes terminal equipment available from the Bell System. In the past, all Bell System terminal equipment was leased to customers, but now, as the result of an order issued by the Federal Communications Commission (FCC) in 1975, customers have the option of
purchasing it (see Section 2.7). Customers may purchase and install
equipment from any manufacturer, provided the sets meet certain registration requirements (see Sections 8.7 and 11.1.2).

2.2.1 TELEPHONE SETS AND VERTICAL SERVICES
The telephone set is an important element of the communications network and provides access to a variety of services. The traditional rotarydial set is available in desk and wall models and in various colors. It
represented approximately 50 percent of all residential sets in service in
1980.
A number of premium telephone sets and decorator telephones are
also available. (Figure 2-1 shows a group of DESIGN LINE2 and other
decorator telephones.) Premium sets have special features that make the
use of the telephone more convenient or pleasant and provide the same
access to the network as standard sets. There is an additional charge for
premium sets, an example of a vertical service. The PRINCESS2 telephone, a premium telephone introduced in 1959 as the "bedroom set,"
offers special features such as a lighted dial and a night light. The
TRIMLINE2 telephone features the dial in the handset and a "recall" button that allows the user to make consecutive calls without hanging up the
handset.
DESIGN LINE decorator telephones use the same internal components
as standard rotary or TOUCH-TONE 3 telephones but have distinctive
housings. These sets are sold outright to the customer-there is no
monthly service charge-and are covered for a limited warranty period.
After this period, customers can purchase a maintenance contract for the
internal mechanism.

2 Registered trademark of AT&T Co.
3 Trademark of AT&T Co.

Figure 2-1. Some DESIGN LINE and other decorator
telephones, including designs from non-Bell manufacturers.

Other vertical services related to telephone sets are TOUCH-TONE 4
dialing and extension phones. TOUCH-TONE dialing is faster and more
convenient for the customer. Faster dialing also offers advantages to the
Bell System, because switching equipment is more quickly available for
other calls. TOUCH-TONE dialing will be required for many telephone
4 Registered service mark of AT&T Co.

42

Chap. 2

Services

43

services that are expected to grow in the future, for instance, banking by
telephone and other services that require access to a computer.
Extension phones make single-line telephone service in the home
more convenient. In addition, the use of a second telephone line is
becoming more prevalent. A series of 2-line sets that provide hold and
signaling features is now available. A customer can place one call on
"hold" and answer a second call, or the customer can signal another person in the house that there is an incoming call on the second line.

2.2.2 MODULAR TELEPHONES AND RETAIL SALES
The telephone sets now available from the Bell System are modular. A
modular set has plug-ended cords that connect the telephone base to the
handset and wall connector, permitting installation and removal by the
customer. Modular sets provide the customer with faster and more convenient service because customers can pick up these sets at retail sales
locations and do not need to schedule installation (see Section 2.7.1).
This also avoids the cost of a home visit.

2.2.3 DATA PRODUCTS
As computers and other sophisticated business machines become more
commonplace, data transmission is becoming an increasingly larger part
of almost all business communications. For many years, the Bell System
has offered a wide variety of data products to satisfy the needs of customers. This section describes two general categories of data products: data
sets and data terminals.

Data Sets
Digital computers and various types of data terminal equipment produce
data in digital form, that is, as a sequence of discrete electrical pulses.
While digital transmission facilities, which transport data in digital form,
are rapidly being deployed in the telecommunications network, analog
transmission facilities, which transport data as continuous electrical
waves,s still represent a greater share of the total network. Data sets (also
called modems) provide the conversion and control functions required to
transmit digital data over analog facilities. The Bell System has a number
of DATAPHONE6 data sets available with different capabilities and
5 Chapters 6 and 9 discuss digital and analog transmission.
6 Registered trademark of AT&T Co.

44

Introduction to the Bell System

Part 1

features suitable for a wide range of applications. Section 11.1.2 contains
a functional description of data sets and some of the specific types and
characteristics.
In general terms, the DATAPHONE data communications service
categories associated with DATAPHONE data sets are based on the type of
analog facilities used: narrowband, voiceband, or broadband/ private
line or switched. Data sets are primarily categorized by the data transfer
rates they provide:
• low speed -

up to 300 bits per second (bps)

• medium speed - 300 bps to 9600 bps (9.6 kbps)
• high speed - over 9.6 kbps.
In the case of the public switched telephone network (PSTN), including
Wide Area Telecommunications Services and foreign exchange lines (see
Section 2.5.1), DATAPHONE data sets permit customers to send data
between any two locations served by the network at speeds up to
4.8 kbps. A telephone set is used for setting up the channel and for alternate voice communications. The calling and answering can also be controlled automatically by the customer's computer or data terminal.
Automatic calling and answering take place through appropriate interaction with the data sets and associated automatic calling units. These units
dial, connect, and terminate calls.
DATAPHONE data communications service is also provided on
voiceband private lines. In this service, as with DATAPHONE data communications service for PSTN applications, a DATAPHONE data set is
used on the customer's premises. Arrangements can be added to permit
voice communications on the private lines and to permit access to the
PSTN for service backup in case of a private-line outage. Voiceband
private-line configurations can be point to point or multipoint; the latter
is more prevalent. Speeds up to 9.6 kbps are offered.
DATAPHONE II data communications service, the most recent offering
on voiceband private lines, employs a series of advanced microprocessorbased data sets. These sets provide considerable built-in diagnostic, testing, and on-line monitoring capabilities for a data network using diagnostic control devices at a customer's central computer site. (See
Figure 2-2.) This capability is particularly important to customers such as
airlines and banks who have data networks with real-time applications.
These cqstomers purchase terminals, computers, etc., from many
suppliers, and they must quickly identify the vendor responsible for
fixing a trouble.

7 See Section 6.2.1.

DIAGNOSTIC CONSOLE

2096C

Figure 2-2. The DATAPHONE II data communications service
new family of data products includes data sets, a diagnostic
console , and a network controller.

High-speed data sets are used with analog broadband services. Both
point-to-point private-line and switched offerings are provided. Speeds
on broadband private-line channels range from 19.2 to 230.4 kbps . Broadband services find the greatest use in applications involving computerto-computer transmission of large amounts of information.
The Bell System also provides several auxiliary sets for use as adjuncts
to data sets. The automatic calling unit mentioned earlier is a commonly
used auxiliary set. Other data auxiliary sets perform functions such as
signal conversion and testing.
Data Terminals
Data terminals are the end points in data communications. They originate and/or receive data transfers. Teletypewriter, telegraph, and remote
metering terminals can be used with low-speed data sets to transmit data
on narrow-bandwidth analog circuits. Cathode-ray tube (CRT) terminals,
certain teletypewriters, and line printers along with medium-speed data
sets use voiceband private lines or the PSTN for data transfer. Some

45

46

Introduction to the Bell System

Part 1

facsimile terminals, CRT terminals, and high-speed line printers are connected to high-speed data sets to transmit and/or receive data over broad8
band analog channels.
The outputs from a number of terminals, each connected to a data service unit, may be multiplexed (combined) for transmission over the Digital Data System (DDS) network, which is described in Section 11.6.1.
The data service units perform functions similar to those performed by
data sets, except that conversion between analog and digital formats is
not required since the DDS provides end-to-end (terminal-to-terminal)
digital transmission. Computers may also appear as high-speed data terminals connected to the DDS network via an appropriate data service
unit. The service provided the DDS network is called DATAPHONE digital service (see Section 2.5.4).
The Bell System offers many different types of terminals. Some of the
more recent offerings are in the DATASPEED 9 terminal set line (shown in
Figure 2-3), a collection of high-speed data communications terminals that
includes CRT terminals, line printers, and intelligent interactive
input / output terminals. (Intelligent terminals typically contain logic and
memory capability.) New disk storage devices, offered with certain

Figure 2-3. DA TASPEED 4540 line of data terminals.
(Courtesy of Teletype Corporation)

8 Table 2-1 in Section 2.5.2 lists narrowband, voiceband, and broadband offerings.
9 Registered trademark of AT&T Co.

Chap. 2

Services

47

teleprinter and CRT terminals, permit message preparation and storage to
improve interaction with a remote host computer. Section 11.1.3 provides
more details on these data terminals.

2.2.4 SPECIAL-PURPOSE TERMINALS
Special telephone sets and auxiliary equipment make other services available to both residential and business customers. One such service is
automatic dialing from a directory of stored numbers. Some telephones
use punch-coded cards (card dialers). Others, like the TOUCH-AMA TlC lO S repertory dialer, automatically dial a number at the touch of
one button. The repertory of numbers is stored in electronic memory
(see Figure 2-4). Many of these telephones automatically store the last

Figure 2-4. TOUCH·A·MA TIC S series telephone , the Bell System ' s first
microprocessor-based telephone for the home. Important or freque ntl y called
numbers can be dialed at the touch of a button. Red and green buttons at the
top of the panel identify emergency numbers .

10 Registered trademark of AT&T Co.

48

Introduction to the Bell System

Part 1

manually dialed number and redial it when the user pushes a button.
Some designs have lights on their panels to help locate emergency
numbers.
Auxiliary equipment, including hands-free telephone equipment
together with electronic graphics, is also available for teleconferencing
between two or more groups. The 4A speakerphone, which is used by
both residence and business customers, has an omnidirectional microphone and an adjustable loudspeaker to permit hands-free conversation
and allow a group of people to participate in a conversation (see
Figure 2-5) . Portable conference telephone units are available for conferences or for communication between a student confined at home and a
classroom.

Figure 2-5. 4A speakerphone. The loudspeaker is at the left . The transmitter
unit has a volume control, an ON / QUIET-OFF switch, and a solid-state ON / OFF
indicator light. The microphone is located under the switches.

The GEMINI ll 100 electronic blackboard system uses a graphics
transmission terminal that can send handwriting over conventional
VOice-grade telephone lines. It can be used for remote teaching and
teleconferencing at a number of different locations simultaneously. (See
Figure 2-6.)
11 Registered trademark of AT&T Co.

Figure 2-6. A demonstration of the GEMINI 100 electronic blackboard.

Special-purpose terminals for business customers, like the Transaction
telephone, can automatically dial a non-Bell System credit service center
or data base for credit authorization or check verification. In a typical
transaction, a cashier slides two cards (the merchant's identification card
and the customer's credit card) through a magnetic strip reader in the
telephone set (see Figure 2-7) and then keys in the sale price on a
TOUCH- TONE telephone dial pad. The terminal automatically dials a
computer in a bank or credit agency and obtains a purchase authorization
as an audio response, a light, or a visual display.

2.2.5 AIDS FOR THE HANDICAPPED

Special Bell System equipment gives disabled persons access to basic telephone service . Someone with impaired mobility may find automatic dialing telephones and speakerphones easier to use than standard sets (see
Section 2.2.4). For people who have lost the use of their laryn x, or voice
box, there is the artificial larynx (invented at Bell Laboratories in 1929),
which replaces the vibrations of normal vocal cords with electronically
controlled vibrations that can be formed into words (see Figure 2-8).
Several different aids are available to persons with impaired hearing.
Among these are amplifying handsets, sets with tone ringers, and sets

49

Figure 2-7. The Transaction III terminal with a Transaction printer.

Figure 2-8. The artificial larynx in use.

Services

Chap. 2

51

that can be equ ipped with loud bells. The CODE-COMI2 se t converts
sound into either visual signals- a flash ing light-or tactile signals-a
vibrating disk (see Figure 2-9) . To alert a d eaf person to an incoming
call, an ordinary household lamp m ay be plugged into the
SIGNALMAN 12 relay switch, which causes the lamp to flash on and off.
Alterna tively, an electric fan can be plugged into the unit to signal someone who is bo th deaf and blind by blowing air on the person .

Fi9ure 2-9. COM· CODE set for the handicapped. When connected to
co nventi onal telephones, this device allows a deaf person to re c eive messages
via flashes of light or vibrations . The circular vibrating pad is shown on the left
of the device , the sending key, which is used like a telegraph key, on the rig ht.
Light flashes come from a recess (the black rectangle) to the left of the be ll
symbol.

12 Trademark of AT&T Co.

52

Introduction to the Bell System

Part 1

2.3 CUSTOMER SWITCHING SERVICES
The term~ustomer switching service describes an arrangement that permits
flexibility· in the connections between one or more telephone lines and
one or more station instruments. 13 Services can be tailored to the individual needs of the customer and may involve additional equipment
either on the customer's premises or in a central office.

2.3.1 CUSTOMER NEEDS
Customer switching services are provided primarily for business customers, although the communications needs of some very large residences
may overlap those of a small business. This section discusses only business applications.
While the specific communications needs of business customers
depend on the size and nature of their businesses, for discussion purposes, these requirements may be divided into four broad categories:
intralocation calling, incoming calls, outgoing calls, and communications management.
The first requirement is intralocation calling. Many businesses need to
communicate between stations on the same premises. This is known as
intercom calling. For voice communication, intercom calling may mean
calls between people in different offices, access to an on-premises paging
system, or dial access to a customer-owned recorded dictation system. For
data communication, it may involve communications between a terminal
and a computer or another terminal on the same premises.
The second category involves varying requirements for handling
incoming calls. For business customers, incoming calls are important since
they often represent new or additional business. Most customers, therefore, want to present a good telephone image to calling parties. This usually means having enough lines and attendants to ensure that incoming
calls are promptly answered and efficiently passed to someone who can
help the calling party.
The third requirements category co·ncerns outgoing calls. The importance and nature of outgoing communications are, of course, a function of
the customer's business. When employees spend a significant part of
their workday making telephone calls, it is important to make those communications as convenient and friendly as possible. Station equipment
with button-operated features and switching systems with automatic callprocessing routines meet these needs. Control of outgoing call possibilities from selected lines may also be important to avoid abuses.

13 Two or more telephone instruments permanently connected to a single telephone line is
not a customer switching system.

Chap. 2

Services

53

The fourth category concerns the different needs of businesses in
managing communications. Since communications can be a major expense,
many customers have one or more managers in charge of communications
facilities. To help these people do an effective job, modern equipment
furnishes them with management information (data) about the use and
performance of the communications service it provides.
The rest of Section 2.3 discusses the service needs of small-, medium-,
and large-size business customers as well as the needs of some special
customers.

2.3.2 SMALL-SIZE BUSINESS CUSTOMERS
The small business customer may be a doctor, a realtor, or a car dealer,
for example. The needs of these customers are relatively simple. For
incoming service, they need one or a few incoming lines. Depending on
the nature of the business, these lines may be in one group accessed by a
single listed directory number, or they may be individually listed so that
the calling party can dial a specific number to reach a specific line. On
the customer's premises, calls will be either answered at a central location
and then passed to the desired person or answered at the specific dialed
locations. In the latter case, there will be less need for call passing.
Call passing can be done in several ways. The desired person may be
paged and requested to pick up a specific line. Another method uses
switching system features that permit the transfer of the call to a telephone near the person.
In very small customer applications, where face-to-face communication
is usually quite convenient, station-to-station intercom calls are often not
required. These applications, however, may need paging or intercom
calling to accomplish call passing. Some means may be required to hold
an ongoing call temporarily so that a person may talk privately to someone else or answer another call.
The small business customer's outgoing call requirements are also
simple. Like residence customers, most small business customers use
their line(s) for both incoming and outgoing calls and are billed for each
toll call they originate. Control of outgoing calls is also of minimal concern to the small customer because the person responsible for telephone
charges can personally discuss problems with the individuals involved.
The needs of small business customers are often met by key telephone
systems, which allow the station user to originate or answer a call on one
of several lines by operating a button (key), and provide features such as
hold, intercom calling, and message-waiting lights. Modern small communications systems (see Section 11.2.2) using integrated circuit technology, multibutton electronic telephones, and microcomputer control provide numerous additional features under software control.

54

Introduction to the Bell System

Part 1

2.3.3 MEDIUM-SIZE BUSINESS CUSTOMERS
The medium-size customer may be a business occupying a small office
building, a manufacturer, or a municipal office such as a police station.
An attendant at a central answering position most often handles incoming calls to the 20 to 200 on-premises lines representative of a mediumsize customer. The physical size of this customer's organization makes
convenient intralocation station-to-station calling more important, and
outgoing traffic can range from a modest communications activity to a
major one. Typically, management of communications services for this
size customer is a part-time assignment for one person.
The customer switching service most often provided for medium-size
business customers is known as a private branch exchange (PBX) service. 14 A
PBX is a relatively small telephone switching system (exchange) located on
the customer's premises (private) and connected to a central office (as a
branch). The basic features of a PBX provide for the central answering
position, convenient station-to-station intercom calling, and whatever
special incoming and outgoing call features are necessary.
Customers in this size category often make and/or receive a substantial volume of calls to or from one or more distant areas of the country.
To meet the needs of these customers, the Bell System offers Wide Area
Telecommunications Services (WATS), which are described in Section
2.5.1. With WATS, calls to relatively large geographic areas are billed on
a bulk basis rather than individually. The actual cost of a given call can
therefore vary considerably depending on both the area called and the
type of outgoing line-WATS or regular direct distance dialing (DOD)selected.
In most cases, a medium-size customer will have a small number of
different outgoing lines to access various geographic areas. Station users
can then usually select the most economical line (that is, the most
economical route or service) for their calls, depending primarily on the
call's destination. Station users can be provided with a map or a list of
area codes as the basis for selection. Other factors that may influence
route selection include the time of day and whether any outgoing lines,
such as WATS lines, are already busy.
In more complex installations, route selection by the individual station
user is often not practical. Installations like these with calls to many geographic locations and with a greater number of different outgoing lines
usually require outgoing calls to be placed through attendants. The
attendants are specially trained to select the optimal route. In the most
modern PBXs, this function can be performed by a computer within the
PBX using software programs to provide a feature known as automatic
route selection (see Section 2.3.4).

14 To distinguish from early manual cord boards, current automatic systems are sometimes
called private automatic branch exchanges (PABXs).

Chap. 2

Services

55

PBX service also meets other needs of the medium-size business customer. For example, it is usually possible to transfer a call from one station to another or to have a secretary pick up a telephone that is ringing
in another office. As an alternative to PBX service, which is a customer
switching service, a medium-size customer may select an exchange service (see Section 2.4), known as centrex service, that provides a PBX-type
service.

2.3.4 LARGE-SIZE BUSINESS CUSTOMERS
Large customers have from about 200 to 10,000 or more lines. Typical
locations are headquarters buildings, large banks, and combined design
and manufacturing locations. At this size, centralized answering becomes
less attractive, and a means of direct inward dialing 15 (DID) to station
lines is often provided. Intercom calling is more important, and outgoing
calls become a major concern. Most large customers also have some
requirement for digital data communications on the same premises. All
of these needs are satisfied by large PBXs.
The number of outgoing calls from a large system usually justifies the
purchase of special long-distance facilities such as WATS lines and
foreign exchange (FX) lines (see Section 2.5.1) that go directly to frequently called areas served by a distant central office. To use these special facilities in the most efficient way, modern PBX systems provide
automatic route selection, a service that automatically analyzes a dialed
number and selects the least-cost type of line (that is, type service) available for the call. Special arrangements are also available to distribute a
relatively large number of incoming calls efficiently to a special group of
lines (such as a customer service department).
In large companies, millions of dollars per year may be spent on communications. For this reason, large companies assign a small staff to
manage communications services. The PBX system must provide management information for use by this staff. Modern systems not only can collect general data on system traffic (calls) but also can make a detailed
record of each outgoing call, so that the costs of the communications
facilities can be allocated to users.

2.3.5 SPECIAL CUSTOMERS
An airline reservation center is an example of a special customer. This
type of business has a large number of incoming calls that must be
evenly distributed among the reservations agents. Automatic call distributors (ACDs) provide this service. The operating costs and performance of
an ACD greatly depend on having the correct number of agents on hand
15 With direct inward dialing, the telephone number of the called party can be dialed directly;
the call need not be passed by an attendant.

56

Introduction to the Bell System

Part 1

at all times. For this reason, modern ACDs also provide the customer
with traffic data (i~ either raw or processed form) that can be used to
manage the work force.
Another special customer is the telephone answering bureau, which
answers other people's (clients') telephone calls. Historically, these customers have been served by providing the bureau with an extension of
the client's phone. However, the service can now be implemented by
means of a Call Forwarding feature available in modern electronic switching systems. With this feature, calls can be redirected (or "forwarded")
from the original number to a special line at the answering bureau,
where messages are taken for relay to the client. (See Section 11.2.7.)

2.4 EXCHANGE SERVICES
The term exchange services describes those services provided through the
local or exchange area network. Access to the network and its services is
obtained by one or more lines that connect the customer's station set(s) to
the central office. Customers can choose from exchange services that
range from basic local calling with standard rotary-dial telephones to
PBX-like business services.

2.4.1 EXCHANGE LINES AND LOCAL CALLING
A single business or residence line connected to a rotary-dial telephone
provides basic exchange services such as local calling and the 911 Emergency Service described in the next section. Local calls are calls to any
customer in the local calling area of the calling customer's central office.
A local calling area, or exchange area,16 is a geograpbic area within which a
strong community of interest exists (that is, heavy calling volume among
customers within the area). It may be served by several central offices.
Basic exchange service also includes operator assistance on local calls and
directory assistance 17 (see Section 2.5.1). This basic service is typically
provided under a tariff18 that allows the customer either flat-rate, or
measured-rate billing. With flat-rate billing, the customer can make an
unlimited number of local calls for a fixed monthly charge. With
measured-rate billing (also called measured service), the customer pays a
lower fixed rate plus an additional charge for all local calls in excess of a

16 This discussion is based on the traditional definition of exchange areas. The term was
used with a different meaning in the 1982 Modification of Final Judgment.
17 Some companies now charge for directory assistance calls when they exceed a fixed
monthly allowance.

18 Tariffs are rates and conditions for various telephone services (see Section 17.3).

Chap. 2

Services

57

specified monthly allotment. The trend in the Bell System is toward
measured service.
In some locations, a customer may have the option of subscribing to
single-party or multiparty service. With multiparty service, the line is
shared by two or more customers; however, only one party may use the
line at a time except when a call is between two customers on the line.
Various ringing combinations are provided to indicate the destination of
incoming calls. In 1981, about 97 percent of the Bell System was single
party, with the trend toward total single-party service.

2.4.2 CUSTOM CALLING SERVICES I
Custom Calling Services I is a group of four features that take advantage
of the stored-program control of electronic switching systems (see Section 10.3.1): Call Waiting, Call Forwarding, Three-Way Calling, and
Speed Calling.
Call Waiting
This feature allows a customer engaged in a call to be reached by another
caller. A short tone informs the customer that another call is waiting to
be accepted. The tone is heard only by the Call Waiting customer; the
caller hears the regular audible ringing. The customer can place the first
call on "hold" and answer the second call by momentarily depressing the
switchhook ("flashing"). By subsequent flashes of the switchhook, the
customer can alternate between the two calls. 19
Call Forwarding
This feature allows customers to "forward" their calls to another telephone number, which they desIgnate by dialing a special code sequence.
While Call Forwarding is activated, all incoming calls to a customer's telephone line are automatically transferred to the designated telephone.
Three-Way Calling
This feature allows a customer involved in an existing 2-way connection
to place the other party on hold and dial a third party for a 3-way connection. When the third party answers, a 2-way conversation can be held
before the earlier connection is re-established for the 3-way conference.

19 In some business services provided by electronic switching systems, the procedures
related to Call Waiting may be different.

58

Introduction to the Bell System

Part 1

Speed Calling
This feature allows a customer to use abbreviated codes to dial frequently
called numbers. Repertories of eight (using a 1-digit abbreviated code)
and thirty (using a 2-digit abbreviated code) stored numbers are available. 2o Speed Calling customers who also have the Customer Changeable
Speed Calling option can assign their own Speed Calling codes to telephone numbers directly and immediately from their own telephones.

2.4.3 TOUCH-TONE SERVICE
TOUCH-TONE service replaces the customary dial pulses with tones for
network signaling. As a pushbutton is depressed, two tones are generated simultaneously and the combined signal, which is clearly audible
to the caller, is transmitted to the central office. Special receivers located
in the central office convert the signals into a form that can be used by
the switching system. TOUCH-TONE service provides customers with
improved speed and convenience in dialing, reduces the number of digit
receivers required by the central office because faster dialing uses the
digit receivers for a shorter time per call, and provides the capability for
end-to-end signaling once a call is established. End-to-end signaling (a
capability that does not exist with rotary-dial service) allows a customer at
one end of a connection to control dictation units and access data bases at
the other end of the connection. The use of TOUCH-TONE service is
increasing and will be the dominant method of customer signaling in the
future.

2.4.4 EXCHANGE BUSINESS SERVICES
Business services offered from the exchange network satisfy many of the
same business customer communication needs served by the PBX and
automatic call distributor (ACD) services described in Section 2.3. With
PBX and ACD services, individual customer lines connect to switching
equipment on the customer's premises, and the switching equipment connects to a switching system in a local central office. With the corresponding exchange services, all of the customer's subscriber lines are directly
connected to the central office, thus reducing the amount of equipment
required at the customer's premises. At the central office, software or
wired logic indicates which subscriber lines are part of the customer's
group. By providing special features in the central office for this group
of lines, both PBX-type and ACD-type services can be emulated.
20 These numbers may be different in some electronic switching system business services.

Chap. 2

Services

59

The primary business services offered from the exchange network are
centrex, ESS21X-l, and ESS-ACD.22 Centrex and ESSX-l provide the same
basic service elements:
• A member of a centrex or ESSX-l group can dial another telephone
number within the same group using only one to five digits.
• A member can dial calls outside the group directly, typically after
dialing an access code, such as the digit 9.
• A member can receive calls that originate outside the group directly.
No attendant is needed.
• Attendant positions can also be provided to allow central answering
positions on the customer's premises to answer, hold, and route
incoming calls to the group when the main centrex or ESSX-l telephone number has been c.alled.
With ESSX-l service, the number of simultaneous incoming and outgoing calls and the number of simultaneous intragroup calls are limited by
software to sizes specified by the customer. For centrex, however, the
only limit is the call-handling capacity of the switching system.
The exchange service counterpart to ACD service (see Section 2.3.5) is
called ESS-ACD. It is a specialized form of centrex service in which central office equipment, specifically an electronic switching system, distributes incoming calls to attendant lines. Typically, ACD attendants work
full time receiving and servicing incoming calls (for example, making airline reservations). Therefore, in order to keep attendant positions
efficiently loaded, there are generally fewer active' attendants than the
maximum number of simultaneous incoming calls.
With ESS-ACD, the central office distributes calls uniformly to the
attendants, thus spreading the workload to minimize caller delay and
maintain attendant efficiency. If no attendant is available, the central
office will queue calls in order of arrival (see Section 5.2) and distribute
them as attendants become available.
Additionally, specially designed customer-premises equipment and
data links between the central office and the customer location make a
large variety of management information, control, and status display
features available. A customer can use statistical performance information and control capabilities to adjust the number of active positions and
thus the average time a caller waits before reaching an attendant.
21 Trademark of Western Electric Co.
22 ESSX-1 and ESS-ACD are services provided by the lESS switching equipment.

60

Introduction to the Bell System

Part 1

2.4.5 911 EMERGENCY SERVICE
911 Emergency Service is designed to provide free emergency calling
capability to the general public and is used in conjunction with dialtone-first service. 23 The cost of implementing and maintaining the service
is typically paid by county and state governments. With 911 Emergency
Service, a single, easily remembered telephone number accesses a variety
of emergency agencies. The service was established by the Bell System in
1968 in response to a recommendation by a Presidential Commission on
Law Enforcement and Justice. The Commission had recommended that
"wherever practical, a single (police emergency) number should be
'established. "
Originally, 911 systems simply routed emergency calls to a centralized
answering point. Later, features were added to this Basic 911 (B911) service to provide for forced disconnect of the calling line (to prevent tying
up the emergency center with nonemergency calls); holding the connection regardless of the calling party's action; emergency ringback to the
calling station; and a visual and audible indication of the switchhook
status of an established 911 call.
The major difficulty in implementing B911 systems is that, in many
places, the boundaries of emergency agencies do not coincide with the
boundaries of the local areas served by a telephone company. In some
places, one local area may have twenty or more different combinations of
emergency jurisdictions. When this happens, emergency calls must be
selectively routed to the correct emergency agency based on the location
of the calling party. The Enhanced 911 (E911) provides the routing logic
required to solve this problem (see Figure 2-10). Other features available
with E911 include the ability to display· the telephone number aI).d the
address of the calling party at the public safety answering point (PSAP),
generally, a police station.
Approximately 800 B911 systems covering 25 percent of the population of the United States were in service by 1980, along with a total of 8
E911 systems covering a population of nine million. The potential exists
for over 100 E911 systems to be in service by 1986. More efficient handling of emergency calls with these 911 systems will undoubtedly result
in significant savings in life and property.

2.5 NETWORK SERVICES
In addition to exchange services, which are limited to the capabilities and
resources of the exchange area network, the Bell System offers network
services that make use of the broad capabilities of the PSTN, including
23 Section 2.6 describes public telephones and dial-tone-first service.

Figure 2-10. Role of an E911 office in routing calls directly to public safety
answering points (PSAPs), police stations in this figure. The area shown
above has four emergency agency jurisdictions. A 911 call from Party A would
be routed to police station A, while a call from Party B would be routed to
police station B. even though both callers are served by the same end office.

the stored-program control network (see Section 11.3.1); private networks; and data networks. The following sections describe these three
types of network services.

2.5.1 PUBLIC SWITCHED TELEPHONE NETWORK SERVICES
A PSTN service, toll service, is used whenever calls are placed to points
outside the local calling area. These calls are referred to as toll calls, and
customers are ordinarily charged for each call. Other major PSTN services include operator services, foreign exchange service, Wide Area
Telecommunications Services, services provided by the direct services
dialing capability, Automated Calling Card Service, DIAL-IT network
communications service, and services provided by the circuit-switched
digital capability.

61

Introduction to the Bell System

62

Part 1

Operator Services
Telephone company operators provide a variety of services to PSTN
customers. 24
• Toll-and-assistance operators directly assist in the completion of calls.
Toll-and-assistance operators interact with customers making calling card, collect, and person-to-person calls. They may also assess
charges and control the collection and return of coins for some
coin calls, and place calls to points that cannot be directly dialed,
such as certain mobile radios, marine stations, and certain foreign
countries. These operators provide special services such as conference and call-back calls and perform manual switching, where
needed. (Very few calls are switched manually because the use of
direct distance dialing is widespread.) Toll-and-assistance operators assist customers who require emergency help or who are having trouble with the network. They also verify the busy-idle status
of lines, accept requests for credits, provide dialing instructions,
and complete calls when the customer cannot or when a customer
experiences transmission problems. In today's environment, many
of the operator functions on a toll call have been automated. 25
Centralized automatic message accounting - operator number
identification operators (CAMA-ONI operators) obtain the calling
customer's number where the switching system does not include
automatic number identification (ANI) equipment. The calling
number is supplied to CAMA equipment (see Section 10.5.4) for
billing purposes. The aNI function, while included for completeness, is not an actual service to the customer.
• Number-service operators provide information necessary for the completion of calls.
Directory assistance operators respond to customers dialing 411
for the local area code and 555-1212 for nonlocal area codes.
Intercept operators handle calls to unassigned or changed
numbers.
Rate-and-route operators assist toll operators.

24 Section 4.2.2 describes how operator services are integrated into the network.
25 Section 10.4 discusses automated systems in greater detail.

Chap. 2

Services

63

Foreign Exchange Service
Foreign exchange (FX) service enables a customer to be served by a distant or "foreign" central office rather than by the nearby central office.
Calls to other customers in the distant exchange area are then treated as
local calls instead of toll calls. For customers who make enough calls to a
particular distant exchange area, the monthly charge for FX service is less
than the sum of the toll charges they would otherwise pay. Customers
who find FX service economical include residence customers who often
call friends or relatives in towns outside their local calling area and
businesses such as firms in New Jersey who often call companies in New
York City.

Wide Area Telecommunications Services
There are two types of Wide Area Telecommunications Services (WATS):
inward WATS (INWATS), also called 800 Service, and outward WATS
(OUTWATS). They permit a customer, respectively, to receive from or
originate to selected service areas long-distance calls that are billed to the
customer on a bulk basis rather than on an individual basis. Both of
these services are available on an intrastate or interstate basis. Subscribers are predominantly businesses with a substantial volume of longdistance calls to or from a wide geographical area.
For an interstate WATS customer, the United States is divided into six
service areas, or bands, that extend outward from, but do not include, the
customer's home state. Service Area One contains the states contiguous
to the home state (but not including it) and sometimes one or two nearby
states. Service Area Two includes Service Area One plus certain other
states. Each successive service area includes the previous service area
plus additional states. Service Area Six encompasses the entire United
States (including Alaska, Hawaii, and Puerto Rico) but not the home state.
Intrastate WATS is also available in most states. Under present tariff provisions, customers must purchase separate dedicated access lines to terminate interstate and intrastate WATS calls.
Expanded 800 Service, an improvement over 800 Service, uses
common-channel interoffice signaling}6 to provide three features that
give customers greater flexibility in defining service areas and determining the treatment an incoming call receives. Single-Number Service provides subscribers with one nationwide 800 number for both interstate and
intrastate calls at one or more customer locations. With Customized Call
Routing, customers can control their call distribution based on callers' area

26 Sections 8.4.2 and 8.5.5 discuss common-channel interoffice signaling.

64

Introduction to the Bell System

Part 1

codes. Variable Call Routing allows the customer to specify call distribution based on the time of day and day of week. Customers with 800 Service can add any or all of these features to meet their needs.

Services Using the Direct Services Dialing Capability
Traditionally, the Bell System has designed services to meet specific customer needs. Recently, however, there has been a trend to meet customer
demand for new network services by providing a collection of serviceindependent network capabilities called the direct services dialing capability.
This approach has many advantages. Customers can modify and control
their services to a degree not previously possible. These capabilities are
in the form of primitives in switching systems that can be summoned into
use for various service applications. Some useful primitives might be
"route the call" and "play an announcement." Services that would use the
"route the call" primitive include:
• routing calls to different locations specified by the customer based on
the location of the calling party, the time of day, the day of the week,
the digits dialed by the caller (in response to a verbal prompt), and
the busy-idle status of the customer's destination numbers. One application routes calls to the nearest retail store when there are several
located in a city or town.
• routing incoming calls to different locations specified by the customer.
One application has calls follow a salesperson who is moving from
location to location, based on input from the salesperson.

Autom~ted

Calling Card Service

Automated Calling Card Service offers customers the ability to charge
telephone calls to a number other than that of the originating station
without operator assistance. This service, available to business and
residence customers, automates calling card, bill-to-third-number, and collect calls. Automated Calling Card Service uses the direct services dialing
capability of the stored-program control network.
A feature that accompanies Automated Calling Card Service, but
which is not strictly part of it, is Billed Number Screening. This feature is
active on any bill-to-third-number call or collect call attempt and
identifies numbers that do not accept any bill-to-third-number or collect
calls. This applies to business or residence customers who have requested
such screening and also prevents those types of billing to public
telephones.

Chap. 2

Services

65

DIAL-IT Network Communications Service
DIAL-IT service is a name for any of several services in which callers dial

advertised telephone numbers to reach an announcement, a live answer,
or both. The '-services fall into two categories: Public Announcement
Service (PAS) and Media Stimulated Calling (MSC).
PAS plays up-to-date recorded announcements for such services as
Sports-Phone, Dial-a-Joke, and Horoscope. For these services, the Bell
System provides access to the announcement; the announcements themselves are provided by other companies.
Services provided by MSC include media promotion, telethons, and
telephone voting service. Media promotion and telethon services give
the customer the ability to connect selected callers to a live answer. Typically, callers not selected for a live answer receive a recorded announcement thanking them for their participation. The telephone voting service
allows callers to respond to questions presented to them by radio or
television. A caller dials one of two telephone numbers corresponding to
the caller's choice or opinion and is connected to a brief acknowledgement. The calls to each telephone number are tallied, and the
result is provided to the sponsor.
DIAL-IT service is available nationwide on a standard basis through
the use of area code 900.

Services Provided by the Circuit-Switched Digital Capability
The circuit-switched digital capability (CSDC) will provide end-to-end
digital connectivity. The CSDC, which is expected to be available as a
tariffed service around the end of 1983, is an important step toward an
integrated services digital network (ISDN), a public end-to-end digital
telecommunications network capable of supporting a wide spectrum of
present and emerging user needs. Like the ISDN, the CSDC will be service independent. The CSDC will provide a 56-kbps digital path over the
PSTN to customers whose lines are terminated on electronic switching
systems with the CSDC feature.
One of the first applications of the CSDC will be the transport of bulk
data. Since the CSDC will operate at a speed of 56 kbps, it will have ten
times the capacity of the 4.8-kbps data sets that are currently used on the
switched network. This feature can be useful to customers like banks
who must transfer large quantities (for example, tens of megabytes) of
information during a limited period, perhaps overnight.
New technologies and new capabilities will help make the integrated
services digital network a reality. Applications of new technology will
provide terminal equipment capable of integrating voice and data into a
single information flow. New service-independent capabilities, such as

66

Introduction to the Bell System

Part 1

those to be provided by the CSDC, will route this flow over the PSTN
and make many new services widely available.

2.5.2 PRIVATE-LINE SERVICES
Private-line services provide point-to-point and multipoint communication channels that are separate from channels of the PSTN. Private-line
circuits are usually used for talking and signaling, but other offerings are
available. These include teletypewriter services, telemetry,27 wired music,
video and television transmission, the connection of computers to other
computers or input/output devices for data transmission, the extension of
alarm or power control circuits from unattended to attended locations,
and the connection of radio or television studios to remote transmitters.
While many private-line services can be approximated using services
available on the PSTN, private lines offer the following advantages:
• Where the traffic is heavy enough and the geographic pattern lends
itself to such use, private lines may be more economical.
• A private line incurs a specified charge that is independent of the
amount of use.
• The time needed to establish a connection can be shorter with a
private line than with the PSTN.
• Private-line services are dedicated to the customer and not shared, as
in the PSTN, thereby ensuring a through (nonblocking) connection at
all times (see Section 5.2).
Private lines are offered in several designated series, which serve as a
basis for service negotiations between marketing representatives and customers. Different series lines have different uses and electrical characteristics. Table 2-1 lists the series numbers and types of service.

2.5.3 PRIVATE NETWORK SERVICES
Large business customers with geographically dispersed locations subscribe to private network services. Each of the customer locations is usually served by a PBX, centrex, or ESSX-l. As long as the calling volume
over the private facility is such that the toll charges for equivalent PSTN
calls are higher than the monthly charge for the dedicated facility, a
private network is cost effective.
27 Low-speed transmission of measured quantities. Generally, telemetry (or telemetering)
refers to an arrangement in which measurements taken in one place are recorded in
another place.

Services

Chap. 2

67

TABLE 2-1
PRIV ATE-LINE OFFERINGS

Examples of Service

Series

1000

Low-speed (narrowband *) data, for
example, private-line telegraph,
teletypewriter, teletypesetter, and
remote metering (telemetering)

2000

Voice

3000

Medium-speed (voiceband *) data

4000

Telephoto / facsimile

6000

Audio (music transmission)

7000

Television

8000

High-speed (broadband*) data

* Section 6.2.1 discusses voiceband, narrowband, and broadband
channels.

The simplest kind of private network would be a transmission facility
dedicated to a customer and interconnecting two geographically separated
customer PBXs or centrex/ESSX-l locations (in Los Angeles and New
York, for example). One PBX/centrex location calls the other by dialing a
code to access the other location and then dialing the extension number
of the station at the distant' location. This example is often not considered to be a true network and is usually referred to as tie-line servicethe transmission facility ties together the customer locations.
If there are three customer PBXs located in Los Angeles, Chicago, and
New York, the customer might acquire tie lines between Los Angeles and
Chicago and Chicago and New York. A caller at the Los Angeles PBX
who wishes to call the New York PBX first dials an access code to reach
the Chicago PBX and then dials another access code to instruct the Chicag~ PBX to connect the Los Angeles tie line to the New York tie line.
After the Chicago PBX makes the connection, the Los Angeles customer
dials the called person's extension number to complete the call through
the New York PBX. This type of service is known as Tandem Tie-Line Service because the Chicago PBX must be able to connect or "tandem" the call
between the two tie lines. With this type of service arrangement, a
different access code is required from each originating location to reach a
particular location. In addition, the customer must not only pay for the

68

Introduction to the Bell System

Part 1

dedicated transmission facilities but must also pay for PBX or centrex
switching capabilities to make direct or tandem connection to tie lines
(see Section 3.3.2).
To establish a network of tie lines with a uniform numbering plan
similar to that which exists in the PSTN, the customer must subscribe to
private network services like common-control switching arrangement
(CCSA), Enhanced Private Switched Communication Service (EPSCS), or
electronic tandem switching (ETS). These services are described in the
next few paragraphs. Each of them allows interlocation dialing on a 7digit basis, where the first three digits uniquely identify each location
and the last four digits identify that location's PBX or centrex stations.
The first three digits do not correspond to the station's normal telephone
number and are only used for private network calls. The result is that
the private network customer has a unique 7-digit dialing plan that is
uniform for all locations on the network.
All of these services-tie lines, tandem tie lines, CCSA, EPSCS, and
ETS-allow the private network the option of carrying calls that go off
the network, that is, calls that do not terminate at one of the customer's
PBX or centrex systems. To enter the public network to complete a call,
tie-line customers dial "9" plus a PSTN number. The CCSA, EPSCS, and
ETS networks recognize lO-digit calls as off-network calls (where the ten
digits are PSTN numbers). The CCSA service, EPSCS, and ETS carry the
call over the dedicated facilities of the private network to a point close to
the desired location where the call enters the PSTN. (Chapter 4 contains
more information about private network configurations and various callrouting arrangements.)
Common-Control Switching Arrangement
CCSA service was the first private network service to offer a customer
with geographically dispersed locations uniform dialing over dedicated
private facilities. The CCSA is primarily an interstate service regulated
by the FCC. Any station within a CCSA network may directly dial any
other station by using a uniform 7-digit dialing plan. The first three
digits identify the location, and the last four digits identify a location's
PBX stations or centrex stations .. The network switching systems that perform the routing function are selected by the Bell System or an independent telephone company, depending on the location, and are never on
the customer's premises; that is, CCSA routing switches cannot be PBXs.
Dedicated access lines from the PBX or centrex provide access to the
private network and to the selected network switches.
To use a CCSA private network, the customer dials an access digit at
the PBX or centrex and, after being connected to the Bell System CCSA
switch, dials the 7-digit on-network number or a lO-digit off-network
number.

Chap. 2

Services

69

Customers with extensive tie-line networks find that the costs of
adding CCSA switches are justified by the convenience of the uniform
dialing plan. Costs are still lower than they would be using the PSTN.

Enhanced Private Switched Communication Service
EPSCS is an improved CCSA-like service introduced in 1978. Like CCSA,
EPSCS is an interstate service regulated by the FCC and uses uniform
7-digit dialing. It, too, utilizes switching systems selected by the Bell System with dedicated access lines to customer PBXs and centrex switching
systems to accomplish network routing functions. However, EPSCS offers
features in addition to those available in CCSA both as part of the standard EPSCS offering and as options at extra cost.
Two unique standard features of EPSCS are 4-wire transmission (to
improve transmission quality)28 within the private network and a Customer Network Control Center, which customers can use to control some
network operations and to obtain private network usage and status information automatically and on demand. Other features include:
• automatic route selection of FX and WATS facilities for off-network
calling
• automatic alternate routing
• time-varying routing to accommodate expected changes in traffic loads
• call queuing when a network, FX, or WATS facility is busy
• authorization code entry when placing a call to provide controlled use
of expensive facilities
• special recorded announcements
• "meet me" conferencing with 6-station capabili ty 29
• automatic dialing.

Electronic Tandem Switching
ETS is another recently introduced (in 1979) private network service. ETS
is regulated by state commissions and is not in itself an interstate service.
It is a collection of features offered by the same switching equipment that
provides PBX and centrex service. There are no special Bell System
28 See Section 6.2.2.
29 "Meet-me" conferencing allows a maximum of six people to participate in a conference
call. The participants dial a special network number (called a conference dial code) at a
prearranged time. Only those people who dial the assigned code have access to the
conference.

70

Introduction to the Bell System

Part 1

switches. To obtain ETS, the customer must be served by PBXs and centrex switches that are capable of being equipped with ETS features. Once
equipped, these PBX and centrex switches offer the basic uniform dialing
plan for dedicated private facilities characteristic of private networks.
Many of the features available to EPSCS customers are also available
to ETS customers, including automatic route selection, call queuing, and
authorization codes. ETS also offers a Customer Administration and Control Center, which is similar to the Customer Network Control Center
used with EPSCS but less sophisticated. ETS offers the customer less
sophistication than EPSCS, but generally costs less.
2.5.4 DATA SERVICES
Section 2.2.3 discussed data products for use on customer premises and
described some data services derived from the use of DATAPHONE data
sets and their inherent capabilities.· In DATAPHONE II data communications service, for example, monitoring and control of a data network is
accomplished through the capabilities of DATAPHONE II data communications service equipment; the telecommunications network provides only
transport functions; that is, it serves only as a communications path or
channel.
This section describes data services that require specific network
implementations or functions, such as synchronization, multiplexing, and
switching. These network services have evolved from the differing needs
of users. DATAPHONE digital service, for example, responds to the need
for very high-quality data transmission. DATAPHONE Select-a-station
satisfies a special class of applications involving the interconnection of a
large number of low-speed data stations on analog private-line facilities.
The other services described in this section are emerging to meet rapidly
growing, diverse needs and are expected to be available in 1983 or 1984.
DATAPHONE Digital Service
DATAPHONE digital service offers data communications over point-to-

point or multipoint private lines at data rates of 2.4, 4.8, 9.6, and 56 kbps.
The objectives of this service are high performance and excellent availability. Availability is provided by a network with high reliability and
rapid restoration of service when failures do occur. It is attractive to customers such as on-line reservation services for airlines, who require low
error rates and high network availability. In general, error rates and network availability are substantially better with DATAPHONE digital service than with other private-line services or on the PSTN.
The service is provided by the Digital Data System (DDS), a synchronized data network (see Section 11.6.1). Data transmission is full

Chap. 2

Services

71

duplex (that is, 2-way simultaneous transmission) and remains digital end
to end (terminal to terminal). There are no restrictions on the data format; that is, the service is transparent to any data sequence. The DDS is
designed for data only-there is no provision for voice transmission.
In 1982, DATAPHONE digital service was available in about 100
metropolitan areas; it is expected to be available in over 125 areas in the
next several years.
DATAPHONE Select-A-Station
VATAPHONE Select-a-station service allows customers to establish a
series of point-to-point connections rapidly between a master location
and a large number of remote locations. It is suitable for users who need
remote telemetry from a large number of remote locations. A typical customer might be a central station alarm company that would use the service to provide security and fire protection by monitoring business and
residential premises. The service is provided over voiceband private lines
and is supported by high-speed switching equipment designed especially
for this application. The equipment is located in central offices but is
controlled by the customer's master station equipment. Section 11.6.3
provides more information on the operation of the system.

Basic Packet-Switching Service30
The Basic Packet-Switching Service (BPSS) is a private-line switching
arrangement for switching data packets among a customer's various locations. 31 BPSS is particularly suited for customers with the following
requirements:
• large numbers of data calls
• large quantities of data that must be transmitted between various locations
• data flow occurring in bursts, with a peak data rate that is high com,;.
pared to the average data rate. (The higher the peak-to-average ratio,
the more efficient the transport of data over BPSS.)
30 Basic Packet-Switching Service has been renamed ACCUNET Packet Service. (ACCUNET
is a service mark of AT&T Co.) Shortly before publication of this book, the names of
several services were changed. Time did not permit changing the names throughout the
book; however, the new names are indicated in footnotes the first time an applicable
service is mentioned.
31 With packet switching, the data are divided into packets, each of which includes
destination and control information (see Section 5.8.1) in addition to data. Section 11.6.2
discusses packet-switching systems.

72

Introduction to the Bell System

Part 1

An airline company, which might use BPSS for reservations and other
operations, is an example of this type of customer.
BPSS is furnished at a telephone company central office. It is accessed
through ports that operate at transmission speeds of 9.6 and 56 kbps.32
Each port is dedicated to a single customer, and a customer may combine
ports from more than one BPSS packet switch with access lines and
trunks to form private packet-switching networks. The design of BPSS
permits different customers to obtain ports on the same switching
arrangement and therefore share its common switching capability.
Although a switching arrangement may be shared by many customers, its
inherent design ensures privacy of communication between different customer networks. When the access lines to BPSS ports and the trunks
between ports on different switching arrangements are provided by the
Digital Data System, BPSS offers high availability and reliability.
BPSS provides two types of fundamental capabilities: virtual call and
permanent virtual circuit capabilities. Virtual call capability allows setup
and clearing on a per-call basis. Once a call is set up, it appears to have a
dedicated connection for its duration, that is, until cleared. Permanent virtual circuit capability provides the same functions as virtual call capability,
except that call-related procedures (setup and clearing) are eliminated.
Permanent virtual circuits are permanently defined in tables in the
switching arrangement(s) when service is established. Data terminals
connected by a permanent virtual circuit appear to have a full-time, dedicated connection. A customer must place an order with the telephone
company to establish, change, or discontinue a permanent virtual circuit.
Customers for the virtual call capability of BPSS may include retailers
who need access to several different data bases periodically during business hours. An automotive parts or plumbing supply outlet may have to
place orders with several different distributors. Customers for the permanent virtual circuit capability of BPSS may include retail stores whose
checkout clerks routinely obtain clearances for credit-card purchases by
customers.
The switching arrangement available with BPSS has a nominal switching capacity33 of up to 500 data packets per second. It supports maximum
packet sizes of either 128 or 256 data octets (8-bit characters) as specified
by the customer. Various software functions are available, with which,
for example, customers may create their own logical subnetwqrks. 34

32 BPSS supports the 1980 access protocol Comite Consultatif International Telegraphique et
Telephonique (CCITT) Recommendation X.25 (see Section 8.8).
33 Actual switching capacity depends on traffic characteristics (see Section 5.8).
34 The capabilities available in BPSS are described in detail in AT&T Long Lines 1982.

Chap. 2

Services

73

High-Speed Switched Digital Service35
High-Speed Switched Digital Service (HSSDS) provides a 1.544-megabitsper-second (Mbps)' network for transmission of voice, data, or video
within or between HSSDS switching nodes. Customers call an
800 number to reserve HSSDS facilities between nodes. Access between
customer premises and an HSSDS switching node is provided through
High-Capacity Terrestrial Digital Service. 36 The network, which is composed of terrestrial and satellite digital facilities, is planned to have nodes
in forty-two cities by the end of 1983. It will support 2-point, multipoint,
and broadcast connections.
High-Capacity Digital Transport Services
These services, first available in 1983, provide customers with 1.544-Mbps
digital circuits on a full-time (24 hours a day) basis. Two services are
included, one using terrestrial facilities, the other satellite facilities.
High-Capacity Terrestrial Digital Service (HCTDS) can be used to
connect two customer locations or to connect a customer location to a
telephone company central office. Equipment may be provided at the
central office that enables the digital circuit to carry twenty-four
voiceband channels, each of which can be terminated on the switching
system.
High-Capacity Satellite Digital Service (HCSDS)37 use dedicated
earth stations at the customer's location or shared earth stations at four
locations in the United States. Customers using shared earth stations
obtain dedicated terrestrial links via HCTDS to one of the four shared stations. Point-to-point and multipoint communications among shared and
dedicated earth stations are permitted. Features such as echo cancellation,
elastic stores,38 and earth station control are also available.
2.5.5 MOBILE TELEPHONE SERVICES
Mobile telephone services utilize radio transmission to provide telephone
service to customers on the move. Until recently, development of these
services has been constrained by limited radio-frequency assignments and

35 High-Speed Switched Digital Service has been renamed ACCUNET Reserved 1.5 Service.
36 High-Capacity Terrestrial Digital Service has been renamed ACCUNET T1.5 Service.
37 High-Capacity Satellite Digital Service has been renamed SKYNET 1.5 Service. (SKYNET
is a service mark of AT&T Co.)
38 A buffer memory that can hold a variable amount of data. The length of time that
specific data items remain in the store depends on the amount of data it contains.

74

Introduction to the Bell System

Part 1

technological complications. Mobile telephone services include: land
mobile telephone service, BELLBOy39 personal signaling set paging service, air I ground service, marine radiotelephone services, and highspeed train telephone service.
Land Mobile Telephone Service
Land mobile telephone service provides 2-way voice communications,
through designated central offices, between mobile units and land telephones or between two mobile units. Users in the mobile serving area
have full access to the PSTN either on a manual (operator-handled) or a
direct-dial basis. In 1981, the FCC approved a new type of system, called
cellular. Beginning in 1983, the Bell companies will serve a much larger
number of customers through this new design, which makes more
efficient use of frequency assignments. Section 11.4.1 discusses this system and other land mobile telephone systems, present and future.
BELLBOY Personal Signaling Set Paging Service

The BELLBOY personal paging service notifies customers when someone
wants to talk with them. Customers carry a cigarette-pack-sized radio
receiver that emits an audible tone when the number assigned to that
unit is called. (A new receiver being offered by some telephone companies also incorporates a visual display of the calling number.) The
receiver is activated by an array of radio transmitters that provide coverage for an urban area. In 1980, BELLBOY personal signaling set paging
service had about one hundred thousand customers nationwide (see Section 11.4.2).
Air I Ground Service
Air I ground service provides 2-way telephone service between customers
flying in private aircraft and customers on the PSTN. The service uses
radio base stations connected to control terminals and mobile service
switchboards that interconnect with the PSTN. All radio equipment
mounted on aircraft is customer owned and customer maintained. In
1980, approximately two thousand aircraft were equipped with this service, and sixty thousand calls were placed during the year.
Marine Radiotelephone Services
Marine radio telephone services include very high frequency (VHF) maritime service, coastal-harbor service, and high-seas maritime radiotelephone service. These services provide 2-way telephone service to water

39 Registered service mark of AT&T Co.

Chap. 2

Services

75

craft. The three services differ in the range of distances over which they
operate. VHF maritime service offers reliable communications up to
50 miles offshore and on inland waterways. Coastal-harbor service communications can range up to 200 miles offshore, and high-seas service is
intended for ships engaged in oceanic operations and transoceanic passages. The radio equipment for all three services mounted on board ships
is customer owned and maintained.
High-Speed Train Telephone Service
High-speed train telephone service provides telephone service between a
passenger train and the PSTN. Operator-handled train telephone service
was inaugurated in 1947, and by 1952, service was provided to nineteen
trains on five railroads. These installations are now out of service, in
most cases because of the demise of the equipped trains. More recently,
in 1968, train telephone service was installed aboard the Metroliner trains
operating between New York and Washington, D.C. The service provides
the public with coinless TOUCH-TONE telephones. Approximately fortyfive thousand calls were handled yearly during the 1970s. This service is
being phased out in anticipation of the new cellular service.
2.5.6 VIDEO TELECONFERENCING SERVICE
In 1964, AT&T offered PICTUREPHONE4o visual telephone service using
small desktop units that contained both a special camera tube and a small
black and white receiver. Public booths were also established, and service was offered between locations in Chicago, New York, and
Washington, D.C. The offering was continued until 1975 as a market
trial.
The experience gained in the trial led to a reorientation of the service.
In 1975, PICTUREPHONE meeting service was introduced (also on a trial
basis), using rooms equipped for conferences of various sizes. The trial of
the reoriented service continued until 1981, when AT&T announced its
plans to offer video teleconferencing as a standard service in forty-two
cities. Long-distance transmission for the standard service is provided by
the High-Speed Switched Digital Service (see Section 2.5.4). Access from
conference rooms to the nodes is offered under full-period High-Capacity
Terrestrial Digital Service (see Section 2.5.4) tariffs. Customers may have
private conference rooms on their own premises or may use public rooms
offered by AT&T.
The objective of the video teleconferencing service is to make conferences both effective and pleasant. As shown in Figure 2-11, six people

40 Registered service mark of AT&T Co.

Figure 2-11. Conference room equipped for
PICTUREPHONE meeting service.

are accommodated at a conference table . Each person is within range of
one of three cameras. The picture selected for transmission depends on
which person is speaking; thus the speaker is on camera. Other cameras
are provided for overviews, various graphic displays, and for a speaker at
an easel. Monitors show both the incoming and outgoing video.
Video teleconferencing service will prove to be economical when
measured against the rising cost of travel and the time lost in travel that
could be applied to other business or personal responsibilities.

2.6 PUBLIC COMMUNICATIONS SERVICES
By the end of 1981, there were approximately 1.6 million telephones providing Public Communications Services and generating about $3 billion
in annual revenues. Most public telephones are installed at locations
where a public need exists such as airports, bus depots, train stations,
hotel lobbies, large office buildings, and on public streets and highways.
They provide the general public with access to all United States and
international telephones, generally on a customer-dialed basis .
About 30 percent of the telephones that provide Public Communications Services are called semipublic because they are not always available
to the public . Semipublic telephones are most often fo und in service stations, delicatessens, self-service laundries, and similar businesses. The

76

Chap. 2

Services

77

proprietor can use the line for business purposes and make a coin telephone available to the public during business hours. Many businesses
also install semipublic telephones to control outgoing calls by employees.
The semipublic station can also be equipped with a noncoin, answer-only
extension to allow the proprietor to answer incoming calls without having to pick up the coin telephone. While most of these have a nondial
telephone bridged across the line without any privacy protection for the
user of either station set, arrangements can be made to provide privacy to
users of the coin telephone. A business with a semipublic telephone can
be listed in both the white pages and the Yellow Pages of telephone
directories.
Revenues from public telephones are generally shared with the owners of the premises where the sets are located. For telephones located on
public property, most state regulatory agencies have set up some formula
to provide a commission to the political unit that grants the franchise.
For example, urban sidewalk installations usually yield a commission to
the city. Similarly, commissions on airport locations are paid to the
government agency operating the airport.
Charging arrangements for semipublic telephones vary considerably
from state to state. Some states require a minimum amount of revenue to
be generated. If that amount is not achieved, the proprietor is billed for
the amount needed to reach the minimum. Other states require a flat
monthly fee with no guarantee. Rarely are any commissions paid on
semipublic telephone service.
There are two types of public telephones-coin and Charge-a-Call, or
coinless (see Figure 2-12). With a coin telephone, a customer may place
either sent-paid calls (paid for at the time the call is mad~ by depositing
one or more coins) or non-sent-paid calls (where payment is not made at
the time of the call). The latter include collect calls, calls charged to a
third number, and calling card calls (that is, calls billed to another telephone). Charge-a-Call stations can complete only non-sent-paid calls.
Coin telephones can provide either coin-first or dial-tone-first service.
Coin-first telephones require a specific deposit (ten, fifteen, twenty, or
twenty-five cents) before the receipt of dial tone. Dial-tone-first telephones allow the completion of service calls (for example, directory assistance, 911, and repair service) and access to Traffic Service Position System (TSPS) operators or TSPS/ Automated Coin Toll Service (ACTS)
equipment without coin deposits. In addition to providing coin-free
emergency access, the dial-tone-first station gives the customer some
assurance that it is working before any coins are deposited.
For toll calls, a toll-and-assistance operator informs the caller of the
charges and confirms the correct deposit of coins. Most toll service
operators are supported by TSPS (see Section 1004.1). Where TSPS has
been enhanced by ACTS equipment, the operator function is automated,
and synthesized voice messages are used for interaction with the caller.

Figure 2-12. Public telephones. Left, a coin set; right, a Charge·a·Call set.

About two-thirds of the public telephone revenues are from non-sentpaid calls. At locations such as airports and railroad terminals, a substantial percentage of the calls are non-sent-paid. These locations are attractive candidates for the installation of Charge-a-Call telephones, which
can handle such calls, and for the provision of Automated Calling Card
Service (see Section 2.5.1), which automatically bills them.

2.7 CUSTOMER SUPPORT SERVICES
The preceding sections describe many of the telecommunications services
available to Bell System customers. These services are provided through
various types of terminal equipment and the capabilities of the network.
This section discusses a different class of services-those provided by an
operating company to support customer needs in the areas of acquisition,
use, and maintenance of telecommunications services.

78

Chap. 2

Services

79

2.7.1 RETAIL SALES AND SERVICE
In 1980, over 20 million requests for telephone service (for example, to
initiate, terminate, or change service), representing more than half of the
Bell System activity, were handled through the 1700 Bell PhoneCenters41
then maintained by Bell operating companies. Before the introduction of
Bell PhoneCenters, residential customers had to contact a telephone company business office to order service and then wait for an installer to
bring the telephone set(s) to the home.
Until recently, telephones and other terminal equipment were leased
from the local operating company. On January 1, 1983, under provisions
of the FCC's Computer Inquiry II decision, sale of new terminal equi pment at Bell PhoneCenters was transferred to AT&T. 42 Operating companies may sell terminal equipment that was in their inventory on that
date, and they continue to operate service centers where customers may
conveniently order service and replace faulty equipment bought or leased
from the operating company.
Customers can select from a variety of basic and DESIGN LINE decorator telephones. They can also obtain the lengths of handset and mounting cords they desire and the particular adapters they need to make their
inside wiring compatible with modular telephone technology (see Section
2.2.2). Once the wiring inside a residence has been adapted, station sets
may be plugged in or removed quite conveniently.

2.7.2 BUSINESS OFFICE SERVICES
Where telephone company customer service centers exist, service
representatives in business offices refer requests for new residential service to them. Business offices continue to initiate service orders for new
phone installations in some cases, for example, when inside wiring
records indicate that customers will be unable to install the phones they
desire. In addition, they answer customers' questions about billing for
residential services.

2.7.3 INSTALLATION AND MAINTENANCE SERVICES
Installation and maintenance services include the inside wiring and connection of station equipment. This may be required by business customers and by those residence customers who either cannot or do not wish to

41 Also called PhoneCenter Stores.
42 Terminal equipment from several manufacturers may be purchased at many retail stores.

80

Introduction to the Bell System

Part 1

install their own equipment. The installer may perform the inside wiring job for new installations and make changes to existing wiring, as
required.
Service problems are handled by a call to Repair Service. The nature
of the customer's complaint is recorded, and after the probable cause is
determined by testing, repair action is initiated. Section 13.2.3 further
discusses maintenance operations.
2.7.4 DIRECTORY SERVICES

Customers are entitled to have their names, addresses, and telephone
numbers listed in white pages directories, but listings will be withheld if
a subscriber so desires. Subscribers not listed in the white pages may also
specify that their addresses and telephone numbers not be published in
Information Services directories for public disclosure. Arrangements may
be made to print listings in bold-face type or in association with special
instructions, such as, "After 5 o'clock, call 555-5555."
An additional fee is required for some white pages services. All Yellow Pages listings require additional fees. Costs depend on the details of
the special listing, in particular, the size of the listing and whether or not
it is accompanied by an advertisement.
To make Directory Services more helpful to users, innovations are
being introduced. For example, in some white pages directories, federal,
state, and county governmental listings are printed on blue paper in a
special section. Upon request, partial addresses or no addresses may be
listed with names and telephone numbers. Four-color advertisements are
being introduced in Yellow Pages directories. New Yellow Pages books
specialize in user-market needs such as Medical Directories, Tourist Directories, etc. White and Yellow Pages data bases have been developed for
user access via the emerging electronic information technology.

AUTHORS
C. E. Betta
H. J. Bouma
P. S. Browning
W. R. Byrne

R. Carlsen

D. C. Franke
W. G. Heffron

T. B. Morawski
K. J. Pfeffer
P. T. Porter

S. H. Richman

J. F. Ritchey
A. T. Vitenas
C. H. Zima

3
Introduction to the Network

3.1 WHAT IS A TELECOMMUNICATIONS NETWORK?
As a starting point in defining a telecommunications network, a general
definition of a network may be helpful. In a broad sense, a network is a
system of interconnected elements. Topologically, it can be representee
by a set of nodes and a set of links that interconnect pairs of nodes. i.
network is needed when certain types of services must be provid p • to
many, widely dispersed customers. Depending on the types of ~ .. vices,
the characteristics of the network elements may differ greatly.
Another concept necessary for the definition of a telecom' .unications
network is the notion of telecommunications traffic or, (' . .nply, traffic.
Traffic is the flow of information or messages throu". the network.
Traffic may be generated by simple telephone convers, ,iOns, or it may be
the result of complex data, video, and audio servic _so (Chapter 5 deals
with traffic theory and its application to engineerir (, the network.)
A telecommunications network, then, is a syster _of interconnected facilities designed to carry the traffic that results irom a variety of telecommunications services. (Chapter 2 discusse~ the various telecommunications services available.) The telecommur..ications network as a whole has
two different but interrelated aspe('·~. In terms of its physical co~­
ponents, it is a facilities netwo"~.. In terms of the variety of telecommunications services that it :- ~vvides, it is a set of many traffic networks,
each representing a t-'articular interconnection of facilities. The distinction between traffic networks and the facilities network is discussed in
more detail later in this chapter.
As stated earlier, a network can be represented by nodes and links. In
the telecommunications network, the nodes represent switching offices
and facility junction points, and the links represent transmission facilities. Traffic is the flow of information in the network.

81

82

Introduction to the Bell System

Part 1

Telecommunications has three characteristics that dictate the basic
nature of the network. First, traffic must be carried among customers
dispersed over large geographic areas. Second, traffic may be generated
between any pair of customers at virtually any time, although the duration of each call may be fairly short. Third, the ability to exchange information between any pair of customers is expected to be available with
relatively short delay.
Figure 3-1 illustrates some key concepts in the design of a telecommunications network. 1 Figure 3-1A shows a highly oversimplified situation in which no switching is used and telephones at all four end points
(customer locations) are directly interconnected in pairs by transmission
paths. A telecommunications network designed in this way would be
inefficient and prohibitively expensive because it would require many
telephones at each end point and many transmission paths as well.
The design depicted in Figure 3-1A can be improved considerably by
the introduction of switching. For example, the use of a switch at each
location would eliminate the need for all but one telephone at each location. This situation is shown in Figure 3-1B. In this case, although the
total number of telephones has decreased, the number of required
transmission paths remains the same, and the implementation of switching at each end point would be expensive.
Figure 3-1C shows that a much more efficient use of network elements
results from the introduction of switching at a central location to interconnect transmission paths emanating from the end points. Both the
number of switches and the number of transmission paths are substantially reduced. In a network with no switching (as in Figure 3-1A), if
there are n end points (n=4 in Figure 3-1A), n(n-1) telephones and
n(n-1)/2 transmission paths are needed. However, the network design
shown in Figure 3-1C requires only n telephones, n transmission paths,
and one central switch.
More efficient network design and lower cost to the customer are
achieved at a certain price: The number of simultaneous connections
through the network is limited. Thus, while the network depicted in
Figure 3-1A would allow the simultaneous connection of all pairs of end
points, thereby supporting six calls at the same time, the centrally
switched network configuration of Figure 3-1C can support only two
simultaneous connections. This limited potential for simultaneous calls is
not a serious drawback, however, because the concurrent use of the network by all or even most users is unlikely. Chapter 5 includes a further
discussion of topics related to network usage.

1 For an in-depth analysis of the design and cost characteristics of telecommunications
networks, see Skoog 1980.

B

c
Figure 3-1. Networking and the tradeoff between transmission and
switching. A, direct interconnection of telephones; B, interconnection through
switches at end points; C, interconnection through a centralized switch.

84

Introduction to the Bell System

Part 1

3.2 THE FACILITIES NETWORK
The telecommunications network was defined in Section 3.1 as a system
of interconnected facilities designed to carry traffic that results from a
variety of telecommunications services. When viewed from the perspective of its physical components, or facilities, the network may be referred
to as the facilities network.
The components of the facilities network may be divided into three
broad. categories.
• Station equipment is generally located on the customer's premises. 2
Its primary functions are to transmit and receive the information flow
and required control signals between customers and the network.
• Transmission facilities provide the communications paths that "carry"
the information between customers. In general, transmission facilities
consist of some sort of transmission medium (for example, the atmosphere, paired cable,3 coaxial cable, lightguide cable)4 and various types
of electronic equipment located at different points along the transmission medium. This equipment amplifies and, sometimes, regenerates
the transmitted signals. In addition, various types of facility terminal
equipment provide functions needed where transmission facilities
connect to switching systems and at facility junction points. 5
• Switching systems interconnect the transmission facilities at various
key locations and route traffic through the network. As mentioned in
Section 3.1, the introduction of central switching into the network
yields cost savings in station equipment and transmission facilities.
In addition to the functions just described, transmission facilities and
switching systems provide for signaling in the network. (Chapter 8
describes signaling, a major network function.)
The Bell System provides a large percentage of the telecommunications facilities in the United States nationwide network. However,
numerous independent telephone companies and other common carriers
also own both transmission facilities and switching systems.
The following sections give an overview of the three basic categories
of facilities, which are discussed in detail in subsequent chapters.
2 Customer-premises equipment (CPE), a broader term, includes more than station or terminal
equipment. For example, private branch exchange equipment located on a customer's
premises performs customer switching functions.
3 Paired or multipair cable contains a number of twisted pairs of wires.
4 Section 6.3 describes these transmission media.
S Chapters 6 and 9, respectively, describe some of the functions and equipment.

Chap. 3

Introduction to the Network

85

3.2.1 STATION EQUIPMENT
Station equipment is the user's interface with the rest of the network and
the available services. The most common station equipment is the ordinary single-line telephone set. The functional components of a telephone
set include a transmitter and a receiver (most often combined in a
handset), a rotary or pushbutton dial, a switchhook mechanism, and a
bell or other alerting device. The telephone converts an acoustic signal
(which may be generated by a human voice or by a device that
transforms data into a series of tones) into an electrical signal, which it
sends over a transmission facility. For reception, the telephone converts
an incoming electrical signal back into an acoustic signal.
In addition to transmitting and receiving information in the form of
electrical signals, the telephone also provides for two kinds of signaling
functions: supervision and addressing. Supervision includes the constant
monitoring by the local switching system or by a private branch
exchange (PBX) of the status (idle or busy) of the telephone and alerting
the user that a call is being made by providing an audio (or visual) signal. Addressing refers to the task of specifying to the network the destination of a call.
The telephone switchhook is used to signal idle or busy status. When
the telephone is idle, or "on-hook," the switchhook contacts are open.
When the telephone is busy, or "off-hook," the switchhook contacts are
closed. These supervisory signals allow certain equipment at the central
office (or PBX) to recognize origination, answer, and termination of a call.
A customer is alerted to an incoming call by a bell or tone ringer (or. ir
some cases, a light).
Addressing is done by eith o - - . _ .....~ y ulal or a set of pushbuttons, producing a signal tJ.._:.~ corresponds to the called number. With a rotary
dial, a serie~ of pulses, equivalent to alternate on-hook, off-hook conditions, represents each digit dialed. With a pushbutton set, a different pair
of tones represents each digit. 6
Many other kinds of station equipment convey various types of information. Some enable computers to communicate directly with one
another over the telecommunications network. Others are used for visual
information such as video, facsimile, and graphics. (Section 11.1 discusses
station equipment in detail.)

3.2.2 TRANSMISSION FACILITIES
Transmission facilities provide the communications paths that carry the
traffic between any two points in the network. These communications
6 There are also telephone sets with push buttons that produce dial pulses.

86

Introduction to the Bell System

Part 1

paths are referred to as channels or circuitl and may be classified as follows into three broad categories.
• Ordinary channels (circuits) that connect customers' station equipment
to a switching system are called lines or loops? Loop is derived
historically from the pair of wires that form a loop between the
customer's location and the switching system. Originally, all loops
were wire pairs. Today, however, many loops do not have an associated physical pair of wires for the entire path. Instead, loop carrier
systems8 are used.
• Ordinary channels (circuits) that connect two switching systems are
called trunks. There may be several switching nodes with trunks connecting them between the calling customer's telephone and the called
customer's telephone. The trunks and switching systems carry traffic
generated by many customers, while loops are dedicated to individual
customers. 9
• Finally, channels (circuits) dedicated to a specific customer to provide
special services are called special-services circuits. They encompass both
circuits to a customer's equipment and circuits between network
switching systems.
Based on the above classification of channels, it is possible to divide
the transmission facilities into two general categories.
• Loop transmission systems, subscriber loop systems, or, simply, loop
systems carry loops and special-services circuits to a customer's premises. Typically, loop transmission systems are paired cable (also called
multipair cable) that is suspended from telephone poles, buried directly
in the ground, or placed in underground cable ducts (often called conduit). The average length of customer loops is about two miles.
• Interoffice transmission facilities carry trunks and special-services circuits between network switching systems. These facilities vary
greatly, but the majority are carrier systems. Interoffice transmission
facilities range in length from less than a mile to several thousand
miles.

7 For the most part, the terms channel and circuit are used interchangeably, as are the terms
loop and line. Where differences are perceived, it is likely to be the result of traditional
usage in a particular technology area.
8 Carrier- systems combine a number of circuits on two physical pairs of wires using a
technique called multiplexing (discussed in Section 6.5).
9 Except in the case of party-line service, where loops are shared by two or more customers.

Chap. 3

Introduction to the Network

87

3.2.3 SWITCHING SYSTEMS
This section provides a broad characterization of switching systems as elements of the facilities network. (Chapter 7 discusses switching functions
and concepts, and Chapter 10 describes a number of systems.)
The primary function of switching systems is to interconnect circuits.
Depending on the types of circuits involved, switching systems fall into
two functional categories: local and tandem. Local switching systems
connect customer loops directly to other customer loops or customer
loops to trunks. Local switching systems,· which may serve many
thousands of customer loops, are also referred to as central offices. The
central office building contains one or more switching systems (central
offices),lo certain transmission and signaling equipment, and other equipment necessary to provide telephone service to customers in the nearby
geographic area. A central office may be divided into two or more 3-digit
central office codes. ll The last four digits of a telephone number provide
up to 10,000 line numbers within each central office code.
The term tandem is used generically for any switching system that
connects trunks to trunks. In a more limited sense, it is often used to
denote systems typically found in metropolitan networks within the public switched telephone network (PSTN). These local tandem switching systems connect local switching systems to each other or to other systems in
the PSTN or interconnect other metropolitan tandem systems. Generic
tandem (that is, trunk-to-trunk) systems that perform class 1 through class
4 functions in the toll switching hierarchy (described in Section 4.2) are
called toll systems. The terms local and toll reflect the tariff distinction
between local and toll traffic.
In addition, some switching systems perform both local and tandem
switching functions. (ChaDtpT A. U.l::;\';U::;::;~::; {ne ImpllcatlOns ot comtnnea
local/tandem switchIng systems.)
A recent development in local switching is the use of remote switching systems that serve small population centers typical of rural areas.
Customer loops are connected to these systems, which, in turn, are connected to central offices in larger population centers. The remote systems
are essentially extensions of the host central office. (Section 10.3
discusses these systems.)
In addition to network switching systems located in telephone company buildings, another type of switching system, the PBX, is typically
located on a customer's premises. As described in Section 2.3, a PBX connects a localized community of users to each other and, through special10 The term central office is sometimes applied loosely to a central office building and its
equipment.
11

Also called exchanges.

88

Introduction to the Bell System

Part 1

services circuits, to a network switching system. An attendant may handle calls involving an off-premises party. The circuits connecting a PBX
to a central office are called PBX trunks since they interconnect two
switching systems. In most cases, the circuits appear as lines to the central office. PBXs and associated PBX trunks are part of the total facilities
network.
Finally, operator services (such as directory assistance, special charging
on toll calls, and intercept of calls to nonworking numbers) are provided
by arrangements of equipment that are considered to be switching systems. (Section 4.2.2 describes how operator services are provided within
the network, and Section IDA describes the related equipment.)

3.3 TRAFFIC NETWORKS
The description of the telecommunications network in the previous section emphasized the physical components of the network, namely, station
equipment, transmission facilities, and switching systems. In that context, the network was referred to as the facilities network. It is also
important to consider the manner in which this network provides the
various telecommunications services. From this perspective, the network
may be thought of as a set of traffic networks sharing common facilities.
For example, the PSTN, which provides public switched telephone network services, is the largest and best-known traffic network. Many other
traffic networks provide a variety of special services such as private-line
voice and data and audio and video program services. Each traffic network is designed to meet a particular set of requirements related to
transmission performance, reliability, maintenance, and the ability to
handle the expected traffic volume. The following sections describe some
of these traffic networks and show how they use common elements and,
in some cases, share them.

3.3.1 PUBLIC SWITCHED TELEPHONE NETWORK
Because of the large volume of business and residential telephone traffic
that it carries, the PSTN is probably the most familiar of the traffic networks. This network provides the public switched telephone network
services described previously in Section 2.5.1.
The various types of traffic in the PSTN represent communications
between any two end points in the network. Traffic is switched through
each switching office, or node, it encounters and travels between nodes
on trunk groups. The offices and trunk groups are arranged in a
hierarchical routing structure, as described in Section 4.2. Circuits 1 and
2 in Figure 3-2 carry this type of traffic. The foreign exchange (FX) line
shown as circuit 5 is noteworthy because it appears as a line at distant

CD
CD

CD
CD

m
t@

TRANSMISSION FACILITIES

CIRCUITS

o

CUSTOMER LOOP
PBX-CENTRAL OFFICE LINE

EJ
L:=J

SYMBOLS

LOOP TRANSMISSION FACILITIES

RESIDENCE

INTEROFFICE TRANSMISSION
FACILITIES

D~~I

PBX TIE TRUNK
PRIVATE LINE

{]

BUSINESS LOCATION

ATTENDANT'S CONSOLE

FXLINE

CD (i)
&

INTEROFFICE TRUNKS

SWITCHING OFFICE B

SWITCHING OFFICE A

LINES

LINES

=

~~ :--===-- ~~=~~";:;t=.:j:J

Figure 3-2. The use of transmission facilities by various types of services.

switching office B. It is generally used primarily for traffic to and from
other lines in the local calling area of office B. It can also use office B as
the access point for the entire PSTN.
The PSTN is, by far, the largest traffic network in terms of both equipment utilization and traffic volume. In 1981, it handled about 270 billion
calls, and at the end of that year, it served about 180 million telephones
in the United States. (These figures include both Bell System and
non-Bell System calls and telephones.)

3.3.2 PRIVATE-LINE VOICE NETWORKS
As described in Section 2.5.3, large businesses with many dispersed locations often use private-line voice networks-dedicated facilities that connect a company's various locations. The circuits used for private networks are special-services circuits.
89

Introduction to the Bell System

90

Part 1

Some of these circuits are switched; others are nonswitched. Among
nonswitched private networks, many serve only two stations and are
called point-to-point networks. Others, called multipoint networks, interconnect a number of stations at dispersed locations and may have signaling
arrangements to alert appropriate stations when communication is
desired. Private-line nonswitched networks may be interconnected at
telephone company switching offices, although they are not switched
through the switching systems. The PBX tie trunk, circuit 3, and the
private line, circuit 4, in Figure 3-2 are examples.
In addition to the nonswitched private networks, several thousand
private switched voice networks serve government organizations and
large business customers. The trunks and access lines in these networks
are private-line circuits that interconnect switching systems either at customer locations (for example, PBXs) or in telephone company switching
offices. A typical private switched network sharing switching systems
with the PSTN is shown in Figure 3-3. The switching systems are partitioned so that only private network access lines may access the private
network trunks. Private switched networks range in size from large
nationwide networks that serve hundreds of locations to small networks
in metropolitan areas that serve fewer than ten locations.

CIRCUITS

TRANSMISSION FACILITIES

CD
CD
CD

CJ

o

LJ

CUSTOMER LOOP
PRIVATE NETWORK ACCESS LINE

SYMBOLS

LOOP TRANSMISSION FACILITIES

~:6~g;:~CE

{]

ATTENDANT'S CONSOLE

TRANSMISSION

PRIVATE SWITCHED NETWORK TRUNKS
PUBLIC SWITCHED TELEPHONE
NETWORK TRUNKS
SWITCHING OFFICE D

SWITCHING OFFICE C
LINES

SWITCHING
NETWORK

TRUNKS

*

TRUNKS

SWITCHING
NETWORK

* DOTTED LINE REPRESENTS PARTITIONING OF THE SWITCHING OFFICES.
Figure 3-3. Example of a private switched network.

*

LINES

Chap. 3

Introduction to the Network

91

Examples of private switched networks are CORNET (the corporate
network used in the Bell System), FTS (the Federal Telecommunications
Service network serving the civil organization of the Federal Government), and AUTOVON (the automatic voice network of the United States
military).
A key point illustrated by Figures 3-2 and 3-3 is that private networks,
both switched and nonswitched, may share many elements of the facilities network with the PSTN. Specifically, the various circuits for the
PSTN and private networks are provided by common transmission facilities. Furthermore, in switched private networks, even the switching systems can be shared, as shown in Figure 3-3; although certain private
switched networks (for example, AUTOVON) have dedicated switching
systems.

3.3.3 PRIVATE-LINE DATA NETWORKS
Section 2.2 describes private-line data services over analog facilities. Section 2.5.4 describes DATAPHONE digital service, and Section 11.6
describes the system aspects of the Digital Data System (DDS) that support the service.
Another example of private-line data networks involves telegraph
channels. These channels employ a fraction of the voice bandwidth 12 and
are used for services such as teletypewriter, remote metering, and burglar
alarms. There are about 8000 interstate private teletypewriter networks.
About half are simple 2-point networks; the rest are multipoint networks,
most of them nonswitched.
In addition to private data networks provided by the telephone companies, customers may form their own switched data networks by leasing
private-line circuits and interconnecting them with computers operating
as data switches. Both the National Aeronautics and Space Administration and the Advanced Research Projects Agency have such private data
networks. In a certain sense, the structure of private data networks is
generically similar to that of private voice networks. The channels
involved are dedicated to the customer, are usually not switched by network switching systems, and typically share transmission facilities with
the PSTN.
Private voice and data channels differ primarily in their transmission
requirements, particularly where high-speed data is involved. Furthermore, because of the current advances in both switching and transmission
technologies, the distinction between voice and data networks is becoming less and less significant.

12 The voice bandwidth is the range of frequencies necessary to transmit and receive
acceptable quality speech signals. See Section 6.1.

92

Introduction to the Bell System

Part 1

3.3.4 PROGRAM NETWORKS
Radio and television broadcasters use program networks extensively to
distribute program material simultaneously to a number of their affiliated
stations. Because of different transmission requirements (see Chapter 6),
both audio and video program networks exist. A key difference between
program networks and private switched voice networks is the directional
nature of the communications. While voice networks support simultaneous 2-way communications, the program networks provide I-way
transmission of audio and video programs from one source to many destinations. Another difference concerns the typical transmission capacities
required. In a private voice network, two customer locations may be connected by a single voiceband channel; the transmission of audio and
video programs, on the other hand, requires much larger bandwidths for
the video information, typically, equivalent to hundreds of voice circuits.
Except for these two distinguishing features, the structure of program
networks is similar to that of private voice networks; hence, program networks share the transmission facilities with other private networks and
the PSTN.

3.4 A TYPICAL TELEPHONE CALL
To introduce the rudimentary operation of the network, this section
presents a functional description of a typical telephone call, the most
familiar service provided by the PSTN. The description illustrates some
of the terms defined in previous sections and introduces some new terms
and concepts.

3.4.1 SETTING THE STAGE (FIGURE 3-4)
Mrs. Cooper, a local realtor, is calling Mrs. Mahon, a prospective buyer, at
her home in a neighboring town. Mrs. Cooper's telephone is served by
central office A, and her central office code is 747. Mrs. Mahon's telephone is served by central office code 951 in central office B. Since many
calls are placed between central offices A and B, a number of trunks provide a direct route between the two offices. An alternate route through
tandem office C is also available. (Chapter 5 describes traffic engineering
aspects, including alternate routing.)

3.4.2 INITIATING THE CALL (FIGURE 3-5)
When Mrs. Cooper picks up her handset, the switchhook contacts of the
telephone set close, signaling its off-hook status. Control equipment in
the switching system at office A detects a change from on-hook to offhook status and interprets the change as a request for service. At this
time, dial tone is connected to Mrs. Cooper's telephone, assuming that a

TANDEM
OFFICE C

DIRECT ROUTE

CENTRAL
OFFICE

CENTRAL
OFFICE
B

TRUNKS

A

I

LINE
(LOOP)

I

I

I

~

~

MRS. MAHON'S
TELEPHONE

MRS. COOPER'S
TELEPHONE

(951-1234)

(747-4321)

Figure 3-4_ Direct and alternate routes for a
call from Mrs. Cooper to Mrs. Mahon.

CENTRAL OFFICE A

MRS. COOPER
HANDSET LIFTED;
SWITCHHOOK
CONTACTS CLOSE

......

REQUEST FOR
SERVICE DETECTED

~
AVAILABILITY OF ORIGINATING
REGISTER DETERMINED;
DIAL TONE CONNECTED

DIALING BEGINS

.....

~
FIRST DIGIT
RECOGNIZED;
DIAL TONE DISCONNECTED

~
DIGITS STORED
IN ORIGINATING
REGISTER

Figure 3-5. Initiating the call.

94

Introduction to the Bell System

Part 1

register, usually called an originating register,13 is available to accept and
store the digits she will dial.
After Mrs. Cooper dials the first digit, the dial tone is disconnected.
The digits dialed by Mrs. Cooper (951-1234) are received and stored in
the originating register.
3.4.3 CALL PROCESSING AT THE ORIGINATING CENTRAL OFFICE
(FIGURE 3-6)
Next, the control equipment in central office A translates the dialed
number. By examining the leading digits, usually the first three,14 it
determines that Mrs. Cooper's call is to another central office code; that is,
it is not an intraoffice call. Her call is an interoffice call and must be connected to a trunk going to another office. Routing information stored in
the system indicates which paths (trunk groups) are appropriate and
translates the desired paths to representations of physical locations or terminations of trunks. If the call is billable, an automatic message accounting (AMA) register is requested (see Section 3.4.5). At this time, control
equipment transfers the call information to a register in another storage
area (the outpulsing register shown in Figure 3-7), releasing the originating register from the call. The control equipment begins scanning the
outgoing trunks to find an idle trunk to office B. An idle trunk is found
directly between offices A and B.
The control equipment could have found that all trunks in the trunk
group(s) to office B were busy. In this case, it would have begun to scan
the outgoing trunks to tandem switching office C, since the call could be
routed on a trunk from office A to office C and from there to office B (as
shown in Figure 3-4). If all trunks to tandem office C had also been busy,
it would have been impossible to complete the call. In that case, Mrs.
Cooper would have heard a reorder tone, often called a fast busy tone
since it has 120 interruptions per minute (ipm), compared to the 60 ipm
of the busy tone.
3.4.4 CALL ADVANCEMENT TO THE TERMINATING CENTRAL
OFFICE (FIGURE 3-7)
The first event shown in Figure 3-7 is the selzlng of an idle trunk to
office B. When a trunk is seized for a particular call, it appears busy to
the switching system and becomes unavailable for other calls. A 2-way
13 To illustrate the functional operations involved in the call, this discussion uses generic
terms for equipment. Because of the variety of switching systems in the network, the
generic term may not fit all cases. For example, step-by-step switching systems, the
oldest systems used (see Section 10.2.2), may not have originating registers and may
complete the switching functions differently. Likewise, stored-program control systems
(see Sections 10.3.1 and 11.3.1) are computer-like in their operation.
14 In some cases, an access code, such as the digit 1, is used as a prefix to the address digits
on calls outside the local area. (See Section 4.3.)

CENTRAL OFFICE A

TRANSLA TlON AREA
CONSULTED

~
LEADING DIGITS
RECOGNIZED AS
CALL TO OFFICE B

+
ROUTING INFORMATION
DETERMINED

+
IF CALL IS BILLABLE,
AMA REGISTER REQUESTED

~
ORIGINATING REGISTER
RELEASED

+
OUTGOING TRUNKS
TO OFFICE B SCANNED

+
IDLE TRUNK FOUND

Figure 3-6. Processing the call at the
originating central office.

trunk may be seized by the switching system at either end to originate a
call, while a I-way trunk may only be seized from one end. Transmission occurs in both directions on either type of trunk.
Mrs. Cooper's line is connected to the outgoing trunk through a path
in the switching network within the switching system. The identity of
this trunk, the number of digits to be transmitted, and additional information that may be necessary for call setup are recorded in an outpulsing
register.
In central office B, an incoming register of the switch will be seized 1S
and will signal readiness to receive address information. The control
15 Seizing the register makes it unavailable for other incoming calls until it is released.

95

CENTRAL OFFICE A

CENTRAL OFFICE B

....

IDLE TRUNK TO
OFFICE B SEIZED

•

INCOMING REGISTER
SEIZED

CALL INFORMATION
RECEIVED BY
OUTPULSING REGISTER

~
OUTGOING TRUNK
SCANNED FOR
READY SIGNAL

~
READY SIGNAL
DETECTED

U

...:--

READY SIGNAL SENT

~
OUTPULSING OF
DIGITS BEGINS

~
CUSTOMER OFF·HOOK
STATUS VERIFIED

+

LAST DIGIT OUTPULSED;
OUTPULSING REGISTER
RELEASED

..

--..

I

"

INCOMING DIGITS
STORED IN INCOMING
REGISTER

Figure 3-7. Call advancement to the terminating
central office.

equipment in Mrs. Cooper's central office will periodically scan for this
"ready" signal. When this "ready" signal is detected, outpulsing of digits
begins. If central office B contains a single central office code, only the
last four digits of Mrs. Mahon's number will be transferred. This is
because all calls on the direct trunk group will terminate at central office
B. However, if office B contains more than one central office code, additional digits must be transmitted to identify the particular central office
code serving Mrs. Mahon.
96

Chap. 3

Introduction to the Network

97

Before the last digit is sent, the control equipment checks to see that
the calling customer's line is still off-hook. If the calling customer has
hung up (abandoned the call), the control equipment will terminate the
call-processing sequence and release associated equipment and circuits.
When the last digit is outpulsed, the outpulsing register is released.
The digits are now stored in the incoming register at central office B.
(Sections 8.4 and 8.5 discuss several techniques for sending the digits
between central offices.)
3.4.5 CALL COMPLETION (FIGURE 3-8)
Once the digits are stored in an incoming register at the terminating
office, many functions are initiated and supervised by the control equipment. The 4-digit line number is translated to Mrs. Mahon's physical
line termination. The status of Mrs. Mahon's line is interpreted and
signifies that the line is idle. (If Mrs. Mahon's line were busy, a busy signal would be returned to Mrs. Cooper.)
The incoming trunk is connected through the switching network to
Mrs. Mahon's line. A ringing register is seized, the incoming register is
released from this call, and Mrs. Mahon's telephone rings. An audible
ring, a tone that has the timing of a ringing signal and that indicates that
a ringing signal is being applied to Mrs. Mahon's telephone, is sent back
to Mrs. Cooper at this time. 16 The control equipment at the terminating
office will scan Mrs. Mahon's line status for an answer (off-hook) indication and, when it is detected, will terminate the ringing signal and return
answer supervision to office A. This will be used to record answer or
connect time for billable calls.
Mrs. Mahon answers the phone, and the conversation begins. As Mrs.
Cooper talks into her handset, the acoustic speech signal is converted into
an electrical signal by the transmitter in the handset. The signal generated by conventional transmitters is an electrical analog of the acoustic
signal. This electrical analog of the speech may proceed through the
switching systems and transmission facilities to Mrs. Mahon's telephone
in that form, or it may proceed through part of its path in digital form.
The latter would then require analog-to-digital and digital-to-analog
conversions.
With conventional technology, the signal reaching Mrs. Mahon's telephone will be analog, and the receiver will convert the analog signal
back to an acoustic signal. The acoustic signal from the receiver is not an
exact reproduction of that at the transmitter. One reason for this is that
the frequency content is limited by the transmission path (see Section 6.2). Also, impairments such as noise and loss occur, and if the call
16 Although initiated at the same time, the audible ring is separate from the ringing signal
that activates the ringer in the called party's telephone.

CENTRAL OFFICE A

CENTRAL OFFICE B

AUDIBLE RING PASSED THROUGH ~--':'----I
CONNECTION TO MRS. COOPER

DISCONNECT

Figure 3-8. Completion of the call.

Chap. 3

Introduction to the Network

99

travels a long distance, an echo effect could occur. (These impairments
and ways of controlling them are discussed in Section 6.6.)
During the conversation, the originating office, office A, monitors the
outgoing trunk to office B for disconnect. If the calling party hangs up
first, the connection is released, and disconnect supervision is sent to the
terminating office. The trunk is idled when the terminating office returns
on-hook supervision.
If the called party (Mrs. Mahon, in this example) hangs up first, a
timed-release period of 10 to 11 seconds is initiated. The connection is
released after this time-or earlier if the calling party hangs up.
Completion of the call is detected and recorded at central office A for
accounting purposes if there is a charge for the call; that is, if it is not
covered by a fixed monthly charge or a flat rate. When the call is first
dialed, the control equipment in central office A determines whether the
call is billable by the routing information associated with the first three
digits (see Figure 3-6). If the call is billable, a register is requested from
an automatic message accounting system to receive information that is to
be recorded about the call. For Mrs. Cooper's call, the information
recorded includes the number of Mrs. Cooper's telephone, the number
dialed, the time Mrs. Mahon answered, and the time the connection was
released. Data on this call and other billed calls from central office A are
forwarded to a data-processing accounting center where they are periodically processed to compute customer charges. If the call is billable, Mrs.
Cooper's next monthly telephone bill will include a charge for the call.
(Section 10.5 describes how the data are processed to determine the
charge, and Section 13.2.2 discusses operations at the accounting center.)
Thus, a basic telecommunications service-the simple telephone
call-requires a relatively complex sequence of events.

AUTHORS

J.

L. Bazley
T. Pecsvaradi

PART TWO
NETWORK AND SYSTEMS
CONSIDERATIONS

The five chapters in this second part present the basics of network structure and planning and explain engineering considerations applicable to
network and customer-services systems. Network functions such as
transmission, switching, and signaling are discussed, and a background of
telecommunications concepts, principles, and technology is presented
with emphasis on relevance to the related network and system topics.
This should provide a general foundation for the description of specific
network and customer-services systems in Part Three.
Chapter 4 expands on the introductory material in Chapter 3 and
describes the structure of traffic and facilities networks. It explains the
difference between local and toll networks, discusses the role of network
planning, and describes the unified numbering plan. Chapter 5 presents
basic traffic concepts through a discussion of topics such as traffic theory
and its application to the engineering of switching systems and trunk
groups. Interrelated traffic considerations that affect network engineering, the management of traffic networks, and considerations in the design
of data networks are also discussed.
Chapter 6 presents fundamental transmission principles and technology. Concepts such as signals, channels, media, modulation, and multiplexing are examined, and transmission impairments and objectives are
discussed. Beginning with a discussion of the basic role of switching,
Chapter 7 explains switching functions, networks, and control. In addition to a description of the basic switching functions involved in establishing a connection, the various auxiliary functions are also discussed.
Part Two concludes with Chapter 8, which explains the fundamentals of
signaling and interfaces in terms of the basic functions they provide in
the network and some of the methods and concepts involved in fulfilling
their roles.
101

4
Network Structures and Planning

4.1 INTRODUCTION
The telecommunications network, as defined in the last chapter, has two
interrelated aspects. In terms of its physical components-station equipment, transmission facilities, and switching systems-the telecommunications network is a facilities network. To provide various services, those
facilities are interconnected in many ways; in this sense, the telecommunications network is a set of traffic networks that share the facilities.
This chapter continues the discussion of the telecommunications network with descriptions of the structures of the facilities and traffic networks. For traffic networks, the emphasis is on the largest and most
complex-the public switched telephone network (PSTN)-although
private networks are covered briefly.1 The discussion also includes a
description of the PSTN worldwide numbering plan. The last section of
the chapter discusses considerations and approaches used in planning the
configuration of the telecommunications network so that it continuously
meets constantly growing and changing demands.

4.2 STRUCTURE OF TRAFFIC NETWORKS
4.2.1 THE PUBLIC SWITCHED TELEPHONE NETWORK
The PSTN actually consists of two strongly interdependent networks: the
local network (sometimes called the exchange area network) and the toll
network. The interdependence results from extensive integration and

1 As noted in Section 2.1.2, special services (which include private networks) have been
growing rapidly. One result of this growth is that about 5.2 million (43 percent) of the
approximately 12 million interbuilding circuits in service in the Bell System in 1982 were
special-services circuits.

103

104

Network and Systems
Considerations

Part 2

sharing of functions to reduce overall network costs. The following discussion, which is designed to convey basic concepts, presents a simplified
view of the local and toll network structures. The actual structure is far
more complex due to the variety of ways in which network functions are
integrated to meet diverse needs in particular segments of the network.

Local Network Structure
The structure of the local network begins with customer station equipment connected by loops to local switching systems. All customers connected to a local switching system (central office) in a particular central
office building2 are said to be located in a wire center area, and the location
of the building is called the wire center. These concepts are illustrated in
Figure 4-1. Customers located within a wire center area communicate
with each other through the local switching system, or systems, at the
wire center. As indicated in Section 3.1, this arrangement reduces network costs by adding some switching costs in return for a large reduction
in transmission costs.
Figure 4-2 illustrates the concept of judiciously combining switching
and transmission to minimize overall costs in the local network. In the
2-level switching hierarchy shown, which is typical of most metropolitan
areas, the switching systems at adjacent or nearby wire centers are connected by trunks, either directly or through one or two tandem switching
systems. Thus, customers in adjacent or nearby wire center areas communicate with each other using their dedicated loops and the trunks
interconnecting their local and tandem switching systems.
Whether it is more economical to provide direct trunks between two
adjacent wire centers,· to interconnect them indirectly using tandem
trunks and tandem switching systems, or to use a combination of both
depends on the traffic volumes, the distances involved, and the opportunities for sharing the facilities among many customers.
In Figure 4-2, there is a strong community of interest (high traffic
volume) between offices at wire centers A and B, justifying a direct trunk
group (represented by the dashed line). Traffic between wire center C
and the other two wire centers does not warrant direct trunk groups and
is carried by tandem groups (represented by solid lines) through tandem
office T. Using tandem trunk groups and switching systems to provide
service in a local area usually involves longer transmission paths and
more switching but proves to be more economical when the traffic
volumes between pairs of local switching systems are very low.
For intermediate traffic volumes, the most economical solution may be
a combination of direct and tandem trunks. The routing technique that
2 As noted in Chapter 3, a central office building may contain one or more switching
systems.

EXCHANGE
AREA 2

EXCHANGE
AREA 5

EXCHANGE AREA 1
•

WIRE
CENTER

A

WIRE
CENTER
AREA
A

\

•

\

\
\

WIRE
CENTER
B

WIRE
CENTER

\

--------\

ARBEA

\,

EXCHANGE
AREA 4

EXCHANGE
AREA 3

CENTRAL OFFICE BUILDING A
AT WIRE CENTER A

CENTRAL OFFICE
CODES

456
567
678

.......

CENTRAL
OFFICE

CENTRAL
OFFICE

1

2

.....
......

CENTRAL OFFICE
CODES

123
234
345

TRANSMISSION
EQUIPMENT

POWER AND BUILDING
SUPPORT EQUIPMENT

Figure 4-1. Local network topology. Top, wire center areas and exchange
areas. Traditionally, a single, uniform set of charges exists for telephone
service in an exchange area, and a call between any two points in the area is a
local call. (Different meanings will apply after divestiture.) Bottom, central
office bui~ding terminology.

TANDEM OFFICE T

DIRECT TRUNK GROUP
(HIGH USAGE)

Figure 4-2. A typical local network configuration.

takes advantage of this network structure is called automatic alternate routing. With automatic alternate routing, a switching office first offers a call
to a high-usage (HU) trunk group (a primary direct route between two
switching systems). If all trunks in the HU group are busy, the call is
routed via a tandem office using a final trunk group. Final trunk groups
are the final routes traffic can take. When all transmission paths on a
final trunk group are busy, the calling customer is sent a reorder tone
and must try again later.
It is important to note that the terms direct trunk group and tandem
trunk group describe the topological structure of the network, while highusage trunk group and final trunk group refer to the manner in which the
trunk groups are used when routing traffic on them. Consequently, a
direct trunk gro.up is not necessarily an HU trunk group. If, for example,
the traffic between offices A and B in Figure 4-2 were not allowed to
overflow to the tandem trunk groups, then the direct trunk group would
not be an HU group. HU and final groups are traffic-engineered
differently, as described in Section 5.4.
Section 5.5 describes the method of determining the most economical
apportionment of HU and final trunks. As discussed there, an important
factor in the determination is trunk group efficiency-the concept that
average traffic carried per trunk increases with trunk group size. This
concept is the basis for efforts to concentrate traffic into larger parcels, a
major consideration in the design of traffic networks.
106

107

Network Structures and Planning

Chap. 4

When the number of central offices in a metropolitan area is large, it
may be advantageous, in terms of trunking arrangements, to group central offices in sectors that reflect communities of interest. Each sector is
served by a tandem office located near the traffic-weighted center of the
sector. This reduces the total length of tandem trunking. The attendant
penalty is that some calls may require three consecutive tandem trunks3
and, consequently, may have a slightly longer set-up time. A more complicated structure with more routing choices may also result if certain
office pairs, such as A and D in Figure 4-3, reside in different sectors but
have enough traffic between them to justify a direct trunk group.

TANDEM
OFFICE
C

TANDEM
OFFICE
B

.... ....

"

,"

.........

,""

,..... >"
.... "

........ ....

DIRECT TRUNK GROUP

LOCAL OFFICE A

•

.....

'

,," '

,

..... .....
......
LOCAL OFFICE D

LOCAL OFFICES
HIGH-USAGE TRUNK GROUPS

Figure 4-3_ A 2-tandem-office local network_

Toll Network Structure
As mentioned earlier, the switching configuration shown in Figure 4-2 is
a 2-level hierarchy. While it provides an economical tradeoff between
switching and transmission costs within most local metropolitan areas, it
is not the most economical structure for interconnecting all switching
offices in the nationwide network. The volume of traffic between widely
spaced local offices is usually small, and direct trunks to serve this traffic
would be too expensive. Therefore, the switching hierarchy now used in
the United States has five levels (illustrated in Figure 4-4), in which
successively higher level offices (also called classes) concentrate traffic
from increasingly larger geographical areas.
3 Trunks interconnecting tandem offices are sometimes called intertandem trunks.

REGIONAL CENTER 0
CLASS 1

FINAL

SECTIONAL CENTER C
CLASS 2

HU6

REGIONAL CENTER E

"""
.oJ

"

.oJ

<
z

ii:

~
I

<
z

I
I

ii:

I
PRIMARY CENTER B
CLASS 3

I

HU7

I
I

I
I
I

.,)

"'" "'"

TOLL CENTER A
CLASS 4

TELEPHONE 1

."

TOLL CENTER H

TELEPHONE 2

Figure 4-4. The switching hierarchy.

Class 5 offices (also called local switching offices and end offices) are part
of the local network. The toll network consists of the class 4 and higher
switching offices (toll centers, primary centers, sectional centers, and'
regional centers as shown in Figure 4-4) and the trunks interconnecting
them. Table 4-1 gives the number of offices at each level of the PSTN
hierarchy. All trunks within the toll network are called intertoll trunks.
Class 5 end offices are connected to the class 4 toll centers by toll connect-

ing trunks.

108

Chap. 4

109

Network Structures and Planning

TABLE 4-1
DISTRIBUTION OF OFFICES IN
THE PSTN HIERARCHY (1982)
Class

Bell operating companies
Independent telephone companies

1

2

3

4

5

10

52

148

508

9803

20

425

9000

An office connected by a final trunk group to a higher class office is
said to "home" on that office, although it should be noted that offices do
not always home on the next higher class office. For example, while most
class 5 offices (end offices) home on class 4 offices (toll centers), some class
5 offices home on class 3, 2, or 1 offices.
Typically, offices in the same homing chain are located relatively near
each other. Consequently, many final trunk groups are only a few miles
long. The final and HU trunk groups that interconnect two different
regions (that is, two different homing chains) are the long-haul trunks of
the toll network. If the volume of traffic between offices of the same class
or differing in class by one is high enough, it is more economical to connect them directly by HU trunks (HUI through HU6 on Figure 4-4) than
to send traffic by some indirect path. Sometimes traffic between offices in
the same homing chain justifies interconnection by an HU group, as illustrated by HU7 in Figure 4-4.
The basic rule for routing a toll call is to complete the connection at
the lowest possible level of the hierarchy, thus using the fewest trunks in
tandem. In Figure 4-4, a call from telephone 1 to telephone 2 first goes to
the local office (a class 5, or end office). That office recognizes the call as
a toll call and sends it over toll connecting trunks to toll center A. Toll
center A searches for an idle HU trunk, first in the HUI group, then in
the HU2 group. If all trunks in those groups are busy, the call overflows
to the final trunk group connecting toll center A and primary cen!er B. If
all trunks in this final group are busy, the call is blocked and the customer receives a reorder tone. Otherwise, the call reaches primary center B,
and the sequential search procedure continues up through the hierarchy,
searching HU3 through HU6 and related final groups until either an idle
path (sequence of trunks in tandem) between the two end offices is found
or all possible routing alternatives are exhausted. The average toll call
uses slightly over three trunks, including toll connecting trunks. The
maximum number of trunks that may be used in a connection is nine.

Network and Systems
Considerations

110

Part 2

The actual structure of the network is far more integrated and complex than the simplified local and toll network structures just described.
For example, advances in switching technology allow the introduction of
local switching systems that can record billing information and perform
alternate routing, thus combining local, local tandem, and toll functions.
Figure 4-5 shows the distribution of the different switching functions
over the Bell System switching systems. The net effect of sharing among
switching systems is a reduction in switching and transmission costs.
LOCAL TANDEM
SWITCHING FUNCTIONS
(550)

LOCAL SWITCHING
FUNCTIONS
(9900)

TOLL SWITCHING
FUNCTIONS
(980)

Figure 4-5. Distribution of switching functions over Bell System switching
systems. The numbers represent the number of switching systems performing
the indicated functions at the end of 1978. The union of any two (or all three)
circles represents systems providing a combination of functions. For example,
400 systems provide a combination of local and toll functions.

One manifestation of the ability to combine or integrate local and toll
functions in one switching system is that toll calls between certain pairs
of end offices are not carried on toll connecting trunks up to the class 4
toll center on which the end office homes. Rather, they are carried on
end office toll trunks. Many of these are direct trunk groups that carry toll
traffic only between two end offices. Other end office toll trunks may
carry toll traffic from one end office to the class 4 toll office on which a
distant end office homes. These trunk groups are HU groups, so that any
overflow from them would be routed through the toll hierarchy in the
usual manner. Because they save on toll switching costs, end office toll
trunks are economical when there is a sufficiently high community of
interest between two end offices that are a considerable distance apart. In
this arrangement, the switching systems involved must be able to record
billing information and perform alternate routing. This type of trunking
is prevalent in certain areas of the country. In New Jersey, for example,
nearly 36,000 of the approximately 168,000 trunks carrying toll traffic at
the end of 1982 were end office toll trunks.

111

Network Structures and Planning

Chap. 4

4.2.2 OPERATOR FUNCTIONS
Section 2.5.1 describes the operator services provided for PSTN customers. The related operator functions are performed primarily in conjunction with computer-based operator systems and automatic call distributors
(ACDs).4 Toll switchboards account for a small and declining percentage
of operator positions. Two systems that automate certain operator functions appear in Figure 4-6. The Traffic Service Position System (TSPS)
automates many functions of the toll-and-assistance operator related to
billing calling card, collect, coin, and person.:.to-person calls. Most intercept functions are now provided by the Automatic Intercept System (AIS).
With AIS, most intercept calls are handled automatically but operators are
available to handle unusual problems and provide additional clarification
to customers, if necessary.s
Connections to the PSTN switching hierarchy from the switchboards
and automated systems are shown in Figure 4-6 and are described below.

r-------------,
I

I
1-----,--1
~---~

TO REST
OF PSTN

TO CAMA
OFFICE

i-tr;--------I

I

L-

-- ---- --

I
RATE·AND·ROUTE
OPERATORS

I

,

I
I
I

,

I
I
I
I
I

--,

TSPS
OPERATORS

I
I
I
I

I

CAMA.ONI
OPERATORS

-r ---~:.:,---- -----------

DIRECTORY
ASSISTANCE
OPERATORS
(TYPICALLY A
COMMON TEAM)

J

END OFFICE
TOLL-AND-ASSISTANCE
OPERATORS

NUMBER·SERVICE OPERA TORS

CLASS 5
OFFICES

e
NPA

CLASS 4 OFFICE (OR HIGHER LEVEL OFFICE IN TOLL HIERARCHY)
NUMBERING PLAN AREA

Figure 4-6. Typical topology of operator functions. Note: Connections to
directory assistance operators show numbers dialed by originator of request.

4 Section 11.2.6 describes ACDs.
5 Section 10.4 describes TSPS and AIS.

:

I

112

Network and Systems
Considerations

Part 2

• Operators at TSPS consoles (or one of the remaining toll switchboards)
are interconnected with the PSTN in several ways. The most common
function, providing assistance on outward calls, requires connection
between class 5 offices and a toll office. Other functions may require
connection to dedicated facilities-such as specialized trunks to class 5
offices for busy-idle testing. These facilities are usually accessed via
toll offices. To perform some of their functions, operators at TSPS
consoles require connections to directory assistance and rate-and-route
operators.
• Centralized automatic message accounting - operator number
identification (CAMA-ONI) operators are temporarily connected to
customer-dialed, station-to-station calls to identify the caller's telephone number to CAMA equipment when the local office does not
have AMA and automatic number identification (ANI) capability.
These operators can be connected to any toll office (class 4 or higher)
that provides a CAMA function. 6
• Directory assistance operators respond to customers' requests for unknown telephone numbers. In most cases, the customer dials different
numbers for directory assistance depending on the location of the
requested number relative to the calling customer as described below.
For numbers in the local calling area, 411 is dialed. Local offices
close to the ACD handling directory assistance are connected by
direct trunk groups; more distant offices typically connect through
a toll office where directory assistance traffic is concentrated on a
trunk group to the ACD.
For numbers in the same or home numbering plan area (HNPAf but
not in the same local calling area, 555-1212 is dialed. These calls
are typically routed to a toll office where they are concentrated on
a trunk group to the ACD.
For numbers in another or foreign NPA (FNPA), NPA-555-1212 is
dialed. These calls are routed through the toll network to a terminating toll office near the ACD and concentrated on a trunk
group to the ACD.
• Intercept operators (or AISs) respond to calls made to nonworking
numbers (changed or unassigned numbers). These operator positions
are connected on direct trunks from terminating local offices.
• Rate-and-route operators provide toll-and-assistance operators with
routing codes, rate information, and lists of numbers that may be coin
6 Section 10.5.4 describes CAMA operation.
7 Section 4.3 discusses the worldwide numbering plan.

Chap. 4

Network Structures and Planning

113

telephone lines. The rate-and-route operators are connected through
the toll network to toll-and-assistance operators.
4.2.3 STRUCTURE OF PRIVATE SWITCHED NETWORKS
The communications, needs of large corporate customers can often be
satisfied at lower cost by private networks. Private networks may also
provide features or capabilities that are not available with the PSTN. (See
Section 2.5.)
In many respects, the structure of private networks is similar to that of
the PSTN. For instance, a typical Enhanced Private Switched Communications Service (EPSCS) network provides large corporate customers with
an advanced, electronically switched voice and data communications service. It links thousands of telephones, private branch exchanges (PBXs),
and data sets in hundreds of geographically dispersed customer locations
throughout the country.8 Customer-premises equipment is connected by
access lines to EPSCS switching systems, which are lESS switching systems (see Section 10.3.3) especially modified to provide customers with
high-quality transmission of voice and data over long distances. The
EPSCS switching systems, in turn, are connected by trunks.
The EPSCS network has a 2-level hierarchy, which means that most
EPSCS calls are completed through only one or two switching systems
and usually involve no more than two or three trunks in tandem. For
example, in Figure 4-7, a call made from the San Diego PBX to the Miami
PBX of a large corporation is routed most directly and economically
through the EPSCS switching systems at Los Angeles and Atlanta over
the high-usage trunks between them. If all trunks on this route are busy,
the call overflows to a less direct route. The trunking arrangement
shown in the figure suggests several alternative routes, with the final
route through all four EPSCS switching systems.
It is important to note that parts of the EPSCS networks are shared
among different EPSCS customers and parts are dedicated to individual
customers. For example, the lESS systems are usually shared because
most customers do not have enough traffic to justify a dedicated switching system. But the access lines from the customer's PBX to an EPSCS
switching system, the trunks between EPSCS switching systems, and the
switch terminations are dedicated to a single EPSCS customer.
Private and public networks also share facilities. Thus, most EPSCS
switching systems also provide ordinary PSTN services. Furthermore,
even the dedicated trunks between EPSCS switching systems are provided by transmission facilities that may contain many other private and
public circuits. It is through such pervasive sharing of facilities that
economies of scale can be realized in the telecommunications network.
8 Sections 11.1.2 and 11.2 describe the operation of data sets and PBXs, respectively.

CHICAGO

PHILADELPHIA

CUSTOMER

CUSTOMER

PBX

PBX

SAN DIEGO

MIAMI

FINAL TRUNKS

o

HIGH-USAGE TRUNKS
1 ESS SWITCHING EQUIPMENT

Figure 4-7_ A typical EPSCS network for a large corporation with offices in
San Diego and Miami. The network has a 2-level hierarchy with the switching
systems at Los Angeles and Atlanta homing on systems in Chicago and
Philadelphia, respectively. This structure creates an efficient alternate-routing
private network.

4.3 NUMBERING PLAN
A fundamental requirement for a switched telephone network is a
numbering system that identifies each main station 9 by a unique address
that is convenient, readily understandable, and similar in format to those

9 A telephone that is connected directly to the central office either by an individual or
shared line. Main stations include the principal telephone of each party on a party line.
They do not include telephones that are manually or automatically connected to the
central office through a PBX or extension telephones (telephones that have been added to
an individual or shared line to extend the telephone service to other parts of the
subscriber's home or business premises).

114

Chap. 4

Network Structures and Planning

115

of other main stations connected to the network. Since 1947, this need
has been met for the PSTN in most of North America by an integrated,
unified numbering plan. Within the plan, each main station is assigned a
unique 10-digit address consisting of a 3-digit area code (representing a
numbering plan area), a 3-digit central office code, and a 4-digit station
number. In some instances, a prefix, a suffix, or both may be added. In
addition to providing telephone numbers for use in all customer-tocustomer traffic (including direct distance dialing [DDD] calls), the
numbering plan provides for:
• calls from a customer to a telephone company service (for example,
calls to directory assistance)
• calls from an operator to a customer (for example, calls established by
an operator at a prearranged [deferred] time)
• calls from one operator to another (for example, calls that cannot be
dialed directly by a customer).
When international DDD (IDDD) became possible, the numbering
plan was expanded to provide access to countries outside North America.
The following paragraphs describe the numbering plan for the
national and international PSTN. Numbering plans used in other traffic
networks (for example, private networks) are outside the scope of this
discussion.

4.3.1 NOMENCLATURE
The following set of symbols is commonly used in discussing the
numbering plan and dialing procedures:
N
X
0/1

==
==
==

Any digit 2 through 9.
Any digit 0 through 9.
Either 0 or 1.

Some related terms are:
• Address -

A unique lO-digit number assigned to a main station.

• Central Office Code - A 3-digit identification number preceding the
station or line number. There may be up to 10,000 station numbers
per central office code. Several central office codes may be served by a
central office, as noted in Section 3.2.3.

116

l'\letWOrK ana ::>ystems
Considerations

Part 2

• Dual-Tone Multifrequency (DTMF) Symbol * - The star (or asterisk) button, located at the lower left in the standard 4-row, 3-column pushbutton array. (Section 11.1.1 discusses DTMF signaling.)
• DTMF Symbol # - The number sign (or pound sign) button, located at
the lower right in the standard pushbutton array.
• Listed Directory Number - The 7-digit number composed of the 3-digit
central office code and the 4-digit station or line number.
• Numbering Plan Area (NPA) - A geographical division defined by the
familiar "area code," within which telephone directory numbers are
subgroups. In North America, a 3-digit NO/IX code is assigned to
denote each NPA or area code.
Home NPA (HNPA) - The NPA within which the calling
line appears at a local (class 5) switching office.
Foreign NPA (FNPA) NPA.

Any NPA other than the home

• Prefix - Any signal dialed prior to the address. Prefixes are used to
place an address in proper context, to indicate service options, or both.
With international calls, for example, the prefix all marks a simple
IDDD call. The prefix 01 calls for special handling such as that
required for a person-to-person call.
• Service Code - A code, typically of the form NIl, that defines a connection for a service. Examples are 411 for directory assistance and
611 for repair.
• Station (or Line) Number - The final four digits of a standard 7- or
la-digit address. These digits define a connection to a specific customer's line within a central office code.
• Suffix - Any signal dialed after the address. Operators use suffixes,
for example, to indicate the end of dialing.
• System Code - A 3-digit code, usually of the form 1XX but including
OXX assignments, available only to operators or to switching equipment for use as part of a special or modified address to influence route
selection.
• Toll Center Code - A 3-digit code of the form OXX that identifies a
specific toll center and is available only for operating company use. It
is always preceded explicitly or implicitly by an area code; therefore,
toll center codes need not be used the same way in all NPAs.

Chap. 4

Network Structures and Planning

117

4.3.2 INPUT DEVICES AND DIALING PROCEDURES
The numbering plan must accommodate all authorized address input devices, for example, customers typically use rotary-dial or pushbutton telephone sets; operators rely largely on multifrequency keysets. The plan
must also accommodate the switching equipment that interacts with the
address input device and all authorized users, including system test and
maintenance personnel.
To reach a particular destination, dialing must follow a prescribed
sequence that may depend on the originating point. First, one or more
prefix digits may be needed; next, the address must be transmitted; and
finally, a suffix may apply. A customer served by a PBX, for example,
may dial the prefix 9 to access the PSTN. A DOD call may require the
additional prefix 1 and a lO-digit address. To reach the same number, an
operator would begin with the prefix KP (key pulse), a keyset signal that
unlocks the device provided to register the address digits to follow. (The
risk of dialing before dial tone and thus reaching a wrong number under
certain circumstances is thereby avoided in operator dialing since nothing
is accepted in the absence of a KP.) The same lO-digit address is used by
both customers and operators, but the operator indicates end of dialing
by the suffix ST (start).
In some cases, a dialing procedure is subject to critical timing. Critical
timing applies when a switching system cannot tell from the digits dialed
whether dialing is complete, as in 0+ (0 followed by other digits) dialing
sequences, since 0 alone may be the complete dialed input. Long-range
plans seek to minimize such timing. For cases where format alone is not
definitive, as in the "dial 0" case cited, the TOUCH-TONE telephone input
# will be increasingly available as an end-of-dialing (cancel timing) indication. Applications in international dialing and custom calling services
already exist.

4.3.3 HISTORY AND EVOLUTION
As noted in Section 4.3, the basic address format used in most of North
America for customer identification consists of ten digits, divided into a 3digit area code, a 3-digit central office code, and a 4-digit station
number. IO To avoid a difficult and traumatic transition for both telephone
companies and customers, growth in telephone service must be accommodated within the limits of this lO-digit format. The evolution of the
address format to accommodate growth is described in the following
paragraphs and the changes are summarized in Table 4-2.
10 In the toll network, many calls are carried in formats that differ from the standard 10digit DDD address. An operator call to another operator may require only the digits 121.
Conversely, an operator-dialed call to Mexico may be served most conveniently with an
ll-digit address. Most toll switching systems can process calls with addresses ranging
. from three to eleven digits.

l'letWOrK ana ~ystems
Considerations

118

Part 2

TABLE 4-2
EVOLUTION OF lO-DIGIT CODE
Period
of
Application

Listed Directory Number
Area Code
Office Code

Line Number

Original

NO/IX

NNX

XXXX

Present

NO/IX

NXX

XXXX

Future

NXX

NXX

XXXX

N
X
0/1

2 through 9
0 through 9
0 or 1

Originally, 10-digit addresses were of the form NO/lX-NNX-XXXX.l1
(Table 4-3 shows the allocations of the one thousand 3-digit codes that
characterized the original plan.) In most NPAs, the NNX-XXXX format is
still used for the 7-digit directory number, but the NXX-XXXX format has
been introduced on a limited basis and will find increasing application
where additional central office codes are needed. This less restrictive
NXX format embodies the concept of interchangeable office and area
codes (that is, the formats of area codes and central office codes will no
longer be mutually exclusive).
The concept of interchangeable codes was incorporated into the plan
in 1962 when it appeared that the basic set of 152 area codes possible
using the NO / IX format (shown in Table 4-2) would be exhausted by the
mid-1970s. 12 (Careful code management has postponed this date to about
the turn of the century.) A more immediate concern in many geographic
areas was the limit of 640 central office codes (NNX) per NPA. Accordingly, network preparation for the additional 152 office codes provided by
the NXX format (less NIl) was completed in 1971, with .Los Angeles and

11 Although all-number calling is now the system standard, telephone numbers have an
alphanumeric tradition. Despite the personal appeal of names (which often had local
geographical significance, for example, MUrray Hill 7-1234) rather than all-number
codes, letters were a basic barrier to the use of the full range of dial-code sequences and
to the uniformity of international addresses. Prior to all-number calling, directory
numbers were commonly referred to as "2L+SN" to call attention to the alphanumeric
usage. It should be noted, though, that the alphanumeric format also used the "3-4"
character subgrouping.
12 When there is no longer an economic alternative (for example, boundary realignment),
the normal procedure to accommodate growth is to split an NPA when it requires relief.
A recent split occurred in 1973 in Virginia, when NPA 703 was divided to make room for
NPA 804. Since the process of splitting involves substantial expense, requires changes
likely to produce customer annoyance, and imposes a need for networkwide
coordination, decisions to split are subject to close scrutiny. Transition plans are a
principal consideration.

119

Network Structures and Planning

Chap. 4

TABLE 4-3
CODE PARTITIONING
200 Toll Center and System Codes

I --------------------------~~,--------------------------~\
000 ...................................................... 099
100 ...................................................... 199
152 Area Codes ~
8 Service
Codes
/

~
200 ... 210
300
400
500
600
700
800
900

. . . 310
... 410
. . . 510
. . . 610
... 710
. . . 810
... 910

""
211
311
411
511
611
711
811
911

640 Central Office
Codes

~
212 ... 219
312
412
512
612
712
812
912

. . . 319
... 419
. . . 519
. . . 619
... 719
. . . 819
... 919

~
220
320
420
520
620
720
820
920

.....
.....
.....
.....
.....
.....
.....
.....

299
399
499
599
699
799
899
999

New York City as early application sites commencing in 1974. With the
NO/1X-NXX-XXXX format, mandatory use of the prefix 1 indicates that a
10-digit number will follow, while an initial digit N indicates that a 7digit number will follow. Nil calls are recognized individually.
When DOD was introduced, areas with step-by-step switching equipment (see Section 10.2.2) required a DOD prefix. Since step-by-step
equipment acts on each digit as it is received, a unique steering prefix
was required to identify toll calls. Calls with the local NNX-XXXX formats were subject to immediate local switching, so prefix choices were
limited to digits or digit combinations starting with 0 or 1. The digit 0
was already being used for operator access, leaving 1 as an apparent
choice. But service codes of the form 11X for repair service and directory
assistance were in common use and had to be accommodated. While
prefixes such as 112 and 115 were among early choices, the DOD prefix
now in use is predominantly 1. Conflicts with 11X service codes were
avoided by changing to codes chosen from the Nil series or by using
standard 7-digit numbers as an alternative.
Eventually, a need for a second prefix was established. The purpose
was to permit not only ordinary DOD but a combination of customer
dialing and options for operator assistance. The prefix 0, the only
remaining single-digit choice, was pressed into service with the understanding that 0+ calls would typically be associated with the "dial 0"
assistance calls that operators were handling in any event. (These calls
are listed in Table 10-1 in Section 10.4.1.)

120

Network and Systems
Considerations

Part 2

4.3.4 THE PRINCIPAL CITY CONCEPT
Both the toll hierarchy and the numbering plan are clearly linked to
geography. But how these plans are linked to one another is not obvious.
The answer lies in the principal city concept.
In seeking suitable routes to a destination switching node, a switching
system examines part of the address or destination code. If every 3-digit
area code defines such a destination without violating hierarchical access
rules, administrative functions are substantially simplified. The principal
city concept provides the needed association of a hierarchical node with
an area code by providing a common point, a principal city, to which all
calls to a given area code may be routed. The principal city, which is
usually located within the numbering plan area served, must be prepared
to route all valid calls delivered to it. Thus, a new central office destination code may be activated anywhere without coordinated preparation at
all possible call sources. It follows that any call for area code NO / IX may
be sent to its one principal city. On the other hand, selected calls for
area code NO / IX may bypass the principal city if analysis of the 6-digit
code (the area code and central office code) confirms the suitability of
such routing.
4.3.5 INTERNATIONAL NUMBERING
1000 depends as directly on numbering as does national ODD, but stan-

dards are now a matter of global concern. The Comite Consultatif International Telegraphique et Telephonique (CCITT) foresaw this need and
organized a study to determine how to satisfy it. The standard, approved
in 1964, establishes eleven digits as a preferred maximum length for
international numbers but allows twelve.
The international number is flagged by a dialed prefix, not internationally standardized, that alerts the switching equipment. The international number itself consists of a country code and a national number.
Country codes are standardized and vary in length from one to three
digits, the first digit of which constitutes the world zone number.
National numbers are the familiar telephone numbers used for domestic
long-distance service. (Table 4-4 gives world zone number assignments.)
The countries or zones anticipating the greatest telephone population
by the year 2000 were assigned the shortest country codes to allow for
longer national numbers. Specifically, the unified North American world
zone is 1 and the Soviet Union zone is 7. In these cases, world zone and
country code are the same. Other zones contain a mix of 2-digit and 3digit country codes (for example, 52 for Mexico and 502 for Guatemala).
Countries where the number of telephones to be served can be handled
by nine or fewer digits are assigned 3-digit codes. Certain combinations
of the initial two digits (22, 23, 24, 25, and 26 for world zone 2; 35 for

Chap. 4

Network Structures and Planning

121

TABLE 4-4
WORLD ZONE ASSIGNMENTS
World Zone

Principal Areas Covered

1
2
3,4
5
6
7
8

Canada, United States
Africa
Europe
South and Central America, Mexico
South Pacific
U.S.S.R.
North Pacific
Far and Middle East
Spare

9

o

NOTE: Specific country code assignments (see AT&T Long Lines 1982,
p. 122) tend to be stable but may be changed by mutual agreement.

world zone 3; 50 and 59 for world zone 5) are selected in forming 3-digit
codes. The other pairs are assigned as 2-digit country codes. Thus, from
the initial two digits, switching systems can determine whether the country code is two or three digits long.
The dialing sequence for an IDDD call is illustrated by the following
call from England to the United States. The customer in England dials
010-1-NXX-NXX-XXXX, where 010 is the international subscriber dialing
prefix used in England; 1 identifies North America as the world zone
(and, in this case, is the country code); and the remaining digits are the
familiar 10-digit address or national number used in North America.
The Bell System has authorized two prefixes for outwardbound IDDD.
The prefix 011 indicates simple coin or noncoin automatic calls. The
prefix 01 indicates a desire for operator assistance.

4.3.6 OTHER SPECIAL NUMBERING

International dialing, of course, is only one of a number of services
dependent on numbering. A selection of representative formats for other
services indicates that numbering is not static and that new services
invariably require some adaptation of numbering.
• Custom Calling Services - for Call Forwarding: *72 with TOUCH-TONE
dialing or 1172 with rotary dialing (followed after dial tone by the
7 -digit number of the telephone to which calls will be sent)

122

Network and Systems
Considerations

• Toll-free service (area code 800) 1212

Part 2

for directory assistance: 1+800-555-

• DIAL-IT network communications service (area code 900) Phone: 1+900-976-1313

for Sports-

• Calling card service - 0 followed by a 10-digit destination address, followed after signal by a 14-digit calling card number.

4.4 STRUCTURE OF THE FACILITIES NETWORK
Section 3.2 introduced the three major elements of the facilities
network-station equipment, transmission facilities, and switching systems. From that discussion, it is evident that station equipment is
dedicated to particular customers; and therefore, the quantity, type, and
configuration are dictated by the services a customer chooses. Transmission and switching facilities; on the other hand, are shared. For this reason, station equipment is omitted from the following discussion, and the
term facilities network refers only to transmission and switching facilities.
Many factors determine the structure of the facilities network. Some
of the more important ones are customer location, telecommunications
services desired, performance objectives, available communications technologies and their costs and performance characteristics, and the need for
redundancy in the network to protect against major service interruptions.
For discussion purposes, it is convenient to divide the facilities network into the local facilities network and the interoffice facilities network, according to the two major classes of transmission facilities: loop
and interoffice. No physical dividing lines separate these facilities; the
distinction is merely an aid in describing the characteristics of the overall
facilities network.
4.4.1 LOCAL FACILITIES NETWORK
The local facilities network consists of the local switching systems located
at wire centers and the loop transmission facilities through which customers are connected to local switching systems. The local switching systems are located at what can be considered the conceptual boundary
between the local facilities network and the interoffice facilities network.
In this sense, they belong to both networks.
A wire center area is divided into several (usually four) distinct geographical areas called feeder route areas. Feeder route areas are further
divided into allocation areas, and each allocation area is divided into one
to five distribution areas. Figure 4-8 shows a portion of the structure of the
local facilities network in a typical suburban area, including a wire center
building and one feeder route area. The inset shows the structure of a
distribution area in detail.

WIRE CENTER
BOUNDARY

NUMBER OF ULTIMATE
HOUSING UNITS TO BE
SERVED BY CABLE
RUNNING DOWN THE BLOCK

,-----I

~~~~

I
I
I
I
DISTRIBUTION CABLE

I

LATERAL CABLE

7 ---

DISTRIBUTION AREA
BOUNDARY

SERVING AREA INTERFACE*
(SAFELY PLACED ON A LOT AWAY FROM
THE ROAD AND OTHER HAZARDS)

• A rearrangeable cross-connect point between feeder and distribution cables
in the loop plant.

Figure 4-8. A suburban feeder route area.

124

Network and Systems
Considerations

Part 2

Loop transmission facilities have a tree-like structure: A feeder route
(typically, a large paired cable 13 ) extends out from the wire center into
each feeder route area; lateral cables are spliced to the feeder cables at
various points along the route; distribution cables are joined to the lateral
cables; and, finally, customers' station equipment is connected to the distribution cable. The feeder cables near the wire center are generally
quite large (containing 1200 to 3600 wire pairs). At each lateral cable
splice, feeder cables become progressively smaller as the distance along
the route from the wire center increases (indicated by the progressively
narrow feeder route line in Figure 4-8).
The primary transmission facility in the loop network is paired cable
suspended on telephone poles, placed in underground conduit, or buried
directly in the ground. In recent years, the application of pair-gain systems has become an important consideration in the design of the loop
network. To reduce the number of cables required, subscriber pair-gain
systems use carrier and concentrator techniques (see Chapter 9), thereby
enabling a number of customer loops to share the same wire pair and
electronics.
Local switching systems may be either electromechanical or electronic. 14 Switching system size depends on the number of customers in
the area served and the area's growth rate. System type depends on area
characteristics and the switching technology available at the time of
installation. Table 4-5 lists the average values of some parameters of the
facilities network in urban, suburban, and rural environments. Trunks
are provided on interoffice transmission facilities, and thus, they are a
part of the interoffice facilities network discussed in the next section.
However, trunks are relevant to wire center parameters because, as stated
earlier, they terminate at local offices.
A number of observations may be made about Table 4-5. The average
area served by a wire center in suburban and rural areas is about an order
of magnitude larger than the average urban area served. Although
greater switching economy might be achieved by serving even larger
rural areas, the savings would be more than offset by the additional cost
of the loop electronics necessary to maintain service quality. While about
85 percent of the customers served by small, unattended switching offices
in rural areas are located within 2 miles of the wire center, the average
length of those rural loops that extend beyond 2 miles greatly exceeds
the global average. Thus, a relatively small percentage of the customers

13 Paired (or multipair) cable consists of pairs of wires twisted together to form the core of
a communications cable. Section 6.3 discusses transmission media, including paired
cable.
14 Chapter 7 describes the concept of switching, and Chapter 10 describes switching
systems.

Chap. 4

Network Structures and Planning

125

accounts for a disproportionately large percentage of the total loop
mileage in rural areas.
In rural areas, the volume of traffic per main station is lower, and the
percentage of intraoffice calls (calls between customers served by the
same switching system) is higher. This aspect of traffic has a noticeable
impact on the average number of lines per trunk, which increases from
eight in urban wire centers to sixteen in suburban wire centers and
twenty in rural wire centers.
The statistics in Table 4-5 illustrate how factors such as geography,
population density, and calling patterns affect the structure of the local
facilities network. The next section discusses the significance of these
factors for the interoffice facilities network.

TABLE 4-5
AVERAGE WIRE CENTER PARAMETERS FOR THE PSTN
Parameter

Number of switching systems
Area served (square miles)
CCS / main station *
Intraoffice calling (percent)
Working lines
Trunks
Trunk groups

Urban

2.3
12
3.1

Suburban

1.3
110
2.7

Rural

1.0
130
2.1

31

54

66

41,000

11,000

700

5,000

700

35

600

100

5.5

* CCS/main station is a measure of traffic in which the unit of measurement is hundred call
seconds (CCS) per hour. For example, a typical urban telephone line is used for 310
seconds (approximately 5 minutes) during the central office busy hour (see Section 5.2.3).

4.4.2 INTEROFFICE FACILITIES NETWORK
The interoffice facilities network consists of interoffice transmission facilities, tandem switching systems,15 and local switching systems. (As mentioned earlier, local switching systems constitute the conceptual boundary
15 Tandem is used here in the generic sense, that is, any system that connects trunks to
trunks. Tandem switching systems may serve as tandem offices, toll offices, or
combinations.

Network and Systems
Considerations

126

Part 2

between the local and interoffice facilities networks.) The interoffice
facilities network can be divided into metropolitan, rural, and intercity
facilities, each of which is described below.

Metropolitan Networks
Table 4-6 shows the distribution of trunks from all class 5 end offices over
various trunk group sizes and distance bands. The information in Tables
4-5 and 4-6 helps characterize the metropolitan network. According to
Table 4-5, the average urban (metropolitan) wire center serves an area of
12 square miles, reflecting the fact that a metropolitan area has a large
number of wire centers relatively close together. Clearly, then, the
shorter trunks in Table 4-6 are mostly in metropolitan areas, and it is reasonable that distances between wire centers of up to 15 miles be used to
represent the metropolitan areas. Table 4-5 shows that an average of 600
trunk groups terminate in a metropolitan (urban) wire center. Applying
the 15-mile distance assumption to Table 4-6, it can be determined that
most of these trunk groups are of modest size, with over half containing
fewer than forty trunks.
These characteristics strongly influence the structure of the interoffice
facilities network in metropolitan areas. Although the structure of the
interoffice facilities network is more complex than that of the local facilities network, the same general principles of sharing facilities are applied.

TABLE 4-6
DISTRIBUTION OF TRUNKS FROM CLASS 5 OFFICES
(EXCLUDING INTERTOLL TRUNKS)
Trunk Group
Size (Number
of Trunks)

Distance Between Wire Centers
(Mileage Bands)

o to 5
(Miles)

5 to 10
(Miles)

10 to 15
(Miles)

15 to 20
(Miles)

20 to 25
(Miles)

Over 25
(Miles)

Total Percentage
of Trunks in
Each Trunk
Group Size
Band

Over 195
180 to 195
160 to 180
140 to 160
120 to 140
100 to 120
80 to 100
60 to 80
40 to 60
20 to 40
Under 20

5.0
0.4
0.4
0.7
0.9
1.0
1.1
1.3
1.6
2.3
3.6

3.4
0.4
0.7
0.5
1.1
1.3
2.0
2.9
2.7
6.3
10.2

0.8
0
0.2
0.2
0.2
0.6
0.4
1.3
2.7
6.3
14.6

0.2
0
0.1
0
0.1
0.1
0.1
0.2
0.6
2.2
10.5

0.1
0
0
0
0
0
0
0.1
0.1
0.3
5.3

0
0
0
0
0
0
0
0
0.1
0.3
2.5

9.5
0.8
1.4
1.4
2.3
3.0
3.6
5.8
7.8
17.7
46.7

Total
Percentage
of Trunks in
Each Mileage
Band

18.3

31.5

27.3

14.1

5.9

2.9

100.0

127

Network Structures and Planning

Chap. 4

Figure 4-9 shows the interoffice facilities network in downtown Chicago.
Many wire centers are located close together and are interconnected by a
transmission facilities network that has a grid structure following the pattern of the streets. From this structure it is clear that a relatively small
number of transmission facilities emanating from each wire center are
shared by the many trunk groups terminating at each wire center. The
TO
WILMETTE
TO
MORTON GROVE
AND SKOKIE
TO
MORTON GROVE

o
o

TO
DES PLAINES AND
PARK RIDGE

ONE WIRE CENTER

!}

TWO OR MORE CLOSELY
LOCATED WIRE CENTERS
TANDEM LOCATIONS
INTEROFFICE FACILITIES
(PRIMARILY CABLE)

TO
SCHILLER PARK
AND O'HARE
RIVER GROVE
TO
BENSENVILLE
ELMHURST

OAKPARKo-~~----~t------t-----t----~~t-~~~==L-~

FRANKLIN (BELL BLDG)
(CHICAGO NO.1)
(DEARBORN TEST)
CANAL
(CHICAGO NO.6)

WABASH
CONGRESS
(CHICAGO NO.2)
TO
LA GRANGE

SUMMITo--L~----f----<~------------~~--+--C>

~:g
§I~

!I~_____ ,
UlL.
I

OAK LAWN

I

......:-I+--+.L...__________+-__+-____~

HICKORY HILLS o-__...L.---1I-__

BEVERLY
(BRAINERD)
CHICAGO

-----

SUBURBAN
TOWARDS
PALOS PARK

BLUE ISLAND
TO
RIVERDALE
TO
HARVEY

Figure 4-9. A metropolitan interoffice facilities networkthe Chicago area.

TO
EAST CHICAGO
HAMMOND

128

Network and Systems
Considerations

Part 2

predominant transmission facilities are voice-frequency facilities (see Section 6.2) and digital carrier systems (see Section 9.4) on paired cable.
Most of the cables are placed in conduit, which explains why the gridlike structure of the network follows the street pattern. A route section
may contain both VOice-frequency circuits and digital carrier channels on
cable (sometimes, in the same cable) and may have a total cross section of
over 20,000 equivalent voice-frequency circuits.
It should also be noted that metropolitan networks contain both local
and local tandem switching offices. Several tandem offices are located in
the downtown Chicago area.
Rural Networks
As shown in Table 4-5, the wire centers in rural environments serve
large, sparsely populated areas and are located far apart. Consequently,
the trunks are usually longer than the IS-mile limit associated with
metropolitan trunks, and the trunk groups are small. (Table 4-6 shows
that 80 percent of the longer trunk groups [15 miles or longer] contain
fewer than twenty trunks.) Table 4-5 shows that rural wire centers are
quite small in terms of both the number of lines and the number of
trunks. There is also more intraoffice calling in rural areas than in metropolitan areas. This, in turn, means that fewer trunks are required to
interconnect rural wire centers.
These characteristics determine the structure of the interoffice facilities
network in rural areas. Figure 4-10 shows the interoffice facilities network in the Greenwood District of Mississippi. The area shown covers
approximately twenty counties and contains forty-six Bell System and
independent wire centers serving 120,000 telephones. The distance
between wire centers, the volume of traffic, and the low percentage of
interoffice calls in a rural area justify few direct trunks between wire
centers. Most trunks are in toll connecting trunk groups connecting class
5 offices to class 4 and class 3 offices. This results in a tree-like facilities
network in which many individual chains of transmission facilities
branch out from the toll offices in routes typically following major highways. Each transmission facility along a chain serves offices that are
located farther away from the toll center. Thus, in general, the closer the
transmission facilities are to the toll office, the larger their channel capacities. In this sense, the interoffice transmission facilities network in a
rural area is similar to the feeder network in the local facilities network.
Although open-wire 16 facilities still exist in certain areas, the primary
interoffice transmission facilities in rural areas are carrier systems (see
Chapter 9) on paired cable either buried directly in the ground or supported on telephone poles.
16 Open-wire lines consist of un insulated pairs of wires supported on poles. Section 6.3
discusses transmission media, including open-wire lines.

TENNESSEE
MISSISSIPPI

>a:
 t)], the probability that the delay exceeds some
number of time units, t. Figure 5-1 shows a possible delay distribution
for a fixed offered load and a varied number of servers.
1.0

0.015

Figure 5-1. Example of delay distribution. P(D> t) depends on how many
servers are available to handle a given offered load. The number is chosen to
meet a GOS objective. For example, the GOS objective for average dial-tone
delay is typically P(D>3 seconds)~0.015. In the case shown, n+1 servers
would be required. If n servers were used, P(D> 3 seconds) would be greater
than 0.015.

4 Section 11.1.1 discusses dual-tone multifrequency (DTMF) signaling.

Chap. 5

Traffic

151

It is also possible to define mixed Bee and BCD systems. If there are,
for example, a fixed number of waiting positions, requests that find all
servers busy and an idle waiting position are presumed to wait for service, while requests that find all servers busy and all waiting positions
filled are presumed to be cleared from the system. A mixed system is
most frequently encountered in PSTN electronic switching systems (see
Section 10.3) where only a given amount of storage area is available to
"remember" calls waiting for a particular equipment item.
In addition, it is not always true that waiting calls should be served in
order of arrival (FIFO). For example, when an electronic switching system becomes congested (there are more calls in progress than the switching system can efficiently process), waiting lines can become long, so
long that the customer may hang up (that is, defect from the queue). The
customer most likely to defect is the customer who has been there the
longest. During periods of extreme congestion, a last-in-first-out (LIFO)
queue discipline prevents the congested system from wasting time trying
to process a call that has probably defected. This makes the best use of
the system, since it increases the probability that the next call processed
from the queue has not defected. In any event, the shape of the delay
distribution depends strongly on whether the queue discipline is FIFO,
LIFO, or some other scheme.

5.2.3 ENGINEERING PERIODS

The decision to specify a GOS for a load on a given system is a good start,
but to which load should the GOS be applied? The formulas developed
in traffic theory typically assume that the load is stationary5 during the
period of time for which a traffic system is being analyzed. For the traffic
theory formulas to be useful, then, those time periods for which the load
is stationary must be determined and the GOS criteria applied to some or
all of them. (This does not imply that the load must be exactly the same
during each engineering period, only that the load variation must be
confined to a small range within a given period.)
In the PSTN, loads do not remain constant during every period of
every day. They may peak during certain periods of the day, particular
days of the week, or particular seasons of the year. However, in the Bell
System, the stationa~y period is generally taken to be an hour, since data
support the assumption that offered load may be considered stationary for
intervals of an hour or so. The load does vary from hour to hour; however, it usually varies in recognizable patterns. Data describing calling
5 Roughly speaking, a stationary process is one in which the parameters describing the
process (A, the average arrival rate, and T, the average holding time, in traffic theory)
remain constant.

Network and Systems
Considerations

152

Part 2

patterns between two local offices show that there are two or three distinct load peaks (busy hours) during the day-one in the morning, one in
the afternoon, and perhaps, one in the evening. Moreover, there will
usually be two or three periods during the year when the daily peaks are
higher than normal (busy seasons). For this reason, the Bell System
defines the engineering period (the period in which GOS criteria are to be
applied) as the busy season busy hour (BSBH), that is, the busiest clock hour
of the busiest weeks of the year. The decision to concentrate on BSBH
periods is an outgrowth of the Bell System service philosophy-to
provide high-quality, economical service during normal, daily use of the
network.
The identity of the busy season busy hour varies by geographical area.
For example, in metropolitan areas where business traffic tends to dominate, the BSBH typically occurs between nine and eleven o'clock on
weekday mornings. (The particular hour may occur from nine to ten, ten
to eleven, or from half past ten to half past eleven, etc.) Another peak in
traffic often occurs in the afternoon, and in some local offices, this may be
the BSBH. In local offices with a significant amount of residential traffic,
an evening busy hour may occur. Figure 5-2 shows the traffic variation
typical of a local office with business traffic predominating. Since the
busy season (as mentioned before, there might be two or three in a year)
normally lasts for a month or longer, loads will be measured in upwards
of twenty busy hours (one particular hour in five weekdays over four
weeks) two to three times a year.
Although traffic theory is useful in predicting the performance of a
given system for a given load submitted to a given number of servers,
considerable engineering judgment is required to select the particular

MORNING
BUSY HOUR

,L

C)

z

i=
c(

z
-0
£210a: 11.

AFTERNOON
BUSY HOUR

EVENING
BUSY HOUR

,L

O::li

u,w

ot:
a:c(
w

III

::Ii

:::l

Z

7 AM

9

11

1 PM

3

5

7

TIME OF DAY

Figure 5-2. Traffic variation with time of day.

9

11 PM

Chap. 5

Traffic

153

load levels (engineering periods) about which to be concerned. One consideration is the type of equipment being engineered. For trunk groups
(circuits between switching systems) the average of the twenty BSBH
measurements is used, giving rise to the concept of the average busy season
busy hour (ABSBH).6 A GOS criterion for such a trunk group might read:
For the ABSBH load, a call requiring a circuit in the trunk group should
encounter all trunks busy (be blocked) no more than 1 percent of the
time. Or the GOS might state: The average blocking over the twenty
BSBHs should not exceed 1 percent? In any event, the GOS for trunk
groups is always stated in terms of blocking because trunk groups are
engineered as blocked-calIs-cleared systems8 (that is, there are no waiting
positions for trunk circuits).
For equipment within a switching system (for example, digit
receivers), the average busy season busy hour load is not always a good
measure of system demand. When switching systems become congested,
they tend to spread the congestion to other switching systems in the network (for example, systems awaiting signaling responses from the congested system). So, peak loads are of more concern than average loads
when engineering switching equipment, and engineering periods other
than the ABSBH are defined. Examples of these periods are: the highest
BSBH and the average of the ten highest BSBHs. Sometimes, the
engineering period is the weekly peak hour (which may not even be a
BSBH).
In addition to selecting suitable engineering periods, it is necessary to
express the GOS criteria in meaningful terms. Since most switching
equipment items provide waiting positions, the GOS criteria are usually
expressed in terms of delay probabilities. For example, the digit receivers
in switching systems are commonly engineered in terms of dial-tone
delay. The number of digit receivers to be provided is chosen such that
the probability of dial-tone delay exceeding 3 seconds is less than a
specified value. To reflect the concern with performance during peak
periods, GOS criteria may be specified for multiple engineering periods.
(For dial-tone delay, criteria for three engineering periods are often
specified.) Figure 5-3 presents a simplified view of the effect of multiple
criteria. It shows two of the frequently used engineering periods: the
highest BSBH and the average of the ten highest BSBHs. One criterion
may dominate (determine the number of digit receivers required),
depending on the ratio of load values for the two engineering periods
(busy hours).
6 The ABSBH concept is limited to trunks.
7 These two criteria are usually not equivalent (see Section 5.4.2). The blocking for the
average of twenty loads (ABSBH) tends to be less than the average blocking across the
twenty loads.
8 At present, this is true for the PSTN, but there may be exceptions in private networks.

NUMBER OF SERVERS

n-1

n

n+1

0.20

"
w

(/)

C?

A
~
~

0.08

OFFERED LOAD

NUMBER OF SERVERS

m-1

"

m

m+1

0.20

w

(/)

C?

A
~
~

0.08

OFFERED LOAD

Figure 5-3. Effect of multiple criteria on engineering digit receivers. Two
frequently used criteria for engineering digit receivers are that the probability of
dial-tone delay greater than 3 seconds, P(D>3 seconds), be as follows:
P(D>3 seconds) =0.08 for the average of the ten highest BSBHs (a 1)
P(D> 3 seconds) = 0.20 for the highest BSBH (a 2) .
Relative values for a 1 and a 2 vary for different parts of the network. In some
cases, top graph, a 1 dominates-requires a greater number of servers (n+ 1).
In other cases, bottom graph, 8 2 dominates-requires (m+1) servers.

Thus, the selection of engineering periods and statements of GOS
criteria are critical components in determining the quality of service
customers will receive. (Section 5.7.2 discusses Bell System service
objectives.)
As mentioned previously, the Bell System service philosophy concentrates on engineering the network for the BSBH. However, the Bell System is also concerned with network performance when load levels are
154

Chap. 5

Traffic

155

substantially higher than, or calling patterns are significantly different
from, any BSBH. These include Mother's Day, Christmas Day, heavy
snow days, and periods of natural disaster such as floods and earthquakes.
Providing enough equipment to handle these infrequent dramatic peaks
would be far too costly. Network management maintains network
integrity at those times (see Section 5.6).
The distinction between BSBH and dramatic peak loads is important
because most of the useful traffic models developed describe systems that
are, in some sense, stable. The formulas resulting from these models do
not predict system performance under heavy overloads well. The following sections then, assume that the traffic demands are those encountered
during the busy season.
5.2.4 TRAFFIC THEORY TECHNIQUES

Telephone traffic cannot be predicted exactly, but it is reasonable and useful to view customer requests and service times as statistical processes
that can be described in terms of probabilities. Thus, the number of calls
that will arrive into the system during the next t seconds may not be
known, but the probability that there will be some number k of them can
be estimated. Similarly, the probability that a given call now being
served will leave the system during the next t seconds can be estimated.
These and other probabilities relevant to traffic analysis can often be
calculated using methods developed in a branch of applied probability
theory known as traffic theory.9 Many of the engineering methods used in
the telephone system are direct applications of traffic theory models.
Even for complex network operations that do not readily lend themselves
to traffic theory analysis, traffic theory concepts and results often form the
basis for engineering procedures.
Traffic theory models begin by ascribing some underlying probabilistic
structure to the processes of call arrivals and call holding times. One
common assumption about call arrivals, for example, is that the probability of an arrival during some small interval (T, T+t) is proportional only
to the length of the interval, namely, t. The constant of proportionality is
some number A (the average arrival rate), so that

Pr[call arrival in (T, T+t)] = At.

(2)

Cooper (1981, pp. SO-51) shows that the simple assumption in equation
(2) leads to a formula that calculates p(k; t), the probability that k calls
arrive in any interval of length t:
(M)k
p( k', t) = -k!- e-At .

(3)

9 This branch of applied probability theory is known by several other names, such as
queuing theory and congestion theory.

156

NetworK and ~ystems
Considerations

Part 2

Equation (3) is the well-studied Poisson distribution (Feller 1968, eq.
[6.1]), and any process described by equation (3) is, therefore, called a
Poisson process. Equation (3) is applicable to many physical situations with
a large (essentially infinite) number of potential users, acting independently, such that the percentage of users actually requesting service at
any time is small relative to the maximum possible. The population of
telephone users in a given central office displays this property, and equation (3) describes well the way prospective calls originate at a central office dur-

ing busy periods.
Of course, in a telephone network, the process describing calls departing from the network after they obtain a server must be considered as
well as the process describing call arrivals. As with equation (2), the
simple assumption can be made that, during an interval of length t, each
call in progress will terminate with probability I.d, where J.L, the departure
rate, is merely the reciprocal of the average holding time, that is, J.L = 1/1'.
H H(t) denotes the probability that a given arrival requires service for t
seconds or less, Cooper (1981, pp. 42££) shows that
(4)

Equation (4) is the well-studied negative exponential distribution (Feller
1966).
One of its more startling properties is its lack of
memory: Equation (4) describes the distribution of the call's remaining
time in the system no matter how long the call has already been in progress.
Of course, that was indirectly assumed by stating that J.Lt completely
defined the departure probability. Though this assumption may seem
unrealistic, it is, nevertheless, true that the negative exponential distribution

describes the distribution of telephone conversation times reasonably well.
Numerous studies through the years have verified this result.
When arrivals follow a Poisson process and holding times are negative
exponential, traffic is said to be random. Traffic theory models have been
developed for other cases, but the two most useful formulas in the Bell
System are based on the assumption of random traffic. These formulas,
developed by A. K. Erlang, are applicable to many traffic problems. The
next sections describe these models and show the resultant formulas
without proof. 10 Instead, the discussion emphasizes the underlying
assumptions and their implications for traffic engineering.
5.2.5 ERLANG'S BLOCKED-CALLS-DELAYED MODEL
H a random load, a = AT, is submitted to a system of c servers such that
blocked calls wait until served on a (possibly infinite) waiting line, waiting calls are served in order of arrival, k is an index of the number of
10 Cooper 1981 gives the derivations of these and other traffic formulas.

157

Traffic

Chap. 5

arriving calls, and Pj denotes the probability that an arriving call finds j
customers in the system (either being served or waiting); then

j == 0, 1, ... , c
(5)

j == c, c+l, c+2, ... ,

where
aC

Po==

[

(c-l)! (c-a)

+

c-I ak

k~

-1

-- ]

(6)

k!

Equations (5) and (6) are valid only when the load is less than the
number of servers (O~a  c), the system is
unstable and the queue length grows without bound.
The quantity (Pc+Pc+I+Pc+2+"') represents the probability that an
arriving call will be delayed. This probability is denoted by C(c, a) and,
from equation (5),
(7)

Then, from equation (6),
aC

aC
C(c, a) == (c-l)! (c-a)

[

(c-l)! (c-a)

c-l

+ k~

ak

-1

k! ] .

(8)

Equation (8) is Erlang's Delay Formula, also referred to as the Erlang C
formula. This formula has been extensively tabulated since many traffic
systems are modeled using the Erlang C assumptions.
Equation (8), however, is not sufficient for engineering purposes.
Because the Erlang C model describes a blocked-calls-delayed system, the
delay distribution is needed as well. Assuming Poisson arrivals, negative
exponential holding times, and :first-in-first-out queue service, it can be
shown that the conditional delay distribution (that is, the probability that
delay exceeds t, given delay occurs at all), denoted by P(D >t ID >0), is
given by the formula
P(D >t ID >0) ==

e-(c-a)fJJ •

(9)

Network and Systems
Considerations

158

Part 2

That is, the conditional delay distribution is negative exponential with
mean

-I D>O
D

T
=--,

c-a

(10)

where T = 1/J.t.
From equations (8) and (9), then, the unconditional delay distribution
P(D > t) is such that

P(D>t) = C(c,

a)e-(c-a)~t,

(11)

whose mean is given by

IT

= C(c,

a)_T_.
c-a

(12)

Equations (9) and (11) depend on the first-in-first-out assumption.
The average delays, however, given by equations (10) and (12), are valid
for any order of service. The negative exponential holding-time assumption in the analysis is important. Exhibiting delay formulas such as equation (11) for other holding-time distributions is complex and invariably
produces formulas that are numerically difficult. This is one reason the
Erlang C model is sometimes used as an approximation even in cases
where the model assumptions are not valid.

5.2.6 ERLANG'S BLOCKED-CALLS-CLEARED MODEL

Assuming the same system as in the previous section (c servers, random
load a), except that there are no waiting positions and blocked calls are
immediately cleared from the system, then Pj (the probability that arriving calls find j customers in the system) is equivalent to the probability
that an arriving call finds j servers busy. Obviously, j ~c for this system,
and it can be shown that

aj
pJ' = _ .....
1·__

.,

c

ak

~k!

k=O

j = 0, I, ... , c.

(13)

159

Traffic

Chap. 5

The probability that an arriving call is blocked, then, is given by Pc
but is commonly denoted by B (c, a), so that

aC
c!

B(c, a) = - - c

ak

(14)

~k!

k=O

Equation (14) is Erlang's Loss Formula,ll referred to as the Erlang B for:mula. It holds even when the load is greater than the number of servers
(a > c) because, unlike the blocked-calIs-delayed model in which all calls
are eventually served, the blocked-calIs-cleared (Bee) model allows calls
to be lost when all servers are busy. Therefore, the Bee system never
becomes unstable.
For a Bee system, then, the offered load can be divided into the load
lost and the carried load. The load lost, a, is easily calculated:
a = aB(c, a).

(15)

The carried load, L, is given by

L = a - aB (c, a) = a [1 - B (c, a)].

(16)

(In a blocked-calIs-delayed system, no calls are lost [a=O]; the carried load
equals the offered load [l=a].)
The efficiency, p, of a Bee system can then be defined as the load carried per server:

L
c

p= - .

(17)

The efficiency, p, is commonly referred to as the occupancy of a group
of c servers. Since a given server can never carry more than one erlang,
L is always less than c, implying that p~ l.
Perhaps the most important property of an Erlang B system is that it
produces an economy-of-scale effect. That is, for a fixed blocking, p
increases with increasing load. Figure 5-4, which plots the number of
servers required to produce 0.01 blocking as a function of offered load,
illustrates this fact. As offered load increases, the number of additional
servers needed to maintain blocking at 0.01 decreases. This property is
extremely important in designing telephone networks (see Section 5.5).
11 This formula is valid for any holding-time distribution with a finite mean.

OFFERED LOAD,

a

Figure 5-4. Servers versus load for B(c, a)

=

0.01.

5.3 ENGINEERING SWITCHING SYSTEMS
A switching system includes many different elements needed during the
processing of a telephone call, but not all elements of a switching system
are traffic sensitive. Trunk scanners, for example, periodically check each
trunk connected to a switching system to determine the trunk's busy-idle
status. The workload of the trunk scanners thus depends on the number
of trunks terminated and the necessary scan rate, but the workload is
independent of the number of calls in progress. For most switching system elements, though, the workload is sensitive to traffic, and these elements must be engineered accordingly. The next few sections provide
some examples of the problems involved in engineering switching system components and describe some engineering approaches.

5.3.1 DIGIT RECEIVERS
Calls encountering all digit receivers busy are, from the customer's point
of view, waiting for dial tone. As long as digit receivers are engineered
so that long delays are unlikely, it is reasonable to assume that the customer will wait for dial tone without defecting. Thus, a blocked-callsdelayed model is appropriate for estimating blocking probabilities. The
grade-of-service (GOS) criteria for digit receivers are related to delay
probabilities, however, not to blocking. Whether the Erlang C delay
equation is appropriate depends on the distribution of digit-receiver
holding times. (As noted previously, the Erlang C delay equations
assume negative exponential holding times.)
Since the negative exponential assumption is reasonable for dial-pulse
receivers, the Erlang C model performs well. Rotary dialing generates
different pulsing times for different digits, ranging from short (1) to long
(0); and the time between dialing successive digits is a random variable
with a wide range, at times, several seconds. This behavior, combined
160

Chap. 5

Traffic

161

with enough partial-dial abandons (where the customer hangs up before
dialing is complete) that generate small holding times, leads to an overall
dial-pulse-receiver holding-time distribution for which the negative
exponential is a good fit. Strictly speaking, the lack of memory cannot,
obviously, apply to dialing times. Once four or five digits have been
dialed, for example, the remaining holding-time distribution cannot be
the same as the initial distribution. But the discrepancy between the
lack-of-memory assumption and actual dialing behavior is not great
enough to render the approximation ineffective.
On the other hand, the Erlang C model cannot be expected to describe
dialing behavior for dual-tone multifrequency receivers accurately.
TOUCH-TONE dialing times are much shorter and less variable than
rotary dialing times. Each digit has almost the same pulse time, and
there is a tendency to dial rapidly. A constant holding-time assumption
might seem more appropriate and would be, except that the number of
digits per call is so variable (seven, ten, possibly three or four with
centrex/private branch exchange systems or from partial-dial abandons)
that the constant holding-time assumption is too optimistic. Despite the
inaccuracy of its assumptions, therefore, the Erlang C model is preferable
because it errs on the side of good service but not to the extent of adding
significant cost.
Similarly, the Erlang C model is used to engineer the digit equipment
needed for communication between switching systems (digit transmitters
and incoming digit receivers, for example). Delays for these items have
nothing to do with dial tone, since the customer has already received dial
tone and has entered the called-party digits; rather, delays in transmitting
and receiving digits between switching systems affect the time needed to
establish a talking path. The considerations, however, are similar to
those in the dial-tone-delay problem for originating users. For example,
the sending system must wait for a signal from the receiving system
before it can send the digits. The waiting time depends on the number
of digit receivers in the receiving system.
5.3.2 OPERATOR FORCE

In the mathematical sense, the number of operators required (the operator
force) can be "engineered" using the same modeling techniques .,as for
equipment items. Since calls directed to an operator that find all operator
positions busy have to wait, determining how many operators are needed
for a given GaS is a traffic problem. Using assumptions of arrival rates
and operator service times, traffic models can be developed to express
delay probabilities, and in fact, the Erlang C model (with adjustments to
account for known deviations from the model's assumptions) is used to
do that. Given the offered load and average operator service times, the
number of operators required is determined by the objective speed of

162

Network and Systems
Considerations

Part 2

answer (the delay in obtaining connection to an operator). Depending on
the type of service and the equipment, the objective speed of answer
ranges from 2 to 6 seconds.
Traffic theory considerations aside, however, operators are not
machines; and this fact plays an important part in engineering procedures. In particular, it is not enough to determine the number of
operators required for peak load periods, put them in service, and leave
them there permanently. The problem becomes one of forcing, that is,
scheduling the operator work force so that the desired grade of service is
met while minimizing the wasted expense of providing too many operators. The forcing problem thus involves careful forecasting of load variations from hour to hour, day to day, and week to week, not just the forecasting of some peak load period. 12
Computer-based systems developed jointly by Bell Laboratories and
AT&T provide operating company personnel with data on traffic volume,
measurements of time to answer, and service time. A modified Erlang C
model is used to forecast operator requirements by IS-minute intervals
during the week, 2 weeks in advance.
5.3.3 STORED-PROGRAM CONTROL SYSTEMS

Stored-program control (SPC) switching systems (see Section 10.3) process
thousands of simultaneous calls on a time-shared basis. The performance
of the stored program is traffic sensitive; that is, the more calls being processed, the longer it takes for the SPC to return to the next stage in processing a given call. It is possible, for example, for an SPC to generate
excessive dial-tone delay even though digit receivers are available,
because the controller may spend several seconds processing calls that are
already in progress before processing the digit-receiver request for new
call originations.
The SPC, then, has a definite capacity, that is, a maximum volume of
calls it can process during a given interval and still satisfy all GOS
requirements. This capacity depends on many factors including the
stored-program organization, the speed of the logic, the amount of
memory and hardware available, and the kinds of calls being processed.
(Different types of calls-rotary or TOUCH-TONE telephone, coin or noncoin telephone, intraoffice or interoffice, etc.-require different callprocessing functions or different processor work times for a given function.) Traffic theory models are not adequate for determining SPC capacity because of the complex interactions between traffic, software, and
hardware.

12 Since forcing is an operating company activity, Section 13.3.1 dis<;usses it further.

Chap. 5

Traffic

163

In the absence of suitable traffic theory models, simulation techniques
have been developed for the various SPC systems to measure system performance as a function of traffic input. Simulation results have been useful in constructing formulas that engineers can use to predict a given
system's capacity before it is placed in service. Once the switching system is in service, the formulas are used with measurement data to adjust
capacity estimates. This allows the engineer to predict when a given system will exhaust (when the forecasted traffic load will reach the estimated
call-handling capacity of the system) and to re-evaluate the prediction
regularly.
One important aspect of SPC traffic is the way the processors react to
heavy load conditions. With many such systems, once the number of
calls in progress becomes too great and the processor begins to give poor
service, deterioration accelerates rapidly. This acceleration occurs because
the deterioration of service in itself generates additional processor work
items for each call (moving calls in and out of waiting lines, for example).
In addition, the number of calls to be processed is likely to increase as
customers reattempt calls that have failed to complete. To compensate for
this effect, logic is included that allows the processor to monitor its performance and to take appropriate action when there are indications that
the load has exceeded capacity levels. These automatic overload control
actions can take many forms, such as deferring nonessential maintenance
tests, reducing the rate of scanning for new call originations, or changing
the order of processing various queues (see Section 5.2.2). Determining
the proper overload detection logic and setting the parameters of the control strategy are traffic engineering problems for which simulation techniques have also been particularly important.

5.3.4 CAPACITY CONSIDERATIONS- LOAD BALANCING

The preceding traffic engineering examples assume that all customers can
reach all idle servers at all times. In fact, depending on the traffic, an
efficiently designed switching network that gives access to the servers
may not provide an idle path at all times. Customer terminations are
grouped as shown in Figure 5-5, and an above-average load in one group
will result in more blocking than for a group with below-average load. It
is a property of the blocking curves that such an imbalance in loads
results in poorer average service than would result if the loads were
evenly distributed. Furthermore, one of the groups may be given very
poor service even when the average appears to be satisfactory. This can
be illustrated in the following example for dial-tone delay where the
servers are dial-tone registers. As shown in Figure 5-6, load balancing
equalizes service to customers, while at the same time, giving the best

SERVERS
CUSTOMERS

CUSTOMER
GROUP A

SWITCHING
STAGES

CUSTOMERS

CUSTOMER
GROUP B

Figure 5-5. Customer access to servers. In this simplified representation, an
above-average load from customers in group A may cause blocking on the links
from group A that exceeds the objective value, even though there may be idle
servers.

overall average service. Line finders in step-by-step systems, horizontal
groups in crossbar systems, and concentrators in electronic switching systems are the most critical components for load balancing. 13
The simplest and most economical way to achieve and maintain load
balance; is through control of routine line assignments: New customers
are assigned to lightly loaded components, and assignments to heavily
loaded components are avoided. 14 As customer disconnects occur, the load
on the heavily loaded components is reduced. However, when imbalance
is severe (that is, when some components are badly overloaded), correction by routine line assignment may take too long. It may be necessary
to disconnect active customers from overloaded components and transfer
them to underloaded components. Such corrective action is clearly more
costly.
The existence of an official Load Balance Index Plan reflects the importance of load balancing. Without load balancing, switching systems

13 Chapter 10 describes switching systems and their components.
14 Assignment of customers to specific line equipment is a network administration function.
Section 13.3.2 discusses network administration functions.

164

POORLY BALANCED SYSTEM

UNACCEPTABLE
SERVICE

+

POOREST·
SERVED ~
CUSTOMER

AcCEPTABLE---.- - - - - SERVICE
AVERAGE SERVICE

LOAD

--+

AVERAGE
LOAD

WELL·BALANCED SYSTEM

UNACCEPTABLE
SERVICE

+ --- ---

-;cCe;";;~--.SERVICE
AVERAGE SERVICE

LOAD

-+

AVERAGE
LOAD

Figure 5-6. Load-service curves. Because the rate of service degradation
increases with increasing load, for identical switching systems operating at the
same average load, average service is better in a balanced system than in an
imbalanced one. In addition, some customers in a poorly balanced system, top,
may receive an unacceptable level of service, while service to all customers is
equalized in a well· balanced system, bottom.

would have to operate at a load level estimated to be about 5 to 10 percent below their design capacity to maintain the same quality of service.

5.4 ENGINEERING TRUNK GROUPS
The previous section explored some of the traffic engineering problems
and techniques associated with switching systems. This section focuses
on some of the problems and techniques associated with the trunk groups
connecting the switching systems.
Several models exist to help determine trunk requirements. Typically,
an application model is used to generate trunking tables that are, in turn,
used by operating companies to determine the number of trunks in a
group for a given demand.

165

166

Network and Systems
Considerations

Part 2

5.4.1 APPLICABILITY OF THE ERLANG B MODEL

In the PSTN, there are two kinds of trunk groups: high-usage (HU)
groups, from which blocked calls are offered to other (alternate-route)
groups for completion, and last-choice or final groups, from which
blocked calls are given a reorder tone indicating that the caller should try
again later. IS Whether an HU group should be placed between two offices
and, if so, how many trunks it should have are discussed in Section 5.5,
which treats the larger question of traffic network design. Final groups,
however, are identified a priori by the switching hierarchy described in
Section 4.2.1 and are engineered to provide an average blocking of 0.01
during the busy season. Because calls blocked on a final group are not
permitted to wait,I6 it would appear reasonable to assume that the Erlang
B model is applicable to these groups. In actuality, however, there are
several effects not accounted for in the Erlang B model that affect its
applicability, so that modifications are needed to engineer final groups
properly. The next few sections discuss these modifications.
5.4.2 DAY-TO-DAY VARIATION

The grade-of-service (GOS) criterion for final groups pertains to average
blocking during the busy season, which typically comprises 20 hours of
data. Generally, the offered load varies between different busy hours in
the busy season. As Figure 5-7 shows, the Erlang B formula is such that
the blocking for the average of loads al and a2 is less than the average
blocking for the two loads. Similarly, blocking for the average busy season busy hour (ABSBH) load is less than the average blocking during the
busy hours of the busy season. This day-to-day variation effect was both
predicted and verified many years ago. Because of this effect, the Erlang
B formula, which underestimates the group size needed to average
0.01 blocking when day-to-day variation is anything but zero, was not
used to build trunk group tables. Instead, the tables were based on the
Poisson Blocking Formula, P(c, ii), where
p (c, ii) = e-ji

};

(18)

k=c

15 HistoricallY, large HU groups with the potential for significant overload were engineered
for 0.01 blocking, and for service protection, calls blocked on these groups were not
offered to other routes. These groups were called full groups or special final groups. But
engineering and overload control techniques have reduced the need for them.
16 Trunk holding times tend to be long because, unlike digit receivers, trunks have to be
held for the duration of the call. Allowing blocked calls to wait for an idle trunk would
involve long delays and inefficient use of system capacity. This is not the case for traffic
with short holding times, typically, certain data applications such as packet-switching
networks. (Section 5.8 discusses data traffic.)

/..
(!)

z

~

AVERAGE
BLOCKING, B

_______ _

...

--..

(,)

o...I

/'

III

/'
~

/'

/'

/'

/'

/'

ERLANG B BLOCKING
FOR AVERAGE
LOAD, B(e,a)

/' .......------t--

/'

a
LOAD

Fig ure 5-7. Day-to-day variation effect.

a

In this formula, c is the required number of servers, is the ABSBH load,
and k is an index of the number of arriving calls. P(c, a) is the tail of the
Poisson distribution, hence the formula's name.
Equation (18) assumes that delayed calls have a duration, including
the delay, that is the same as it would be if there were no delay. This
assumption is mathematically convenient, but it has no physical
justification in actual customer behavior. The practical value of the Poisson Blocking Formula in equation (18) is that, by chance, it produces
approximately the correct answer. It predicts a number of servers
significantly smaller than the Erlang C formula (calls wait forever) but
greater than the Erlang B formula (calls do not wait at all). Consequently, it serves as a suitable compromise between the two formulas.
Engineering tables that considered day-to-day variation were first
introduced in the mid-1960s, based on work by R. I. Wilkinson (1956).
More refined techniques were applied to provide later tables (see Hill
and Neal 1976). These later tables, known as the Neal-Wilkinson B tables,
are replacing Poisson tables in the operating companies.

5.4.3 REATTEMPTS
Customers whose attempts are blocked frequently try again. If the
number of blocked attempts is small, the reattempt effect will be
insignificant. The effect is potentially significant on trunk groups that
operate at very high occupancies (see Section 5.2.6). Because these groups
167

168

Network and Systems
Considerations

Part 2

have very little built-in slack, blocking on them increases dramatically
when an overload is present. Reattempts cause a departure from the random arrival assumption used in the traditional models and, if they are
not accounted for, cause an overestimate of offered load based on measurements. In practice, the accounting has been done by assuming that
only 35 percent of blocked calls are lost (do not return to the system).
The potential problem caused by reattempts is minimized because the
trunking tables use the models (originally, the Poisson, now, the NealWilkinson B) only for groups of fewer than 250 trunks. When the B
model calls for more than 250 trunks (that is, for the higher occupancy
groups), the group is engineered to the same (lower) occupancy that a
250-trunk group provides.
5.4.4 NONRANDOM LOAD

Both the Erlang B and the Poisson models assume that calls arrive in
accordance with the Poisson distribution. For final groups serving
overflow from HU groups, however, the Poisson arrival assumption is
demonstrably false. Poisson load has a variance-to-mean ratio equal to 1;
overflow load has a variance-to-mean ratio significantly greater than 1.
The variance-to-mean ratio of the offered load is called the peakedness and
is denoted by z. More trunks are required to provide an equivalent GOS
for peaked (z > 1) load than for Poisson (z=l) load.
The effect of peaked (non-Poisson offered) traffic is modeled using a
technique known as the equivalent random method. It assumes that all
peaked traffic will behave the same way as traffic overflowing a single
trunk group that meets the same assumptions as the Erlang B Loss Formula. The model is constructed by calculating a load, A, and a number of
trunks, C, that will have an overflow traffic with mean, a, and variance,
v, (v> a), equal to the traffic to be modeled. The desired overflow from a
group of trunks to serve the peaked traffic is expressed as overflow load, a.
For example, the objective may be I-percent blocking, or a/A = 0.01. If
A erlangs offered to C+c trunks produce a erlangs of overflow, then the a
erlangs of peaked traffic offered to c trunks will also produce a erlangs of
overflow.
Calculated values of C and A for a wide range of overflow loads and
peakedness are tabulated. Calculation of c, then, involves iteration of the
equation:
a
(19)
(-A) = B(C+c, A).
The disadvantage of the equivalent random method is that it presumes
that the group of c trunks is engineered using the Erlang B model. But,
as mentioned previously, engineering final groups using the Erlang B

Chap. 5

Traffic

169

model fails to account for day-to-day variations. N or can inserting the
Poisson blocking probability in equation (19) compensate for day-to-day
variations, since A and C are estimated by assuming blocked calls are
cleared on the C trunks.
The development of alternate-route networks, in which non-Poisson
offered traffic reaches the final groups, spurred the effort to produce
models that accurately account for peakedness. This effort, in turn, generated a need for models (for example, Neal-Wilkinson) that properly
account for both day-to-day variation effects and non-Poisson offered
traffic. The next section discusses the advantages of alternate-route networks and some factors complicating their design.

5.5 TRAFFIC NETWORK DESIGN
The previous two sections described the problems of engineering switching systems and engineering trunk groups as if they were independent
activities. However, when the nodes (switching systems) and the links
(trunks) are combined to form a network, optimizing overall network
design becomes the key problem. This optimization involves providing
satisfactory (from the customer's viewpoint) end-to-end service at the
lowest possible cost. Because the telephone network is so large and complex and must be flexible enough to respond to increasing demands, new
services, and changing technology, a precise formulation of the design
problem is impossible. However, alternate routing has proven to be a
key factor in optimizing the network. The next few sections describe one
of the simple models developed to produce cost-optimal (in the static
sense) alternate-route networks and how this model is used as part of a
process to provide the required number of trunks in the network-a process that captures dynamic (year-to-year) effects too complex to include
directly in a single optimization model. 17
5.5.1 THE ECONOMICS OF ALTERNATE ROUTING
Every central office (CO) in the PSTN is connected by a final trunk group
to some tandem 18 switching office, so that routing between COs is always
possible over these "backbone" final groups (see Section 4.2). Figure 5-8,
in which COs 1 and 2 have final groups to the same tandem office A,
illustrates this construction. Traffic between COs 1 and 2 can always be
17 Section 13.3.1 describes the process for providing trunks and special-services circuits in
the network (provisioning).
18 Tandem is used in this discussion in the generic or functional sense, that is, an office that
switches trunks to trunks. In the switching hierarchy, this office may be a toll office or a
local tandem office.

TANDEM OFFICE A

C01

C02
HIGH-USAGE GROUP

Figure 5-8. Alternate-route triangle.

routed on groups I-A and A-2 and would encounter (roughly) 0.02 blocking on the average during the busy season, assuming 0.01 blocking on
each final group.
Where enough traffic exists, there are significant advantages in also
building a direct group between COs 1 and 2 and allowing calls blocked
on the 1-2 group to overflow to the alternate route l-A-2. This routing
strategy effectively balances the conflicting forces of cost and service.
Trunks in the direct group, being shorter, tend to be cheaper than trunks
in the alternate route, which seems to suggest that all traffic between the
offices should be carried on a direct group. If this were done, however,
the group would have to be engineered for objective blocking, requiring
a larger number of trunks. In addition, there would be no service protection if a serious failure in the trunk group occurred. The alternate-route
network in Figure 5-8 provides better service at lower cost than either the
alternate-route-only or the direct-route-only strategies, provided the
correct number of trunks is determined for the direct group.19
The model for determining the correct number of HU trunks is based
on the simple triangle network of Figure 5-8. The following discussion
assumes that there is a cost per alternate-route trunk, CA (which includes
the cost of switching through tandem office A plus the transmission cost
of trunks I-A and A-2), and a cost per direct trunk, CD' The marginal
capacity, 'Y, of a trunk group engineered to 0.01 blocking can be defined as
the additional load that can be offered to the group, without changing
the blocking, when one trunk is added. (Mathematically,

aa
ac

'Y- -

l

=0.01

19 An exception occurs when the traffic between COs 1 and 2 is small. Because small trunk
groups are inefficient compared to larger trunk groups (that is, small groups operate at
low average occupancies), there are cases when no direct group can be justified
economically.

170

Chap. 5

171

Traffic

where a is the offered load and c is the number of trunks in the group.)
Since the I-A and A-2 groups are tarrying other traffic (group I-A carries
traffic from COl to other COs; group A-2 carries traffic from other COs to
C02), they tend to be reasonably large and efficient; that is, they have a
high marginal capacity. The following discussion assumes both groups
have the same marginal capacity, denoted by 'YA. (In practice, 'YA is usually between 26 and 28 CCS.)20
If n trunks are put into the COl-C02 group, the cost of these trunks
will be nCD . Moreover, the overflow from this n-trunk HU group, a(n),
will be offered to the alternate route, which will require the addition of
a(n )/1'A trunks to group I-A and to group A-2 to maintain 0.01 blocking.
The incrementaZ 21 cost, then, of putting n trunks in the COl-C02 HU
group, denoted by C(n), is given by
C(n)

=

nCD

+

a(n) CA.
'YA

(20)

Figure 5-9 shows C(n) and its two components in graph form, treating
n as a continuous variable. As shown, C(n) has a unique minimum,
which occurs at the value of n such that
-

aa(n)
an

---="'A
I

CD

'YA

-=CA
CR '

(21)

TOTAL COST, C

(n)

DIRECT·ROUTE COST
~

en
o
u

n =OPTIMAL HU
NUMBER OF HU TRUNKS,

GROUP SIZE

n

Figure 5-9. Cost function for alternate routing.

20

Hundred call seconds per hour (see footnote 3).

21 The background traffic on trunk groups I-A and A-2 requires trunks too, of course, but
the cost of those trunks is assumed to be independent of n. Thus, only the additional cost
of routing the COl-C02 overflow via l-A-2 must be considered.

172

Network and Systems
Considerations

Part 2

where CR = CAIC D . CR is referred to as the cost ratio of the HU group.
(If there is no n such that equation (21) holds, then the minimum occurs
at n =0; that is, tandem route only is the optimal strategy.)
The quantity -aa(n )Ian has a practical interpretation: It is the load on
the last trunk of an n-trunk group, assuming the Erlang B model applies
to the HU group. Thus, the quantity "'tAICR is the most economical load
on the last trunk, and the optimal group size is such that the last trunk
carries "'tAIC R load. Since the telephone companies measure load in CCS,
the quantity "'tAICR is referred to as the economic CCS (ECCS). The technique of sizing HU groups such that the last-trunk load is l'A IC R is commonl y called ECCS engineering.
5.5.2 MODIFICATIONS TO THE ECCS ENGINEERING TECHNIQUE
The cost function in equation (20) represents an approximation to the
actual incremental cost of providing n trunks in the HU group. The marginal capacity, for example, is not truly a constant but depends on both
the peakedness (variance-to-mean ratio) of the background traffic on the
alternate route and the peakedness of the overflow from the HU group.
Nevertheless, the ECCS technique yields a good approximation to the
optimum HU group size. This approximation can be improved by recognizing certain cost realities that the ECCS model ignores. These realities
are discussed in the following paragraphs.
Minimum Group Sizes
There is an administrative cost associated with an HU group that is not
reflected in equation (20). This cost applies when n >0 but not when
n =0. When this administrative cost is added to the total cost for the
optimal group size as determined by the ECCS method, it may well be
that n =0 is a cheaper solution. Empirical studies have determined that
this happens when the ECCS method suggests small HU groups and can
be corrected effectively by using the EeCS method in conjunction with a
minimum HU group size. The recommended minimum, determined
empirically, is typically three trunks per group for local networks and six
trunks per group for toll networks; smaller HU groups are not built. The
difference between the two minimum values is explained by the fact that
cost ratios (CAIC D ) in local networks tend to be higher than in toll networks, so that smaller groups are economical.
Modular Engineering
The cost function of equation (20) assumes that there is a unit cost, CD'
for each trunk added to the direct group. However, many of the
transmission components that implement trunk groups have the capacity
to handle multiple trunk circuits. For example, trunks can be added to a

Chap. 5

Traffic

173

digital interface frame (see Section 9.4.3) in increments of twenty-four.
To compensate for these effects, a technique known as modular engineering
is used.
With modular engineering, the optimal HU group size as determined
by EeeS engineering is rounded to the nearest relevant module size. For
the Long Lines network, the relevant module size is twelve trunks
because the basic multiplex unit (the channel group) implements twelve
circuits at a time. 22 For local T-carrier (see Section 9.4.2) networks using
dedicated digital terminations, the relevant module size is twenty-four
trunks for 2-way groups and twelve trunks for I-way groups. (Using
twelve trunks for I-way groups maximizes the chances that the trunk
requirements of two opposing I-way groups will add up to a multiple of
twenty-four circuits.)

N oncoincidence of Busy Hours
The triangular cost model illustrated in Figure 5-9 assumes that the busy
hour of the direct group coincides with the busy hour of the two
alternate-route groups. This need not be true. For instance, the busy
hours of the two alternate-route groups need not be the same. In that
case, overflow from the direct HU group might force additional trunks to
be added to group I-A but not to group A-2 in the group I-A busy hour.
Similarly, trunks might have to be added to group A-2 but not to group
I-A for overflow during the group A-2 busy hour.
One method of compensating for possible noncoincidence effects is
the cluster busy hour technique, shown in Figure 5-10, in which a cluster
of potential HU groups overflows to a common final group. The offered
loads to each group are summed for each busy hour, and the maximum
hourly sum identifies the cluster busy hour. Each HU group is then
sized, using the offered load for that group that corresponds to the cluster
busy hour. This technique works well in many applications, but it
ignores the possibility that the alternate-route busy hour may, in fact,
depend on the HU group sizes selected. When this happens, no single
hour can be identified a priori as the "correct" hour for all HU groups. A
technique known as multihour engineering (see Elsner 1977) has been
developed to solve the problem generally, but the algorithm is complex
and must be implemented by means of a highly sophisticated computer
program.
5.5.3 DYNAMICS OF NETWORK DESIGN

The EeeS cost model used to size HU groups determines an optimum
group size only in the static sense, that is, for a given load or collection
of hourly loads and a given alternate-route cost ratio. As trunk group
22 Sections 6.5, 9.3.5, and 9.4.3 discuss multiplexing and channel groups.

TANDEM

TANDEM

OFFICE

OFFICE

COS
C01

C04

Figure 5-10. Originating cluster concept.

requirements change from year to year, other costs are incurred that are
not included in the ECCS model but that can strongly influence economic
decisions.
The impact of trunk requirements on the facilities network layout (see
Section 4.4) is particularly important. For the facility-provisioning process to position correctly the facility routes needed to implement trunk
groups, estimates of trunk requirements are needed years in advance.
The ECCS model is used to forecast trunk requirements using projections
of offered loads and routing configurations.
Forecasting errors are to be expected and must be corrected to minimize total network cost while providing the grade of service required.
The network design process therefore involves three distinct but interrelated phases: forecasting, engineering, and servicing. The following
paragraphs discuss some of the traffic considerations involved in these
operations. Activities involved in each operation are described in Section
13.3.

Forecasting
The key to accurate trunk forecasting is accurate projection of loads
offered to the trunk groups. Load projection, in turn, requires a measurement system to obtain basic load data and an extrapolation process that
translates basic data into expected loads.
174

Chap. 5

Traffic

175

There are two sources of basic load data: automatic message accounting (AMA) tapes and trunk group measurements. AMA tapes record
information related to telephone calls for billing purposes (see Section
10.5). The Centralized Message Data System (CMOS) analyzes the tapes
to determine traffic patterns, but the process is expensive. So, CMOS
samples only 5 percent of all completed, or billable, calls (messages) to
generate a point-to-point load base. Although the resultant estimate of
total load is based on a small sample of calls that excludes incomplete
calls, it is, nevertheless, useful because it is relatively independent of the
network switching configuration. This is particularly important if it is
known that the switching configuration will be changing during the
period of the trunk forecast.
Trunk group measurements are more reliable than point-to-point estimates and are usually less expensive to obtain. But such measurements
are only possible, obviously, on groups that exist. They become less useful where network configuration changes occur and new trunk groups
come into existence.
Engineering
The network trunk requirements produced by the forecasting process are
sensitive to errors in load projection. Such errors can generate significant
and frequently unnecessary year-to-year "churning" (fluctuation) of trunk
group requirements, the cost of which can be high. Furthermore, the
decision regarding the number of trunks to place in a given group should
depend on both the number of trunks already in place and the expected
future of the group. For example, disconnecting trunks is expensive, but
so is maintaining more working circuits than are required to meet service
objectives. An optimal disconnect policy, therefore, depends on whether
the reduced requirements of a given group are expected to return to their
original size in the near future. If so, it is usually more economical to
leave the unneeded circuit(s) in place rather than pay a disconnect cost
now and a reconnect cost later.
Another important consideration in engineering is proper reserve
capacity. If final trunk groups, for example, were always engineered to
provide exactly 0.01 blocking, then, on the average, half of the final
groups would actually generate greater than 0.01 blocking due to forecasting errors. Trunk servicers (discussed in Servicing below) would
have to augment (add trunks to) these underprovided groups to correct
for the errors, and this augmentation is expensive.
An optimal reserve capacity· balances the cost of overproviding against
the cost of servicing. Too much reserve is wasteful; too little can lead to
severe and expensive servicing problems. It thus becomes economical to
overprovide final groups, since the cost of some additional (reserve)
trunks is offset by the savings in labor cost during servicing.

Network and Systems
Considerations

176

Part 2

Using the trunk forecast as input, the engineering process attempts to
smooth the forecast to eliminate churning due to statistical error and to
reflect disconnect and reserve economics. Historically a manual process,
engineering is now being automated to provide more accurate, stable, and
economic (in the total cost sense) trunk requirements to the facility
planners and engineers.
Servicing
In trunk servicing, performance measurements of final trunk groups are
used to determine where circuits have to be added or deleted to meet service objectives. This process is supported by the computer-based Trunk
Servicing System (TSS). TSS makes the servicing of HU groups as well as
finals possible since it analyzes overall network performance and can
determine when, for example, the overflow from a poorly engineered HU
group is responsible for the poor performance of a given final group.

5.6 NETWORK MANAGEMENT
Telephone networks employing alternate routing and common-control
switching systems (see Chapter 7) provide efficient use of facilities. This
efficiency, however, can lead to operating penalties when heavy traffic
overloads or major equipment failures occur. The solid curved line in
Figure 5-11 illustrates this phenomenon. The figure shows simulated
results for a network engineered to carry 1600 erlangs but subjected to a
substantial overload. Without network management controls, as the

2000 r - - - - - - - - - - r - - - - - - - - - - - - - - ,
NO SWITCHING DELAYS - " ! - -_ _ _

iii'

1600 ~--------~F----~~---~
ENGINEERED LEVEL

ystems
Considerations

178

Part 2

and experimentation with poor service conditions violates the operating
companies' dedication to providing the best possible service.
Computer simulation techniques have proved to be the best tools for
studying network congestion and evaluating network management techniques. Simulation programs can take many forms. The program used to
construct Figure 5-11 is a call-by-call simulator; that is, each call entering
the simulated network is followed through the network. This simulation
shows that, without controls, when the offered load exceeds the
engineered load (1600 erlangs) by 50 percent or more (2400 erlangs), the
carried load falls below the engineered level. Moreover, as the dashed
line indicates, increasing switching capacity and, thereby, eliminating
queue delays would significantly improve the overload characteristics.
The results of another run on this simulator are shown in Figure 5-12
and illustrate several principles. This figure shows the time response of
the Figure 5-11 network when a 100-percent overload (3200 erlangs) is
offered and no controls are used. The vertical scales show both carried
load in erlangs and ineffective (blocked) attempts. The carried load is initially about 1800 erlangs, this being the level at which the network was
operating when the overload was introduced. As the congestion builds,
the number of attempts blocked by trunk congestion increases rapidly.
Ineffective attempts due to timeouts (switching congestion), on the other

12,000

J

2000

/
/

--1\

10,000

/

CARRIED LOAD
Q

I

w
~

()

8000

I

0

....I

lEI

...en
2i
w
......
~

;,'

TRUNK
;'
BLOCKING / '

;'

6000

\

I

/

Ij

o

I

/

2000

/

~

20

30

"
50

,,

w

«
()

1000

"

Q

II:

\

\

o
....I
iE

\

I

40

1400

1200

1',

1----- --- 1 - - - -

10

~

I

I
I

SENDER
TIMEOUTS

II:
Q

1\,

/
I

/

1600

u;
~
z
«
....I
«

\1

./

4000

/

"\ ~1

;'./

«

1800

I

/

60

TIME (MINUTES)

Figure 5-12. Transient network congestion-100-percent overload.

Chap. 5

Traffic

179

hand, are initially small. As reattempts due to trunk blocking combine
with the heavy overload of initial attempts, the switching system
becomes congested. Timeouts increase rapidly after 40 minutes and carried load falls off accordingly. As a result of switching congestion, the
trunk congestion disappears, since so many short-holding-time calls
(namely, digit transmitter timeouts) are using the trunks.
If special network management measures were not taken, the poor
performance illustrated in Figure 5-12 might occur several times a year.
Christmas Day, Mother's Day, and, occasionally, other holidays, for example, produce intertoll trunk congestion. While the total message volume
on these days is well below that of the average business day, the calling
pattern is different. Local and intrastate toll traffic is light, but long-haul
interstate traffic far exceeds engineered capacity levels. Such skewed patterns of overload also occur in metropolitan areas during snowstorms, for
example, when calls from the suburbs to the center city exceed normal
levels.
Another type of overload that can cause service deterioration is the
focused overload-heavy calling from many points to one point. Telephone calling after the Mount Saint Helens eruption in May 1980 is a
good example. For a few days following the eruption, call attempts from
the rest of the country to the Pacific Northwest Bell area were three to
five times higher than they had been on Mother's Day the week before,
typically the company's peak day. Similar focused overloads may occur
in metropolitan areas when, for example, a radio or television station
encourages mass calling, or when, during a power failure, a power company becomes the target for large call volumes.

5.6.4 DETECTION OF CONGESTION
Network congestion is detected and often predicted by interpretation of
appropriate traffic data. As shown in Figure 5-12, overloads often begin
with trunk congestion. Signals indicating all-busy status for trunk
groups are provided in Network Management Centers. In more sophisticated systems, data are used to calculate attempts per circuit per hour (ACH)
and connections23 per circuit per hour (CCH) for trunk groups as frequently
as every 5 minutes. ACH and CCH patterns can be revealing. For example, a higher than normal ACH coupled with a normal CCH implies that
demand is heavy, but calls are being completed. If CCH becomes higher
than normal, however, trunk holding times are short. This implies that
ineffective attempts are being switched, and some control may be
required.

23 Connection implies only that a call attempt has been switched through a switching system
to an outgoing circuit. It does not imply that the call has reached'the called party.

180

Network and Systems
Considerations

Part 2

The most effective indicators of switching congestion are dial-tone
delay (for central offices) and incoming digit-receiver delay (for toll and
local tandem offices). If excessive dial-tone delay is the result of heavy
originating traffic, the network manager can do little. If, however, dialtone delay at an end office is the result of heavy traffic terminating at that
office, as is possible with electronic switching systems (see Section 5.3.1),
controls can relieve terminating congestion. When digit-receiver delay
becomes excessive, timeouts become a problem (see Figure 5-12). Controls can maintain delays at reasonable levels.

5.6.5 CONTROL OF CONGESTION
Network management controls may be restrictive or expansive. Restrictive
controls eliminate or reduce routing alternatives to prevent traffic from
reaching congested areas. Expansive controls permit calls to be completed using additional nonstandard routing to bypass congested trunk
groups and switching systems.
Table 5-1 gives some examples of the various controls and a brief
description of the function of each. As the table indicates, controls such
as directionalization and code blocking are particularly effective during
focused overloads. Without such controls during the Mount Saint Helens
eruption, for example, incoming traffic would have made it impossible to
complete emergency calls from the affected area.
For more general overloads, such as those that occur on Mother's Day
and Christmas Day, changing normal routing patterns can be an effective
technique. If, for example, the eastern seaboard is experiencing congestion at ten o'clock on Christmas morning, traffic between Eastern timezone offices can be routed via Central, Mountain, or even Pacific timezone offices. Because of time-zone differences, circuits to those areas and
the offices themselves probably have unused capacity at that time. Under
normal circumstances, such circuitous routing would be inefficient, but
network managers can temporarily redefine alternate-route sequences
when the need arises.

5.7 TRAFFIC MEASUREMENTS AND SERVICE
OBJECTIVES
The preceding discussion frequently referred to the various service objectives that govern the engineering of equipment. The existence of such
service objectives presumes that measurements will be taken to determine
whether the network is meeting them. The interplay between traffic
measurements and service criteria is important: There is no point in
establishing service objectives that cannot be adequately verified by the
measurement process. In fact, the service objectives must be such that

TABLE 5-1
NETWORK MANAGEMENT CONTROLS
Control

Type

Function

Directionalization
of Trunk Groups
by Making Trunks
Appear Busy*

Restrictive When operated at one switching system, reduces call pressure to a distant switching system. For a 2-way trunk
group, additional equivalent I-way outward trunks for the
distant switching system are provided. Commonly used
in focused overloads to hold traffic away from the focal
point.

Cancellation of
Alternate Routing*

Restrictive Removes alternate-routed traffic from a group and, hence,
reduces the load on the distant switching systems. Since
alternate routing is reduced, the average number of links
per call is also reduced. There are two forms: CANCELFROM (selectively cancels traffic overflowing from a
high-usage group to another high-usage group or final
group); CANCEL-TO (cancels all traffic overflowing to the
controlled group). Usually used in peak-day overloads
(such as Christmas Day) to increase network efficiency by
reducing links per call and to control the amount of
alternate-routed traffic that reaches the upper levels of the
hierarchy.

Rerouting*

Expansive

Code Blocking*

Restrictive Blocks calls (routes to reorder tone or to recorded message) according to destination code. Useful during
focused overloads, especially if calls can be blocked at or
near origination. Blocking need not be total unless the
destination office is completely disabled through natural
disaster or equipment failure. Switching systems
equipped with code-blocking features typically can control 50 percent, 75 percent, or 100 percent of calls to a particular code. The controlled code may be as broad as the
destination numbering plan area or as restricted as the
office line.

Dynamic Overload
Control (DOC)

Restrictive An automatic control system that senses congestion (as
measured by the number of calls waiting for a sender) at a
toll or local tandem office and that sends an electrical signal to subtending offices (those that home on it).
Responses of the subtending switching systems depend
on how they have been programmed or wired. Commonly, subtending offices cancel alternate-routed traffic or
make trunks busy. When DOC is properly deployed,
about 40 percent of the sub tending traffic will be controlled, and a control action may last only about 10
seconds. DOC controls are preferred over manual controls because they are activated as soon as congestion
appears and removed as soon as the need disappears.

Selective Dynamic
Overload Controls
(SDOC)

Restrictive

Routes overflow traffic to a trunk group not in the normal
route advance sequence. Usually used when all normal
routes are busy.

DOC does not discriminate between traffic with a high
completion probability and that with a low completion
probability. SDOC combines DOC and code blocking to
form a system in which the sub tending offices either calculate or receive completion statistics by code, and, if the
control office encounters congestion, the subtending
offices respond to a DOC signal by canceling alternate
routing for hard-to-reach codes or by making trunks
appear busy for these codes.

* These controls require manual activity at a key or switch and sometimes involve wiring
changes. Hence, they may be applied too slowly and are often left in effect too long.

182

Network and Systems
Considerations

Part 2

measurements can be developed to pinpoint the corrective action that
must be taken when objectives are not met.
Intuitively, it is expected that service objectives are set first, then the
network is engineered and equipped, and then measurements are taken
to evaluate performance. Frequently, with switching equipment in particular, this natural progression is followed. For the traffic network as a
whole, however, this progression is not feasible. Limitations inherent in
the measurement process can influence both the setting of service criteria
and the design of the traffic network itself. The next section discusses
some traffic measurement procedures, after which it is possible to explain
the motivation behind the apparently arbitrary decision to engineer every
final group to 0.01 blocking rather than to design the network to some
end-to-end service objective.

5.7.1 TRAFFIC MEASUREMENT PROCEDURES
A key measurement that must be taken for any group of servers is offered
load. Not only are load estimates used as the basis for projecting future
demands, but when a given service objective has not been met, it is
important to know what the load was so that the proper number of additional servers needed can be calculated.
One method of measuring load on a trunk group requires a traffic
usage recorder, which scans the trunks every 100 seconds and counts the
number of busy trunks. The sum of the counts during the measurement
interval, usually an hour, gives a direct estimate in hundred call seconds
(CCS) per hour of the usage of the trunk group.24 This usage measurement, however, estimates carried load, L. To convert the estimate, L, to
one of offered load, it is necessary to use equation (22),

L = a[1- B (c ,el) J'

(22)

iteratively to solve for a, the estimate of offered load. This procedure,
though, requires an assumption about the peakedness of the offered load.
Further, the measurement includes usage on trunks made busy (taken out
of service) because of maintenance activity.
Another method of estimating trunk group offered load uses peg
count and overflow (PCO) data. Peg count refers to the number of
attempts offered to the group during the measurement interval; overflow
refers to the number of attempts encountering all trunks busy. The ratio
of overflow to peg count directly estimates blocking}5 from which offered
24 There is, of course, statistical error because of the discrete scanning.
2S PCO measurements do not distinguish initial attempts from reattempts; hence, the ratio
is only an estimate of first-attempt blocking.

Chap. 5

Traffic

183

load can be deduced, assuming Erlang B behavior and an estimate of
offered load peakedness. The combination of both usage and pea data
gives the best estimates of offered load and grade of service, since
peakedness assumptions are not required when both data sources are
available.
Similar measurements, usage and peg count, are taken for switching
equipment such as incoming digit receivers. Usage data for switching
items, however, are typically based on 10-second scan rates, since the
holding times of these items are shorter than trunk holding times. For
those items that must meet delay criteria, it is also necessary to record
delay data.
Older common-control equipment, such as crossbar systems (see Section 10.2), require expensive external measuring devices to record data.
For electronic switching systems (see Section 10.3), traffic measurements
can be taken directly by a stored program. Such data are more accurate
and complete (different types of telephone calls are easily distinguishable,
for example) but require processor real time. Therefore, the amount of
data taken must be limited so that the system's capacity for call processing is not restricted.
However measurements are taken, there are errors attributable to
discrete sampling, maintenance activities, hardware failures, and incomplete knowledge of the underlying traffic parameters (peakedness and
reattempts, for example). Also, the more data taken to minimize errors,
the more expensive the process of analyzing the data and interpreting the
results. The current trend is toward mechanizing data collection and
using computer programs for analysis and most of the interpretation.
5.7.2 SERVICE OBJECTIVES

Some service criteria used in the Bell System, such as dial-tone delay and
incoming digit-receiver delay, are self-explanatory, deriving directly from
the server's function, customer expectation, and the potential effect of
poor service on the network. However, setting proper service criteria for
the traffic network as a whole is not so simple. Ultimately, the customer
is most concerned with the probability that a telephone call will reach its
destination, independent of where along the way it might be delayed or
blocked. If service objectives were set to govern that probability, a reliable measurement process to estimate the total number of initial attempts
between every pair of end offices and the number of those attempts that
failed in the network would be needed. Given that data and some endto-end service objective, the question would be: What to do if the objective were not met? Failed attempts could result from a problem in one or
more of many trunk groups from which an end-to-end path might be
formed and/or from a problem in one of several switching systems. The
specific location(s) and cause(s) of the problem would have to be determined. The amount of data that would have to be collected, coordinated,

184

Network and Systems
Considerations

Part 2

and analyzed is staggering, as is the problem of deciding upon corrective
actions.
From this perspective, the effectiveness of a hierarchical network with
its backbone finals engineered to 0.01 blocking becomes clear. The worst
service that a call can receive during the average busy hour is roughly
0.09 blocking (there are, at most, nine finals in tandem in any route).
Because of the noncoincident busy hours of the network groups and the
economical placement of high-usage groups to provide alternate routing,
the blocking seen by the customers is substantially less. Empirical studies
show that end-to-end blocking averages about 0.02, and almost all the
traffic experiences blocking of 0.05 or less under normal busy season
conditions.
Equally crucial, the problems of measuring performance and then taking corrective actions are simplified as much as possible. For example,
attention is focused on individual network components, rather than on
overall network activity with its many possible interactions. If, for example, a given final group is not giving 0.01 blocking, that fact can be
determined relatively easily; and solutions are immediately known. The
hierarchical design with probability-engineered final groups thus greatly
simplifies the problems of measuring and servicing the network and, in
so doing, significantly reduces the cost of administering and maintaining
that network.

5.8 TRAFFIC CONSIDERATIONS IN DATA NETWORKS
Data communications are growing rapidly both in terms of traffic volume
and the diversity of data applications. The Bell System initially met the
growing need by developing equipment that allows data to be transmit"'-.. ted over the PSTN, private switched networks, and private-line multipoint networks. Later developments led to increased PSTN capabilities
and other networks specifically designed to carry data traffic and support
data applications. 26 The following sections describe characteristics of data
traffic, performance parameters, and some concepts peculiar to engineering data networks.
5.8.1 THE NATURE OF DATA TRAFFIC
Section 5.2.1 defined the traffic on a telephone network in terms of average call arrival rate and average call holding time. These parameters are
also applicable to data networks, although the models that describe them

26 Section 11.6 describes some of the network capabilities the Bell System provides to
accommodate the many different types of data traffic.

Chap. 5

Traffic

185

are different. A simple example of an established data call is a user
logged on to a time-sharing computer. The call rate, as in telephone networks, is the number of call ("logon") attempts per unit time, and the call
holding time is the duration from call establishment (logon) to call disconnect ("logoff"). Call holding time, of course, includes times when no data
are being transmitted, just as a telephone call holding time includes
pauses in conversation.
Call rate and call holding time can vary significantly from one data
application to another. An airline reservation desk application, for
instance, may require only one call per day-a connection is established
at the beginning of the day and is maintained until the end of the day.
A reservation clerk may send or receive data many times during that one
call. Alternatively, in a time-sharing application, a data call might have
to be made each time the application is run.
When data calls are transmitted over networks sensitive to the quantity of data transmitted per unit time, two additional parameters are
useful-transaction rate and transaction length. Each time data is
transferred to or from a customer, a transaction occurs. There may be
many data transactions during a data call (as in the airline reservation
desk application) or only one to and from a computer (as in the timesharing application). Each transaction may itself contain variable quantities of data. The quantity of data is called the transaction length and may
be described in terms of bits, bytes or characters, or packets.
Packet-switching networks,27 in particular, are sensitive to the quantity
of data in a call and take advantage of the "burstiness" (that is, the data is
sent and received in bursts of high intensity interspersed among periods
of little or no activity) of that data traffic. In packet switching, data are
divided into packets, each of specific format (including destination and
control information) and of maximum length. The transmission path is
occupied by a particular packet for a time that is dependent on the packet
length, and the path is then available for use by other packets that may
be transferring data between different data terminals. This contrasts with
circuit-switching data networks where a physical end-to-end transmission
path is established at the start of the call and is dedicated to the call for
its duration. The end-to-end transmission path in packet-switching networks is called a virtual circuit. As they traverse this virtual circuit, packets are switched on a per-packet basis. from one network node (packet
switch) to the next, using the specific control information they carry.
Each type of data network is sensitive to call rate, call holding time,
transaction rate, and transaction length in a different way. Figure 5-13
depicts the relationship between the four data traffic parameters.

27 Sections 2.5.4 and 11.6.2 discuss packet-switching network services and systems.

CALL INTERVAL
LOG
ON

LOG
OFF

CALL HOLDING TIME

-I

I'

•

........
,........A-,.

.
,. +

AVERAGE CALL RATE =
1/AVERAGE CALL INTERVAL

t'

I'

I

LOG
ON

,.

CALLS

TRANSACTION LENGTHS

+

~

~

tl ...--....

~

~

TRANSACTIONS

TIME

t

t

t

TRANSACTIONS BEGIN

t

t

Figure 5-13. Data traffic parameters.

5.8.2 DATA TRAFFIC MODELS
Telephone networks are dominated by voice traffic, facilitating the
modeling of call rate and holding time. Modeling traffic on data networks is more difficult. No general models for data traffic exist because
of the diversity of data applications and the short experience in measuring data traffic compared to many years of measuring voice traffic. The
data source for a data network may be another network or a customerprovided node that concentrates its own end-users. In these cases, the
individual data calls would not necessarily be explicitly observable. Call
rate and holding time are, thus, modeled only after the particular data
application is understood as well as possible.
Data transaction rate and transaction length, however, lend themselves to modeling by the telephone traffic call rate and holding time
models, respectively. Transactions are generally expected to arrive
according to a Poisson process. Thus, the probability that k transactions
arrive in any time interval, t, is given by equation (3), where A is now the

average transaction arrival rate:
(3)

Typical transaction arrival rates are shown in Table 5-2. The rate is
assumed to apply to both directions in 2-way data applications (for example, inquiry-response applications) and for one direction in I-way data
applications (for example, batch applications).

186

187

Traffic

Chap. 5

TABLE 5-2
SAMPLE DATA TRAFFIC CHARACTERISTICS
Application
Time Sharing

Call rate (average
number of calls per
busy hour)

0.50

Inquiry-Response

Message

0.25

3.33

Batch

0.15

Call holding time
(average number
of minutes)

12

16

3

11

Average number of
transactions per call

10

10

1

1

3.33

0.15

Transaction rate
(average number
per busy hour)
Transaction length
(average number
of characters per
transaction)
In
Out

5.00

35
200

2.50

120
200

600

40,000

Assuming the transaction has arrived, data transaction length in characters is treated as a geometric probability function (see Fuchs and Jackson 1970). As in the assumptions for equation (4), it is assumed that the
probability of a call ending in a small interval, t, is a constant, /-Lt, and
that the probability that the character last received is the last one in the
transaction is a constant, 1/c. Under these assumptions, the probability of
a transaction having m characters is:
1
1 m-l
P(m) = -(1--)
c
c

m = 1, 2, 3, ... ,

(23)

where c is the average number of characters in a transaction.
For packet-switching networks, the data transaction length is stated in
terms of the number of packets, since packet-switching systems are more

188

Network and Systems
Considerations

Part 2

sensitive to the number of packets than to the number of characters.
Assuming a geometric distribution of transaction lengths, if the characters
of a transaction are loaded into packets D at a time, except for the last
packet, which may have any number from I to D, the distribution of
packets per transaction will also be geometric. The probability, given a
packet contains at least one character, that this packet will be the last one
in the transaction is one minus the probability that D or more characters
remain in the transaction. From equation (23), this probability is
(1-IIc)D. Therefore, the probability that any packet, once stated, will be
the last one is constant-the condition for the geometric probability function. The number of packets per transaction is, therefore, a geometric
distribution with mean N:
N =

1

-----.;:~

1

D·

(24)

1-(1--)
c

The data traffic models are useful as inputs to other models of the data
network and network components. They can be used directly for
engineering access lines and switch access-line ports and as direct inputs,
for example, to load-service models of access lines and switch access-line
ports. Traffic loads at other data network components (for example, at
interswitch trunks or switch trunk ports) can be derived from the data
traffic models and models of the network components through which the
data traffic flows.
There is no universal flow model applicable to all networks, since the
model depends on the network and packet-switching system architecture,
method of operation, and traffic-handling capabilities. The models of
data traffic and related models for data network components can be used
to design, for example, new data network control strategies and to specify
data network engineering rules. The engineering process includes adjustments to network and component sizing that are based on data network
measurements.

5.8.3 DATA PERFORMANCE CONCERNS
Telephone traffic performance concerns, such as call blocking, are also
relevant to data traffic. In addition, data accuracy, throughput, and network delay are important in data network performance. Data accuracy
and throughput are concerns in both circuit switching and packet switching, whereas network delay is usually a concern in packet switching. 28
28 The recommendation of categories and levels of performance for packet-switching
networks is a current activity of the Comite Consultatif International Telegraphique et
Telephonique (CCITT) 1980-1984 plenary period. Performance categories and levels are
reasonably well understood at this time, although the 1980 recommendations treat only a
few performance categories.

Chap. 5

Traffic

189

Accuracy is only indirectly related to traffic engineering through objectives such as lost packet rate. Throughput and network delay, allocated
on a network component basis, are closely coupled to traffic engineering.

Data Accuracy
Data accuracy objectives are typically bit- or character-related or are
reflected in a specified number of error-free seconds as in the Digital Data
System (see Section 11.6.1). On packet-switching networks, accuracy performance objectives also include errored, misdelivered, lost, duplicated,
and out-of-sequence packet rates. Typically, rates for each of these are on
the order of 1 in 109 to 1 in 107 under normal service conditions.

Throughput

Throughput is the quantity of data a user can transfer during a time interval. On packet-switching networks, the end-to-end transmission path
(hence, channel capacity) is not dedicated. Contention for the channel
capacity occurs when multiple users attempt data transmission. Thus,
throughput and the related performance objectives are functions of the
network parameters that control the sharing of the channel capacity.
Once a call is established on circuit-switching data networks, the end-toend transmission path, or channel capacity,29 is dedicated to the data call.
In both types of switching networks, the achievement of throughput
objectives also depends on the round-trip network delay.

Network Delay
Network delay is usually measured from the last character input to the
network to the first character received from the network. In both
circuit-switching and packet-switching networks, network delay is a function of data transmission speed, propagation time, and distance through
the network. In packet-switching networks, network delay is also a function of packet-switch processing time and queuing delay, both of which
can be influenced by traffic engineering. Other factors related to network
delay either are under customer control (for example, transmitting at
different data rates) or are determined by network design (for example,
limiting end-to-end transmission paths to a small number of packet
switches).

29 Bandwidth for analog channels, bit rate for digital channels. Section 6.2 describes analog
and digital channels.

190

Network and Systems
Considerations

Part 2

Network delays are usually divided into call-related times (for example, call establishment) and data transfer times. 3o Typical packet callrelated times are approximately 500 milliseconds. Packet data transfer
times typically range from 150 to 500 milliseconds.

5.8.4 TRAFFIC-ENGINEERING CONCEPTS
The quantity of data traffic carried by the PSTN is small compared to the
voice telephone traffic it carries. Therefore, little or no explicit planning
is done for data traffic. If it is known that a significant quantity of data
traffic will be present (for example, when a large data user moves into a
new area), planning may include accommodating the data traffic on the
PSTN, engineering special data networks, or both. This section addresses
traffic-engineering concepts for data networks. Moreover, since engineering circuit-switching data networks is similar to engineering telephone
networks, only concepts related to packet-switching data networks are
described.
Typically, data networks are engineered to support some peak busy
hour condition. Extreme value engineering (see Barnes 1976) is sometimes
used. Another method models the highest traffic hour of the busiest day
considered by excluding some number of busiest days of the year and
then multiplying average busy hour traffic by a factor from 1.2 to 1.5.
Estimates of peak busy hour traffic and performance are used to
engineer data network transmission facilities and packet switches. The
data traffic models in Section 5.8.2 are a basis for the engineering rules
and models. For example, peak busy hour traffic and performance estimates are applied to the rules and/or models to predict time of exhaust
for packet switches and other network components and to determine
appropriately sequenced network growth steps.
Network access lines are engineered using the total traffic entering the
network at the access points. Network trunks and switches are
engineered using the total traffic entering the network at the access
points and the traffic flow patterns. Network access lines and trunks are
engineered to a utilization level, typically from 50 to 70 percent in
packet-switching networks, to achieve allocated network performance
objectives.
Packet-switch engineering involves determining the appropriate
switch configuration so that the allocated network performance can be
achieved. Typical resource limits on state-of-the-art packet switches
include maximum number of ports, throughput of each port, switch realtime processing, and port-and-switch memory. Ideally, all of these
30 Data transfer time is the network delay for switching data and does not include, for
example, routing and address translation time as does call establishment.

Chap. 5

Traffic

191

resources should exhaust simultaneously when switching system traffic
reaches a certain level. This rarely happens, though, and some limits are
never reached. Typically, packet-switch real-time processing is most critical and is used as the engineering load-versus-service (for example,
cross-switch delay) relationship.

AUTHORS
N. Farber
R. A. Farel
W. S. Hayward
V. S. Mummert
S. H. Richman

6
Transmission

Chapters 3 and 4 introduced the network elements (for example, switching systems and transmission facilities) involved in transmitting information over the telecommunications network. This chapter introduces some
of the fundamental transmission concepts and principles underlying the
way those physical facilities carry information. Sections 6.1 through 6.3
describe types of signals, the concept of a channel, and important
transmission media, respectively. Sections 6.4 and 6.5 discuss two
processes used to prepare signals for transmission (modulation and multiplexing) and some of the techniques used in the Bell System. The last
section describes some impairments that degrade the performance of
transmission facilities, how the impairments are controlled to meet performance objectives, and how transmission performance objectives are
allocated to the network components.

6.1 SIGNAL TYPES AND CHARACTERISTICS
The telecommunications network can transmit a variety of information,
which originates in either of two basic forms: analog signals and digital
signals. Analog signals are in the form of a continuous, varying physical
quantity (for example, voltage) that reflects the variations of the signal
source in time (for example, changes in the amplitude and pitch of the
human voice during speech). Speech and video signals are examples of
analog signals. Digital signals, on the other hand, are discrete and have
well-defined states at any point in time. Telegraph signals and pulses
transmitted from a dial-pulse telephone are examples of digital signals.
An increasingly common digital signal is that originated by computers
and data terminals. It is important to note that signals may originate in
one form, but may be converted to the other form for transmission over
the network (see Sections 6.4 and 6.5). The following sections describe

193

194

Network and Systems
Considerations

Part 2

three types of analog signals (speech, program, and video) and one type
of digital signal (data).

6.1.1 SPEECH SIGNALS
The speech signal is the most common signal transmitted over Bell System facilities. The transmitter in a telephone set transforms sound waves
(acoustic signals) generated in the speaker's larynx into electrical analog
signals. The analog waveform 1 of a typical speech signal can be defined
in terms of frequencies over a band-from 30 hertz (Hz) to 10,000 Hz, or
10 kilohertz (kHz)-with most of the energy between 200 Hz and 3.5
kHz. It is not necessary, however, to reproduce speech waveforms precisely to achieve acceptable transmission quality because the human ear is
not sensitive to fine distinctions in frequency, and the human brain can
correct for inaccuracies such as missing syllables. Since transmission costs
are directly related to the range of frequencies transmitted (bandwidth),
the electrical signal used to provide commercially acceptable quality for
telephone communications is, thus, bandlimited to the range of frequencies
in which most of the energy occurs-between 200 Hz and 3.5 kHz (see
Figure 6-1).
The characteristics of the speech signal that vary with time are not
easily described. For example, the dynamic range is large; that is, the
amplitude varies rapidly and widely from moment to moment and from
speaker to speaker. There are talk spurts followed by silent intervals during which the speaker pauses for breath or stops to listen to the other
person. 2 Because of these variations, it is difficult to evaluate the speech
signal precisely, and the design of transmission systems capable of
achieving acceptable voiceband channel performance relies on measured
survey data. (Section 6.2 discusses channels.)

6.1.2 PROGRAM SIGNALS
Program signals include high-fidelity radio broadcasts, the audio portion
of television programs transmitted between a broadcaster's studio and the
station transmitters, and "wired music system" material that is distributed
to subscribing customers. The average volume, dynamic range, duration,
and bandwidth of program signals are usually greater than for ordinary
telephone speech signals. The bandwidth of a program signal, for
instance, may be as narrow as that required when speech alone is
transmitted, such as during a newscast; or it may be as wide as 35 Hz to
15 kHz when high-quality music for frequency modulation (FM) stereo
1 Green 1962 provides a good introduction to the concept of waves and their characteristics.
2 The Time Assignment Speech Interpolation (T ASI) System exploits this characteristic of
human conversation. TASI switches one conversation onto the transmission facility
during the idle time that occurs in another conversation.

LONG-TERM AVERAGE ENERGY
INCLUDING MALE AND
FEMALE VOICES

0.1

>-

c"

IE:
W

z

w
w
>
i=

0.01

«
..J

w

COMMERCIAL
BANDLIMITED
RANGE

IE:

0.001

0.0001
0.1

0.2

0.3 0.4

0.6 0.8 1.0

2.0

3.0

4.0

6.0 8.0 10.0

FREQUENCY (kHz)

Figure 6-1. Average speech signal energy. Experiments have shown that
frequencies below 1 kHz contain the intelligibility information and the higher
frequencies convey the articulation and improve the naturalness of the
received signal. There is a significant amount of energy below 200 Hz, but
these frequencies are excluded because they are subject to noise and
inductive interference (see Section 6.6.1).

broadcasts is distributed. The energy distribution is also difficult to
specify because of the variety of program material transmitted (for example, speech, music, sound effects).
6.1.3 VIDEO SIGNALS
Most video signals carried over Bell System facilities today are color
television signals. The color information (hue and saturation) is combined with the black and white (luminance) information in the television
picture and encoded into a highly complex electrical signal for transmission. The picture image is scanned rapidly at the transmitter in a systematic manner and then reproduced at the television receiver.
The synchronous scanning process used in the United States is illustrated functionally in Figure 6-2. The basic process consists of a sequence
of nearly horizontal scanning lines that run from left to right, beginning
at the top of the picture image. When one scan, or field, reaches the bottom of the image, the process is repeated, the next set of scanning lines
falling between the lines from the previous field. Two successive fields
195

Network and Systems
Considerations

196

Part 2

total 525 scan lines and make a complete picture frame. At a rate of 30
frames (15,750 scan lines) per second, the received picture appears not to
flicker because of the persistence of the human eye.
To permit decoding of the video signal at the receiver, it is necessary
to indicate the beginning of horizontal scanning lines and fields. This is
done by transmitting synchronizing pulses interleaved with the picture
information. To prevent the synchronizing pulses and traces of returns
to the next horizontal scan line from being visible on the screen, the picture signal is driven to a very low brightness level during the intervals
when pulses are transmitted. Since synchronizing pulses are essentially
digital signals and picture information is an analog waveform, video signals may be thought of as hybrid signals.

TRANSMITTER
IMAGE

,r-----"""'""\

COLOR
AND
LUMINANCE INFORMATION

RECEIVER
IMAGE

HORIZONTAL
SCANNING
LINES

SCANNING SYNCHRONIZING INFORMATION

SCANNING
PATTERN
GENERATOR

__ -:J
- - - -

-

INTERLACED FIELDS
VERTICAL FL YBACK
(RETURN TO TOP
OF FIELD)

Figure 6-2. Broadcast television scanning process. Horizontal picture
resolution is determined by the finest detail that can be resolved along a line
and vertical resolution by the number of horizontal lines per frame.
Subjective viewing tests have determined that a signal bandwidth of about
4.3 megahertz (MHz) is required to achieve horizontal resolution about
equal to the vertical resolution.

6.1.4 DATA SIGNALS
The signal produced by a computer or data terminal usually consists of a
stream of pulses that represents information coded into binary digits, or
bits (for example, 0 and 1). Unlike the speech signal, which contains
much inherently redundant information, the basic data signal has little or
no redundancy. For protection, bits are often added to the basic signal to

Chap. 6

Transmission

197

permit error detection (error-detecting codes). By adding more bits
(error-correcting codes), it is possible to correct many errors (that is, to
reconstruct the original signal without retransmitting the data).
Since data signals are often transmitted over systems that were originally designed for speech signals alone, the data signals must be processed at the source in a manner that ensures compatibility with the system. A train of "rectangular" digital pulses (see Figure 6-3, top) is not
directly suitable for transmission over analog facilities because it has frequency components extending from 0 frequency (dc) to a high frequency
(determined by the sharpness of the pulses and the rate at which they are
transmitted). To fit the bandwidths available in the analog transmission
channels, high-frequency components are restricted by modifying the
shape of the pulses (see Figure 6-3, bottom) and limiting the pulse rate.
The pulse rates most commonly used in the Bell System are those compatible with transmission over voiceband channels and range up to several
thousand pulses per second.
The characteristics 9f actual analog channels are accommodated by
changing the form of the signal through a process called modulation. An
example of modulation is frequency-shift keying (FSK), a technique in
which the transmitted frequency is shifted back and forth between 1.2
kHz (corresponding to the presence of a pulse or a binary 1) and 2.2 kHz
(corresponding to the absence of a pulse or a binary 0). (Section 6.4.4
contains more information on FSK.)
Digital facilities, coming into increasing use, can be a more effective
means for transmitting data signals. Even so, the digital pulse stream
requires processing at the source so that it fits into the format of the Digital Data System (see Sections 6.6.2 and 11.6.1) or the time-division multiplex hierarchy (see Sections 6.5.2 and 9.4.3).

TIME

Figure 6-3. Digital pulse streams.
modified pulses.

Top, rectangular pulses; bottom,

198

Network and Systems
Considerations

Part 2

6.1.5 OVERVIEW
In summary, signals may be characteristic of the producer or the receiver
of the signal (for example, a person or a teletypewriter). They may be
analog, digital, or, to some degree, hybrid (for example, video signals).
And signals may be converted from one form to another to make them
compatible with the transmission path over which they will travel with
no significant loss of information.

6.2 CHANNELS
6.2.1 BASIC CONCEPT
A channel is a transmission path for providing communications between
two or more points within the netwbrk or between the end points of a
connection (for example, from customer to customer or from one customer to several customers).3 A channel may be dedicated to a particular customer full time, may be switchable, or may even be changed in assignment during a call. A channel is the means by which information is
transmitted over the network; the signal transmitted is that information.
This distinction is important to remember throughout the rest of this
discussion.
Channels may be classified by their intended use and the corresponding kind of signal transmitted, for example, voiceband channels, program
channels, telegraph channels, and video channels. Signal requirements
may dictate the channel characteristics required to meet transmission
objectives; for example, I-way broadcast television requires wideband
channels to carry the signal bandwidth described earlier. In other cases,
the signal format that is actually transmitted over the channel may have
to be tailored to meet the characteristics of the channel. For example, a
data signal must be processed (coded, pulse-shaped, modulated) to be
suitable for transmission over a voiceband channel in the public switched
telephone network (PSTN).
Channels may also be classified by the broad nature of the transmission facility that provides the channel. If the transmission facility accepts
a band of frequencies and is compatible with the transmission of analog
signals, it is said to provide analog channels, and the facility is called an
analog facility. Analog channels can be further characterized by bandwidth:
narrowband channels (for example, 100 Hz, 200 Hz); voiceband channels
(4 kHz);4 broadband channels (for example, 48 kHz, 240 kHz).
3 The term multipoint is used for connections involving more than two end points.
4 The nominal voiceband channel is defined as 4 kHz although the speech signal is
essentially bandlimited to between 200 Hz and 3.5 kHz. The additional bandwidth allows
for a guard band on either side of the speech signal to lessen interference between
channels.

199

Transmission

Chap. 6

If the transmission facility accepts a train of pulses and carries signals
in digital form, it is said to provide digital channels and is called a digital
facility. Digital channels can be further characterized by pulse rate or bit
rate. For example, channels with a bit rate of 64,000 bits per second, or
64 kilobits per second (kbps), are often used in the Bell System. Digital
signals at lesser bit rates (2.4, 4.8, 9.6, and 56.0 kbps) may be carried on
these channels. For the lower bit rates, several signals may be combined
(multiplexed, as described in Section 6.5) onto the 64-kbps channel.

6.2.2 VOICE-FREQUENCY AND CARRIER-DERIVED CHANNELS
Channels in the Bell System network are provided by either voicefrequency (VF) or carrier transmission facilities. A VF facility is an analog facility that provides just one voiceband channel. In this case, the
voiceband channel is termed a baseband channel because the information
signal is carried at· its original frequencies. VF facilities are operated in
either a 2-wire or a 4-wire mode. In the 2-wire mode, both directions of
transmission are carried on the same pair of wires. Local loops and short
trunks are operated in this way to save copper wire and to be compatible
with local 2-wire switching systems, as shown at the top of Figure 6-4.
Because long 2-wire facilities requiring excessive gain would suffer from
instabilitys and singing,6 longer VF trunks and certain VF special-services

2-WIRE
CENTRAL
OFFICE

TELEPHONE
SET

TELEPHONE
SET

4

TELEPHONE
SET

2-WIRE
CENTRAL
OFFICE

T
W R

I

U

R N
E K

2·WIRE
CENTRAL
OFFICE

Figure 6-4. Voice-frequency 2-wire and 4-wire channels.
channel; bottom, a combination 2-wire /4-wire channel.

TELEPHONE
SET

Top, a 2-wire

5 The reflection of some portion of the transmitted signal back toward its source in
sufficient amplitude causes gain instability (see Section 6.6.1).
6 An undesirable self-sustained oscillation in a channel resulting from excessive positive
feedback. In a telephone connection, singing manifests itself as a continuous whistle or
howl.

Network and Systems
Considerations

200

Part 2

circuits (see Section 3.2.2) are often operated on a 4-wire basis (that is,
with a separate transmission path for each direction of transmission). As
shown at the bottom of Figure 6-4, a transmission circuit element called a
hybrid is necessary as an interface between short-distance 2-wire trunks
and long-distance 4-wire trunks.
In carrier transmission, signals are processed and converted to a form
suitable for a particular broadband medium. The facility and,
correspondingly, the channels may be analog or digital. A number of
channels are combined (multiplexed) in one carrier system, and they are
called carrier-derived channels (see Figure 6-5). Providing a number of
channels on one facility results in savings (for example, in copper wire)
for trunk lengths greater than some prove-in length. Therefore, although
the electronics required to combine channels may be expensive, carrier
systems become increasingly economical as their length and channel
cross section increase.
Carrier systems -consist of three major functional building blocks, as
shown in Figure 6-5:

· Multiplex/demultiplex terminals combine voiceband channels at the
carrier system input and separate them at the output.
· Carrier facility terminals process signals into a form suitable for
transmission over the high-frequency line and process signals that
have been transmitted over the high-frequency line into a form suitable for separation.
• The high-frequency line provides a broadband facility for the simultaneous transmission of a large number of voiceband channels. In this

MULTIPLEX
TERMINAL

-+

CARRIER
FACILITY
TERMINAL

t\

n

HIGHFREQUENCY
LINE

-+

-+

CARRIER
FACILITY
TERMINAL

-+

n

,
'" '" ..... .... ..... ">

VF 4-WIRE
CHANNELS

::+

DEMULTIPLEX
TERMINAL

., .,

.,..
".

CARRIER-DERIVED
CHANNELS

-+

VF 4-WIRE
CHANNELS

".

~

I
~

DEMULTIPLEX
TERMINAL

+-

...

I

CARRIER
FACILITY
TERMINAL

+

+-

HIGHFREQUENCY
LINE

+

CARRIER
FACILITY
TERMINAL

+-

MULTIPLEX
TERMINAL

n
VF VOICE FREQUENCY

n

ANY NUMBER

Figure 6-5. Example of carrier-derived 4-wire channels.

...

Chap. 6

Transmission

201

eontext, microwave radio qualifies as a high-frequency line, as does a
pair of wires in the T1 carrier system (see Section 9.4.2).
Carrier-derived channels operate in the 4-wire mode with a separate
path (for the combined channels) for each direction of transmission.
Most multichannel carrier systems on paired cable (see Section 6.3.2) use
a separate pair of wires for each direction of transmission. Some shorthaul carrier systems and submarine (undersea) carrier systems, on the
other hand, use an "equivalent 4-wire" mode of operation. Here, as in
certain radio systems, the paths for the two directions of transmission are
separated by frequency rather than physically.

6.3 TRANSMISSION MEDIA
Transmission media may constrain and guide communications signals or
permit signals to be transmitted but not guide them. Five types of
guided media used in the Bell System are open-wire lines, paired cable
(also called multipair cable), coaxial cable, waveguides, and lightguide
cable. The atmosphere and outer space are unguided media for transmitting terrestrial microwave radio signals and satellite communications,
respectively. This section describes fundamental applications and basic
advantages and disadvantages of each medium.

6.3.1 OPEN-WIRE LINES
Open-wire lines consist of uninsulated pairs of wires supported on poles
spaced about 125 feet apart. Five open-wire pairs may be mounted on a
single crossarm on a pole with a foot of space between wires to prevent
momentary short circuits in high winds. .Most poles are limited to about
ten crossarms. The wires range in diameter from 0.08 inch (12 gauge) to
0.165 inch (6 gauge) and may be copper, copper-clad steel, or galvanized
steel.
Open-wire lines have been largely supplanted by paired cable (see
Section 6.3.2), except in some rural areas, because of the limited number
of wire pairs that can be accommodated on a pole; the susceptibility of
the wires to storm damage, atmospheric corrosion, and interference from
power lines that share the poles; and the high cost of maintenance. A
singular advantage of open-wire is its low attenuation 7 at voice frequencies compared with paired cable.

6.3.2 PAIRED CABLE
Paired cable is composed of wood-pulp or plastic-insulated wires twisted
together into pairs. In some cable, many twisted pairs are stranded into a
rope-like form called a binder group; several binder groups are, in turn,
7 A decrease in the signal amplitude during transmission.

Network and Systems
Considerations

202

Part 2

twisted together around a common axis to form the cable core; and a protective sh eath is wrapped around the core. The amount of twist applied
is varied amon g pairs within each bind er group to reduce crosstalk.8
Paired cable is manufactured in a number of standard sizes and may contain from 6 to 3600 wire pairs. Typically, wires range in diameter from
0.016 inch (26 gauge) to 0.036 inch (19 gauge). Figure 6-6 shows typical
paired cable.
The main applications for paired cable are in the loop and· exchange
areas . The cable may be strung on poles, installed in underground conduit, or buried directly in the ground. To protect the cable in these

BINDER

GROUP

....._ _ _ CABLE

CORE

Figure 6-6. Typical paired cable.

8 Th e interference in on e communica tio n ch ann el ca used by a sign a l trave lin g in an
ad jacent chann el (see Section 6.6.1).

Chap. 6

Transmission

203

environments, sheaths of different materials or combinations of materials
are used. For example, cables that may be exposed to water are sheathed
in aluminum, steel, and polyethylene coated with a waterproofing compound. In addition, they are filled with a petroleum compound to keep
out water if the sheath is damaged. Most older paired cable is made
waterproof with lead sheathing and pressurized air within the cable core.
The lead sheath also provides protection from noise, but it is subject to
corrosion and ultimately may admit moisture.
On a given facility route, more circuits can be accommodated by
paired cable than by open-wire lines. Increased circuit demands have
also led to the extensive use of paired cable in carrier systems where,
however, the higher frequencies result in greater crosstalk between pairs
within a cable. To combat crosstalk, it is sometimes necessary to use
different cables for the two directions of transmission, to select pairs
within the same cable for opposite directions of transmission according to
special engineering rules, or to use a cable with an electrical screen
separating binder groups.
6.3.3 COAXIAL CABLE

Coaxial cable contains from four to twenty-two coaxial units called tubes.
Each coaxial tube consists of a O.IOO-inch copper inner conductor ke'pt
centered within a O.375-inch cylindrical copper outer conductor by
polyethylene insulating disks spaced about I inch apart. The outer conductor is formed into a cylinder around the disks and is held closed by
interlocking serrated edges along its longitudinal seam. Two steel tapes
are wound around the outer conductor for added strength. In addition to
coaxial tubes, coaxial cable contains a small number of twisted wire pairs
and single wires that are used for maintenance and alarm functions. Figure 6-7 shows typical coaxial cable.
An important advantage coaxial cable has over paired cable is its capability to operate at very high frequencies, which permits it to carry a relatively large number of carrier-derived channels. In addition, since the
copper outer conductor is grounded and provides a shielding that
improves with increasing frequency, crosstalk decreases rather than
increases with frequency as is the case with paired cable.
Coaxial cable is used primarily on intercity routes in the long-haul
network (see Section 4.4.2) where heavy cross sections of traffic exist. In
these applications, the installed first cost (materials, manufacturing, labor,
etc.) per circuit is substantially lower than for paired cable, although the
overall installed first cost is much higher than for paired cable.
Coaxial cable is also used in undersea cable systems (see Section 9.3.4).
However, the unique operating environment dictates design, operational,
and reliability requirements different from those for cable used on land.

".\~----

"3~~~~~~~--

"

COAXIAL TUBE

TWISTED WIRE PAIRS
AND SINGLE WIRES

INSULATING
DISK

OUTER
CONDUCTOR

Figure 6-7. Typical coaxial cable. (Redrawn and adapted with permission
from Martin 1976, p. 162.)

6.3.4 WAVEGUIDES
A waveguide is a rectangular or circular copper pipe that confines and
guides radio waves between two locations. Its main advantage is very
low attenuation at microwave frequencies compared, for instance, with
coaxial cable. Very wide bandwidths can be achieved as well. The application of waveguide is limited, though, because (1) it must be manufactured to extreme uniformity and (2) extreme care is required during
installation to minimize sharp bends for hilly terrain or right-of-way curvature, which result in transmission mode 9 conversion and increased
attenuation.
Circular waveguides have been used to transmit 4-, 6-, and 11gigahertz (GHz)lO microwave signals simultaneously from the bases of
microwave radio towers to the antennas on top. (The position of a circular waveguide is shown in Figure 6-10.) The waveguide diameter is
chosen to achieve low attenuation across this band of frequencies, considerably lower attenuation than for the same length of rectangular pipe. In
addition, the circular cross section of the waveguide permits the transmission of two signal polarizations so that two signals at the same frequency

9 A transmission mode is a particular form of signal propagation within a waveguide
characterized by the configuration of the electromagnetic field in the plane transverse, or
perpendicular, to the direction of propagation. Major modes are the transverse electric
(TE) and the transverse magnetic (TM). In the TE mode, the electric field is perpendicular
to the axis of the waveguide and the magnetic field is parallel to it.
10 One gigahertz (GHz) equals one billion hertz.

204

Transmission

Chap. 6

205

can be transmitted simultaneously, effectively doubling the channel
capaci ty .11

6.3.5 LIGHTGUIDE CABLE
In lightguide transmission systems, an electrical signal is converted to a
light signal by a light source (for example, a light-emitting diode or a
laser) and then coupled into a glass fiber. Two types of lightguide cable
are currently in use: ribbon-fiber cable and stranded cable. One form of
ribbon-fiber cable consists of from one to twelve flat ribbons, each containing twelve glass fibers, thereby providing up to 144 one-way paths
for optical signals. As shown in Figure 6-8, the ribbons are stacked into a
rectangular array and twisted to improve the flexibility of the cable. Two
sheaths, both consisting of polyethylene with steel reinforcing wires, protect the hair-thin fibers (0.005 inch) from excessive pulling tensions. The
cable diameter is about 0.5 inch, regardless of the number of optical fibers
it contains. Single-unit stranded cable has up to 16 fibers woven around
a central strength member. A 72-fiber multiunit cable consists of six
single units, each having 12 fibers.
Optical fibers (see Figure 6-9) are composed of three concentric
cylinders made of dielectric materials: 12 the core, the cladding, and the
jacket. The jacket is a light-absorbing plastic that prevents crosstalk and
protects the surface of the cladding. The core and the cladding are transparent glass designed to guide the light within the core and to reflect it

STEEL REINFORCING
WIRES

Figure 6-8. Typical ribbon-fiber lightguide cable.

11 Signal polarization refers to the direction of a signal's electric field. The number of
signals carried by a channel may be doubled if the polarizations of two signals are
perpendicular to one another.
12 Materials that do not conduct electrical current.

SINGLE·MODE FIBER
INPUT

PATHWAYS THROUGH
THE FIBER

OUTPUT

JACKET
SINGLE MODE
CLADDING

JUL
CORE

MULTIMODE
(GRADED INDEX)
MUL TIMODE FIBER

N\

JACKET

CLADDING

CORE

Figure 6-9. Optical fibers.

as it grazes the interface between them. The refractive index 13 of the core
is greater than that of the cladding, so light in the cladding moves faster
than light in the core. As a light signal in the core moves toward the
cladding (from a region of lower velocity to one of higher velocity), it is
bent back toward the core and is, thus, guided along the fiber.
The two basic types of optical fibers are single mode and multimode.
The core diameter of a single-mode fiber is sufficiently narrow to restrict
light to one mode of propagation (that is, light travels only one path
through the fiber). Because of this restriction, transmission over singlemode fibers is not subject to modal dispersion,14 but it is more difficult to
couple light from one single-mode fiber to another. Multimode fibers
have much wider cores that allow light to enter at various angles and
travel through the fiber along different zigzag paths of varying lengths.
Since the path lengths are different, so are the times required to travel
them at the same speed (that is, with a uniform index of refraction). This
results in modal dispersion. To reduce dispersion, the index of refraction
is graded across the core (that is, the index of refraction in the core

13 The ratio of the velocity of a lightwave in free space to its velocity in a given medium.
14 Signal distortion caused by multiple modes combining at the output of a fiber and
spreading out the signal in time. For example, if the signal were in the form of pulses,
the pulses would spread out and, thus, overlap causing them to be less distinguishable at
the receiving end of transmission.

206

Chap. 6

Transmission

207

decreases as distance from the center of the core increases). Thus, light
traveling a shorter path near the center moves more slowly, while light
traveling the longer path moves more quickly. The result is that the
travel times tend to be equal.
The main advantages of lightguide cable are:
• immunity to electromagnetic interference and crosstalk.
• small size (0.5-inch outer diameter) and light weight (approximately
80 pounds per 1000 feet of cable). Paired cable of equivalent capacity
is approximately 3 inches in diameter and weighs 100 times as much.
• low attenuation allowing repeaters 15 to be positioned far apart. Maximum spacings with current technology are 25 miles for single-mode
fibers and 15 miles for multimode fibers.
• high communication capacity. With currently available systems, 144fiber cable can accommodate approximately 250,000 voice-frequency
circuits. This capacity will increase rapidly in the future as a result of
new technology.
• wide bandwidth: 100 GHz for a single-mode fiber and 1 to 2 GHz for
a multimode fiber at the longer wavelengths.
In many potential applications, the advantages outweigh the disadvantages. These applications include intraoffice data busing between computer peripherals and equipment frames in digital switching systems,
loop carrier systems, medium- and large-capacity interoffice trunks,
large-capacity long-haul intercity routes, special video hookups, and
transoceanic cable systems.
6.3.6 TERRESTRIAL MICROWAVE RADIO

The terrestrial microwave radio medium consists of a line-of-sight propagation path 16 through the earth's atmosphere and associated towermounted transmitting and receiving antennas shown in Figure 6-10.
Microwave energy can be focused into a narrow, strongly-directional
beam, similar to a beam of light. Antenna towers must, therefore, be
rigid enough to withstand high winds without excessive antenna
deflection, which results in increased attenuation. The Federal Communications Commission (FCC) makes a portion of the radio-frequency bands
of the electromagnetic spectrum available for common-carrier service.

15 Devices that perform amplification or regeneration and associated functions. In
transmitting digital signals (for example, lightguide signals), repeaters are regenerative;
that is, they reconstruct each digital pulse to its original form.
16 The straight path between the radio transmitting antenna and the receiving antenna.

ANTENNAS

CIRCULAR
WAVEGUIDE

Figure 6-10. A typical microwave radio station.

The Bell System applications have been primarily in the 4-, 6-, and 11GHz bands. Under normal atmospheric conditions, the received
microwave energy on a line-of-sight path decreases in proportion to the
square of the distance between two radio towers. The decrease in energy
is relatively constant over a wide band of frequencies .
The two main characteristics of radio transmission are that no physical
facility is required to guide the microwave energy between repeater stations and that these stations may be positioned many miles apart. The
spacing between repeater stations is determined by the geography of a
given route, the technology used in the terminal equipment, and the
transmitter power permitted by the FCC. Typical repeater spacings in the
Bell System are 20 to 30 miles, but much longer spacings are possible
where there is little fading activity. (Fading is discussed in the following
paragraphs.) Microwave radio, thus, has the advantage of .being able to
span natural barriers such as rugged or heavily wooded terrain and large
bodies of water.

208

Chap. 6

Transmission

209

However, using the earth's atmosphere as the transmission medium
results in problems not experienced with other media. For example,
heavy ground fog or very cold air over warm terrain can cause enough
atmospheric refraction to reduce the power of the signal noticeably. This
fading affects a wide band of frequencies and may last several hours. The
remedy is to use higher antennas or position them closer together.
A second type of fading, called multipath fading, occurs mostly at night
and at dawn during the summer, when there is no wind to break up
atmospheric layers. Normally, microwave energy radiated outside the
line-of-sight path (called off-axis energy) is bent by atmospheric refraction
into the receiving antenna. This energy travels a longer path than the
line-of-sight signal and, therefore, has a longer travel time. Depending
on the amount of off-axis energy and its phase (the amount of delay relative to the primary signal), instantaneous reduction and even cancellation
of the primary signal may occur. Multipath fading may recur frequently
but lasts only a few seconds and, generally, affects only certain frequencies. It can be minimized by using frequency diversityJ7 Since the use of
frequency diversity requires more of the limited radio-frequency spectrum available, it is generally not applied to low-density routes. An alternative approach, space diversity, IS is used in this case and for paths that
experience severe fading.
Another problem is reduction of received signal power from rain
attenuation, which is particularly severe at 11 GHz and higher frequencies. Greater absorption of microwave energy occurs with increasing frequencies as the wavelength of the signal approaches the size of the rain
drops. Neither frequency diversity nor space diversity is effective in
countering rain attenuation. The only remedy for maintaining acceptable
transmission quality is the use of shorter repeater spacings.
6.3.7 SATELLITE

The transmission medium of a satellite system consists of a line-of-sight
propagation path from a ground station to a communications satellite and
back to earth. The ground station includes the antennas, buildings, and
electronics needed to transmit, receive, and multiplex signals. The satellite is usually placed in a geosynchronous orbit (about 22,300 miles above
the earth) so that it appears stationary from any point on the earth. It

17 Transmission of the same radio signal over different microwave frequencies at the same
time. For a given set of atmospheric conditions, radio signals at different frequencies
will experience different degrees of multi path fading.
18 Space diversity uses two receiving antennas usually mounted on the same tower and
separated vertically by several wavelengths. By switching from the regular antenna to
the diversity antenna whenever the signal level drops, the received signal is maintained
nearly constant.

l"lt:{WUTK lUlU

210

~y:s{t:lIU;

Considerations

Part 2

acts like a radio relay station in the sky and uses the same frequency
spectrum as terrestrial microwave radio.
The satellite's microwave antennas are directional and may cover the
whole earth or just a part. A beam of energy from the satellite about 17
degrees wide will just cover the earth's surface. To cover a portion of the
earth's surface only 500 miles across, a beamwidth of 1 degree would be
required. While narrow beamwidths result in gre(~.ter received power,
they require sophisticated antenna control systems to keep the antennas
oriented properly. One method is to spin the satellite in one direction
and its antennas in the opposite direction at an equal rate.
One advantage of satellite transmission is the capability to send large
amounts of information to almost anywhere on the earth, regardless of
how remote or distant. Another is that transmission cost is independent
of the distance between sending and receiving locations. Further, each
satellite system requires only one satellite repeater. A comparable terrestrial microwave system between distant locations requires many repeaters
placed in tandem because the curvature of the earth interferes with the
line-of-sight transmission path. Amplification of the signal by each
repeater increases the effects of distortion and noise. In some cases, this
characteristic of satellite systems may make transmission via satellite
superior to transmission via terrestrial microwave radio.
However, because of the great distance between the earth and the
satellite repeater, satellite transmission produces greater attenuation and
longer delays (approximately 0.25 second one way) than transmission
using terrestrial systems. The effects of attenuation can be overcome by
using high-gain repeaters and high path-elevation angles (above 20
degrees) for the ground antennas (to reduce the path length through the
atmosphere) together with narrow-beam transmission. In the 4- and 6GHz frequency bands, attenuation is not a significant problem. At frequencies above 10 GHz, however, rain attenuation is a problem on satellite paths. Since heavy rain is not likely to occur simultaneously at two
widely separated ground stations, rain attenuation can be minimized by a
form of space diversity.
The long delay is another matter. Besides impairing the quality of
voice communications, long delay can seriously affect data transmission
unless protocols 19 and control of user terminals are designed to match
this media.

6.4 MODULATION
Modulation converts a communication signal from one form to another

more appropriate form for transmission over a particular medium
between two locations. Demodulation restores the signal to its original

19 Strict procedures for initiating and maintaining communications. (See Section B.B.)

Chap. 6

Transmission

211

form at the receiving end of the transmission medium. Modulation is
often necessary because a signal, even when bandlimited, is rarely in the
best form for direct transmission over the frequency spectrum available
on a given medium.
A simple example of modulation is the transformation of an acoustic
signal into an electrical analog signal by the telephone transmitter (discussed in Section 6.1.1). When the telephone is off-hook, a direct current
flows over the customer's loop from the central office to the station set.
This direct current can be conceived of as a O-Hz carrier signal whose
amplitude is varied by changes that occur in the sound-wave pressure
when a customer speaks. The modulating signal in this case is the acoustic signal; the modulated signal is the varying amplitude direct current,
and the ac component of this modulated current is usually referred to as
the speech signal.
Some reasons for modulating a communication signal prior to
transmission are:
• to enable a number of signals from voiceband channels to be combined (multiplexed, as described in Section 6.5) and transmitted simultaneously over a common broadband medium, thereby reducing the
per-channel transmission cost.
• to shift the signal frequencies upward to simplify the design of
repeater components and, in the case of radio communication, to
translate the signal spectrum into one of the frequency bands allocated to common carriers by the FCC. Demodulation shifts the signals
down.
• to convert voice or other analog signals to digital form to achieve
lower noise or less distortion as well as economies in system
implementation.
• to convert digital signals for transmission over analog channels that
would destroy the digital waveform.
The following sections describe some commonly used modulation
methods applied in the Bell System network. These include doublesideband amplitude modulation, single-sideband amplitude modulation, pulse-code modulation, and methods of modulating digital data
signals for analog facilities.

6.4.1 DOUBLE-SIDEBAND AMPLITUDE MODULATION
One of the more familiar forms of amplitude modulation (used in AM
broadcasting) is double-sideband amplitude modulation (DSBAM). In
amplitude modulation, the amplitude of a sinusoidal carrier signal is
varied continuously by the information signal, and the modulated signal
is translated in frequency to occupy a distinct frequency band for

l''Ilet:WUTK ana

212

::>y~t:eIIU;

Considerations

Part 2

transmission over a particular medium. DSBAM generally uses product
modulation; that is, the product of an analog baseband information signal, a(t), and a sinusoidal carrier signal, cos wet, is formed. To obtain a
modulated signal, M (t), from which the baseband signal is easily
recovered by demodulation, the carrier signal is added to the product:
M(t) . cos wet

= [1

+ a(t).

cos wet

+ a (t)]cos wet.

Here, We = 27rf e and f e is the carrier frequency. Figure 6-11 illustrates
this process.
BASEBAND
MODULATING
SIGNAL

artJI .----....

MODULATED
SIGNAL

~
CARRIER

I

:IG~AAL
cosw,t~
A A

A

A A

o

A ,

M_~_~O_U~_~C_T6_R_a_(_t_)._co_s_w_ct_~+cos wJ ~

-+_ _

t

t

Figure 6-11. Double-sideband amplitude modulation process.

Figure 6-12 is a representation of the modulated signal in the frequency domain. DSBAM, translates the frequency spectrum of the
baseband signal symmetrically about the carrier frequency. The upper
sideband constitutes a pure frequency translation above the carrier frequency, while the lower sideband inverts the baseband frequencies below

w

CARRIER
SPECTRUM

BASEBAND
SPECTRUM

Q

:::I
I-

::::;
Q.

~

-<

~---FREQUENCY

Figure 6-12. Frequency representation of DSBAM.

Chap. 6

Transmission

213

the carrier frequency. Both sidebands contain the same information. The
carrier frequency is chosen so that the lower sideband does not overlap
the baseband signal spectrum.
This form of modulation is attractive because filters are not required to
eliminate an unwanted sideband as in single-sideband amplitude modulation (see Section 6.4.2), and therefore, terminals are less expensive. In
addition, since the carrier signal is transmitted in these systems, no
separate carrier signal needs to be supplied by the receiving terminal for
demodulation.
However, since the carrier signal and the redundant sideband are both
transmitted, the power-handling requirements of the transmission equipment are much greater than if only one sideband were transmitted. Also,
DSBAM transmission requires twice as much bandwidth as the baseband
signal. Both of these factors increase the line-haul cost (cost per mile) per
voiceband channel of these systems and limit their economic application
to shorter distances than those served by single-sideband systems. 20
6.4.2 SINGLE-SIDEBAND AMPLITUDE MODULATION

To obtain a single-sideband amplitude modulated (SSBAM) signal, either
sideband of the product-modulated spectrum described in Section 6.4.1
may be selected for transmission by a bandpass filter. In most Bell System
applications, the lower sideband is selected.
Single-sideband transmission requires only one-half the DSBAM
bandwidth per voiceband channel, allowing twice as many voiceband
channels to be carried in a given frequency band (by multiplexing). (Figure 6-13 illustrates this.) For this reason, SSBAM is the technique usually
found in the multiplex terminals used with long-haul broadband facilities
incorporating frequency-division multiplexing of voiceband channels (see
Sections 6.5.1 and 9.3.5). Also, since the individual carrier frequencies are
not transmitted over these facilities, the ratio of information power to
total transmitted power is greatly improved. Thus, the cost of line
repeaters is significantly less than for DSBAM.
Suppression of the carrier frequency at the transmitting terminal
improves the power efficiency of SSBAM but requires the insertion of a
precise and stable carrier frequency at the receiving terminal for demodulation. This must be done with extreme phase accuracy to avoid an
impairment called quadrature distortion. 21 Data signals are particularly sensitive to distortion. Improved control of the phase as well as the frequency of the carrier supply signals in SSBAM systems is provided by the

20 A short-haul system (N2) is described in Section 9.3.3.
21 Waveform distortion caused by a phase error in the demodulation process (see Bell
Laboratories 1982, pp. 100-102,300).

Network and Systems
Considerations

214

Part 2

nationwide Bell System Carrier Synchronization Network, which
distributes synchronization signals throughout the United States based
upon two standard reference signals (see Section 6.5.3).

W
Q

::I
I-

::::i
a.

:::IE


en

L

SYSTEM LENGTH

Figure 6-16. Multiplex cost tradeoff.

6.5.1 FREQUENCY-DIVISION AND TIME-DIVISION
MULTIPLEXING
The two basic multiplexing methods used in the Bell System are
frequency-division multiplex (FOM) and time-division multiplex
(TOM). FOM divides the frequency spectrum of a broadband transmission system into many full-time communications channels. Signals from
these channels are transmitted at the same time but at different carrier
frequencies. In the Bell System, for example, the standard analog communications channel is the nominal O-kHz-to-4-kHz voiceband channel.
At the transmitting (multiplex) terminal (shown in Figure 6-5), voiceband
channels are shifted up in frequency using amplitude modulation and
occupy specific frequency slots associated with broadband transmission
systems. At the receiving (demultiplex) terminal, all the voiceband channels are shifted back in frequency using amplitude demodulation, and
then separated for distribution.

Network and Systems
Considerations

220

Part 2

In TOM, a transmission facility is shared in time rather than frequency. This is accomplished as indicated in Figure 6-17, by interleaving
PCM samples from a number of voiceband channels on a common bus.
At the transmitting terminal, several slow-speed pulse streams (digital
channels) are combined into a composite high-speed pulse stream. At the
receiving terminal, the slower speed streams are separated. It is important to note that transmitting and receiving terminals must be synchronized so that the pulses can be correctly identified and kept in the proper
relation to one another. Network signaling bits and framing bits are
added to the coded information bits as noted in the figure.

6.5.2 MULTIPLEX HIERARCHIES
The time-division multiplexing of twenty-four voiceband channels illustrated in Figure 6-17 is typically accomplished in the local network. In
the intercity network, the concentration of traffic provides economic

--t-t-J:::::7
.~

CHANNEL 3

:

CHANNEL24

INTERLEAVED
PAM SAMPLES

I I

I

II:~II

~

~

I I
I

~

I I 1 I

I 1
I L ___ -,

-------,
LL
------,
---,
I
1

~'C7~~4n~
I 1 1 I
I I

I

I

I ------'l'f--t-t_1-".....
--;'-+1....,.i1___---li!~!-1-1L.-_
L_...,

I

PCMOUTPUT~~~
OF ENCODER

I 1 I

I 1 I

"

L. •

I
1
I
~F ~~

1111" 1,11.111" ,,11.1.1 ":;. ,,11.1, I" I 1,11...1
CH 1

CH2

CH3

"",,~--{ 1 FRA~~~f:3 BITS

CH24

CH 1

II

JII
CH2

} - - . . . . ,....,

NOTES:
1.

FRAMING BIT IS 193RO BIT (F).

2.

SIGNALING BIT IS 8TH BIT IN EACH CHANNEL IN ONE FRAME OUT OF SIX (S).

3.

THE OUTPUT PULSE RATE SHOWN IS 1.544 Mbps, THE OS1 LEVEL IN THE
TOM HIERARCHY. (SECTIONS 6.5.2 AND 9.4.3 DISCUSS THE TDM HIERARCHY.)

Figure 6-17. Time-division multiplexing.

•••

Chap. 6

Transmission

221

opportunities to combine channels into larger and larger bundles for
long-haul transmission. A number of different size bundles are needed
to make the most efficient use of the available transmission facilities to
satisfy the range of route capacities and distances found in the network.
A hierarchy of multiplex levels or steps has been developed to meet this
need and to provide for orderly and efficient growth of the network. In
fact, there are two hierarchies: one for analog transmission systems and
one for digital transmission systems.
In the FOM hierarchy, multiplex levels correspond to increasingly
wider frequency bands. The initial level is a grouping of twelve
voiceband channels by an A-type channel bank. Additional levels eventually lead to the multiplexing of a large number of channels compatible
with broadband transmission systems. (Section 9.3.5 includes information
on A-type channel banks and the FOM hierarchy, and Figure 9-19 is a
graphic representation of it.)
The TOM hierarchy consists of multiplex levels corresponding to
increasingly higher pulse rates. The initial level is shown in Figure 6-17.
It is called the DSl level and is a grouping of twenty-four voiceband channels by a O-type channel bank. Additional levels are designed to be compatible with high-capacity digital facilities. (Section 9.4.3 includes information on the O-type channel bank and the TOM hierarchy, and
Figure 9-26 is a graphic representation of it.)

6.5.3 MULTIPLEX SYNCHRONIZATION
In the Bell System, the need for accurate frequency and stable timing signals to coordinate the numerous analog and digital transmission and
swi tching systems now in service is filled by the Bell System Reference
Frequency Standard (BSRFS) and a nationwide distribution network
called the Bell System Carrier Synchronization Network. The center of this
tree-like network is located in an underground repeater station in
Hillsboro, Missouri. Here, the BSRFS (consisting of three interlocked
cesium frequency standards accurate to one part in 1011) generates two
reference signals: one at 2.048 MHz and another at 20.48 MHz.
The FDM Synchronization Plan
As noted in Section 6.4.2, modulation using single-sideband (SSB) FDM
terminals requires the insertion of a precise and stable carrier frequency
at the receiving terminal for demodulation. If the frequencies of the
transmitting and receiving terminals differ, an impairment called carrier
frequency shift (see Section 6.6.1) that appears as a frequency offset in the
information signal may result. To synchronize modulation and demodulation operations in SSB FDM terminals, reference frequency signals are
transmitted from the BSRFS to regional synchronization centers deployed
throughout the country. At each center, regional frequency supplies provide

222

Network and Systems
Considerations

Part 2

various control and operating signals (for example, carrier frequencies) to
regional central offices needing FOM synchronization. These regional
central offices are equipped with primary frequency supplies, whose signal
frequencies are synchronized to the incoming signals from the regional
frequency supplies. Multiplex synchronizing signals are then transmitted
independently over coaxial or microwave facilities.

The PCM-TDM Synchronization Plan
As mentioned previously, a synchronization plan is a crucial component
of time-division multiplexing because digital signals cannot simply be
combined at the transmitting end and correctly identified at the receiving
end unless transmitting and receiving multiplex terminals are locked to a
common clock (that is, they are in synchronism). Two forms of synchronization are important at each level in the digital hierarchy. The
first is synchronizing each bit stream to a nominal pulse rate. The
second, and more fundamental, is synchronizing the relative timing of
several bit streams so that they can be multiplexed together into one bit
stream at a higher rate, and correctly identified and separated at the
receiving end.
Since digital signals often originate in different locations that are
separated by long distances and have independent timing clocks, synchronizing the relative timing of bit streams is a problem. The solution
above the OS! level is a process called pulse stuffing, in which bit streams
to be multiplexed are each stuffed with additional dummy pulses to raise
their rates to that of an independent, locally generated clock. The outgoing rate of the multiplexer24 is therefore higher than the sum of the
incoming rates. The dummy pulses carry no information and are coded
so that they can be recognized and removed at the receiving terminal.
The resulting gaps in the pulse stream are then closed to restore the original bit stream timing.
At bit rates above the OSI level, delay variations caused by various
transmission media can, in turn, cause timing discrepancies on the order
of ± 100 pulses in the received pulse stream. In principle, the departure
from nominal timing could be handled by buffer storage,25 but this
becomes expensive at higher bit rates. With pulse stuffing, differences
are corrected immediately, so that only a small amount of storage is
needed to accommodate these variations. Another advantage of the
pulse-stuffing method is high system reliability. The failure of a multiplexer or associated transmission facility will affect only those signals

24 The term multiplex is often used interchangeably with or in place of multiplexer to mean
the equipment that performs multiplexing.
25 A·
· storage.
n mtermed·late storage me d·!Urn b etween d
ata·mput an d
active

Chap. 6

Transmission

223

passing through the failed elements, since the timing at other multiplexers is independently derived.

6.6 TRANSMISSION IMPAIRMENTS AND OBJECTIVES
Telecommunications signals are degraded by the practical limitations of
channels (such as bandwidth), impairments arising from within the channel (for example, attenuation and echo), and impairments introduced
from outside of the channel (for example, power-line hum and radiofrequency interference). The impact of many impairments depends on
the type of signal carried over the channel and whether the channel is
analog or digital. For example, impulse noise26 from electromechanical
switching systems has little effect on the reception of analog speech signals because of amplitude limiters in the circuit and the relative tolerance
of the human ear. Digital data signals, on the other hand, can be seriously affected by impulse noise; blocks of data can be obliterated, resulting in high error rates in the received signal.
The planning, deSign, installation, operation, and maintenance of analog and digital transmission facilities are based on a thorough understanding of impairments and an appropriate set of transmission objectives
which, when achieved, lead to customer satisfaction at a reasonable cost.
Transmission objectives are dynamic: They respond to changing subjective reaction to impairments, new technology, new measuring techniques,
and improved analysis tools. The steps involved in determining a new
objective or adjusting an existing objective to achieve a balance between
customer satisfaction and cost include:
• identifying the impairment that needs to be controlled, including
defining the impairment in terms of a measurable parameter
• specifying a method to measure the impairment parameter
• evaluating the impact of the impairment on transmission performance
• finding the means to control the impairment
• formulating an end-to-end performance objective for the impairment
that will provide a satisfactory quality of service
• allocating the end-to-end objective to individual parts of the network
in terms of equipment design requirements, engineering application
guidelines and rules, and maintenance requirements and limits for the
use of craft forces
• establishing methods to monitor network performance to ensure that
objectives are being met (see Chapter 16).
26 Short bursts of high-level noise that sound like clicks over a telephone line. (See Section
6.6.2.)

Network and Systems
Considerations

224

Part 2

For convenience, the impairments and objectives considered here are
divided into those that primarily affect analog speech signals and those
that primarily affect digital data signals. 27

6.6.1 ANALOG SPEECH SIGNALS
Early in this century, the emphasis in communications was on the conquest of distance. The basic goal in telephone transmission was to provide a satisfactory signal volume at the receiver so that the talker could
be heard. Once this goal was achieved, attention turned to improving
the quality of transmission through technological advances.
Transmission objectives were initially established for the transmission
of speech signals over the PSTN. As new types of signals and services
evolved, these objectives were modified and new objectives were formulated. The following discussion begins with the problem of received
volume and proceeds to other impairments and applicable objectives.
Subjective tests have been used to determine listener reactions toa
range of received volumes and other impairments. Satisfactory transmission performance, in turn, is expressed in terins of a grade -of service that
is based on expected customer satisfaction with the quality of telephone
connections provided by the network; that is, what percentage would rate
the quality as excellent, good, fair, poor, or unsatisfactory. Objectives for
impairments are generally set so that the majority of listeners would rate
the transmission as excellent or good. Chapter 16 describes the setting of
objectives in more detail.
Received Volume and Loss28 Objectives
With many impairments, the more intense the impairment, the poorer the
received signal. Received volume, in contrast, can impair transmission in
two ways: The signal may be too weak or too loud. If it is too loud, the
27 Bell Laboratories 1982, pp. 136-139, discusses impairments peculiar to video signals.
28 In the Bell System, the term loss refers to insertion loss, a quantity that represents a
specific relationship between the input and output of a network (for example, a customer
connection or a circuit). The figures below illustrate a basic insertion loss calculation. In
Figure 1, the generator is connected directly to the load, and the power delivered to the
load is P l' In Figure 2, a network exists between the generator and the load; with the
network in place, power P 2 is delivered to the load. Insertion loss is the ratio of PI to P 2
and is expressed in decibels (dB):
PI
insertion loss (in dB) = 10 log 10 P2

GENERATO~
Figure 1

LOAD

LOAD

(POWER P 1)

(POWER P 2)

Figure 2

Chap. 6

Transmission

225

listener may be uncomfortable; if it is too faint, the listener may have
difficulty understanding the received message. Received volume depends
on several factors including telephone speaking habits, the acoustic-toelectric conversion efficiency of the station set, the amount of sidetone29 in
the telephone, and the attenuation in the built-up connection between
speaker and listener.
Loss objectives (the amount of loss that provides a satisfactory grade of
service) have been derived for individual portions (loops, trunks) of
built-up, end-to-end, customer connections. Original analysis considered
received volume only. This was followed by a model that combined
received volume and idle circuit noise. Later the effects of talker echo
were included to provide for performance evaluation in terms of a combined received volume-noise-echo grade of service.
Transmission objectives for loop loss have been derived based on
achieving satisfactory performance in terms of received volume and noise
impairments for a variety of built-up connections. Loop loss is controlled
by specifying design methods to produce a satisfactory distribution of
losses. (See Section 9.2.1.) When properly applied, the methods, together
with trunk objectives, ensure that overall objectives on built-up connections will be met. 30
Loss objectives for trunks are allocated on the basis of the need to provide satisfactory volume, the need to minimize the contrast as perceived
by the customer between different types of calls (local, long distance,
operator-assisted, etc.), the position of the trunk type in the switching
hierarchy (direct, tandem, toll connecting, etc.),31 and the need to control
echo on trunks involved in longer connections through the network. On
shorter trunks (less than about 10 miles for 2-wire voice-frequency [VF]
trunks, 25 miles for 4-wire VF trunks, and 300 miles for carrier trunks)
where echo is not a problem, the loss objectives are as listed in Table 6-1.

Talker Echo
Talker echo results when an electrical speech signal travels a long distance and some portion of it is reflected back toward the speaker. If the
elapsed time (delay) is long and the echo path loss32 is inadequate (that
is, there is not enough loss to attenuate the echo), the echo can interfere

29 The portion of the signal from a telephone transmitter that appears at the receiver of that
telephone. Some sidetone appears to be desirable to assure a customer that the telephone
is working and to help the talker adjust the level of speech.
30 Based on measurements of I-kHz loss made in 1960 and 1964, the Bell System average
loop loss was 3.8 dB, and the standard deviation was 2.3 dB.
31 Section 4.2.1 describes the types of trunks in the switching hierarchy.
32 The return loss (a function of the hybrid balance) plus twice the circuit loss between the
speaker and the point of reflection.

Network and Systems
Considerations

226

Part 2

TABLE 6-1
RECEIVED VOLUME LOSS OBJECTIVES
FOR SHORTER TRUNKS IN
THE LOCAL NETWORK
Loss
(dB)

Trunk Type

Direct
Tandem
Toll connecting

Nominal

Maximum

3
3

5
4

3

4

with the talker's speaking process. In extreme cases, echo may be so
great that speaking is nearly impossible.
Figure 6-18 illustrates a frequently encountered situation in the PSTN
that results in echo. The figure shows a 4-wire intertoll trunk terminating in a toll office. This trunk may be switched to anyone of several
class 5 end offices over 2-wire toll connecting trunks and, in turn, be connected to one of a number of loops with widely different impedances. 33
The transition between 4-wire and 2-wire transmission is provided by a
hybrid circuit. 34
If the hybrid termination is ideal (perfectly balanced; that is,
impedances Z and Zc are equal), the transmission loss across the hybrid
(from B to C on the figure) is infinite, and no echo returns to the talker.
This means that none of the signal in the transmission path A to B is
returned in path C to D. Achieving perfect hybrid balance is difficult
because the impedance Z is different for every connection through the
switches, but the hybrid balance impedance Zc is fixed. Nevertheless, the
potential effects of highly variable loop impedances can be and are
masked because the impedance of the 2-wire toll connecting trunks is
carefully controlled. If, however, the 4-wire-to-2-wire interface is at the
end office instead of at the toll office, there is no buffering by the toll
connecting trunks and more serious echoes occur. Because of this lack of
precise control over loop impedances and the insufficient degree of
impedance matching possible at the class 5 end office, other means of
controlling echo must be used.
33 The opposition to the flow of alternating current by an element in a circuit.
34 Hybrid circuits are used for various purposes. In this application (that is, as a transition
between 2-wire and 4-wire circuits), it is also called a 4-wire terminating set.

END OFFICE
TOLL OFFICE
(CLASS 2,3, OR 4)

r-- - - - -

I

-

-

-..., LOOPS

I
,

-r---------...,.,4-___,...4J1
I
I

o
4·WIRE {
INTERTOLL
TRUNK

r' - -

I SWITCHING

I
I

2·WIRE
TOLL
,CONNECTING
I TRUNKS

I

I

®

I •

I NETWORK I

,
SWITCHING
NETWORK

1

•

I

I

•

I
I

I

n

L.: _____ :J

END OFFICE
. - : - - - - - , LOOPS
I

TO
OTHER

I
I

I

~~~

,

•

:

L _______

TOLL
SWITCHING:.
}.CONNECTING I NETWORK
I
TRUNKS
I
:
•

~_-_-_-_~_~

I
I

I

L.: _____ .J

m

Figure 6-18. Talker echo path-4-wire intertoll
trunk to 2-wire toll connecting trunk.

The impact of talker echo depends on the echo's amplitude, the echo's
delay in transmission (which is proportional to trunk length), and the
speaker's tolerance of the echo. The echo becomes increasingly bothersome as echo amplitude and delay increase. Echo amplitude depends on
the loss of the signal up to the point where reflection begins, the hybrid
balance (return loss) at the reflection, and the loss in the path back to the
talker. Figure 6-19 shows how the grade of service perceived by the
talker varies with echo path loss and delay.
An inexpensive means of meeting grade-of-service objectives for echo
is to increase the loss in the transmission path. (This loss appears
twice-once in the path to the listener and again in the return path to
the talker.) Additional loss, however, causes a reduction in the received
talker volume. A loss administration plan called (rather arbitrarily) the
Via Net Loss (VNL) Plan is a compromise between echo and received
volume loss impairments. To control echo, the VNL Plan contains a loss
component that depends on trunk length and trunk type. The trunks
involved in longer connections where echo is a problem are in the toll
portion of the network. They include intertoll trunks, toll connecting
trunks, and intertandem trunks. Table 6-2 lists the loss objectives for
these longer trunks, and Table 6-3 shows the VNL component versus
length for high-usage intertoll trunks provided on carrier facilities.
Under the VNL Plan, whenever the echo path delay of a built-up connection exceeds 45 milliseconds, the 4-wire intertoll trunk that is part of
this connection is operated at O-dB loss and is equipped with an echo
suppressor or echo canceler. This is also done when a high-usage intertoll carrier trunk is longer than 1850 miles or would require a VNL
greater than 2.9 dB, as indicated in Table 6-3.

227

ECHO PATH DELAY (ma)

1.5

99.99
99.98

L

/
/

w

III
Q

60.00
SO.OO
40.00
30.00
20.00

~
~

zw
CJ

II:

/

/

I

/

/

/

/

I

I

~

/

V'/

I

L

o

L

I

L

L
L

10

I

100

/

/

L

300

V /*~

L

I

L

LL

ILLL

/

//

I"L L

~

600
1200

L L LL
I

II

I

L

L
L1f

20

/

/

/.
//

//

/ / //
I

I

I

I///.

/ ///
I

II

V/

//

I

/

/1/

//

/

V /

//
1"/

I

V/

/

//

LI

/

/

/

I

//

LL

/ //

/
/

V / / !///

0.01 .,.

LL lL LLV

V VL LL/

V

/

5.00

/

/ / 1/ /

/
I

L
L

/

L

/

/

L

I

V V~ VV/
VV VV
V LL lLLL L L 'L

/
/

3040 60 80
L

/

L

L

V
/

10.00

1.00
0.50
0.20
0.10
0.05

L

/

/

/

/

W

Q.

/

/
/

70.00 V

0
0

/

/

90.00
85.00

II:

0

20

L

/

~

/

II:

~

L

I

99.90
99.80
99.70
99.00
98.00

w

10

5

/

/.

LL

30

40

50

60

TALKER ECHO PATH LOSS (dB)

Figure 6-19. Grade of service versus talker echo path loss and delay.
Grade of service is determined by observing the percent good or better
(horizontal lines) at the intersection of the echo path delay (oblique lines)
and echo path loss (vertical lines) for a given connection. With satellite
communications, for example, a customer's signal travels over 45,000 miles
from talker to listener. The echo signal travels twice that distance, resulting
in an echo path delay of about 500 milliseconds. To achieve a grade of
service of 90 percent good or better with so great a delay would require a
talker echo path loss of about 57 dB (asterisk [. D.

An echo suppressor is an electronic device that compares the speech signals traveling in the two directions during a long-distance conversation.
The suppressor decides which person is talking at any given time and
inserts a high loss (35 dB or more) in the circuit in the opposite direction.
This prevents the echo from looping back to the talker over the listener's
speaking path. When the two persons talk at the same time, the suppressor inserts a loss in both directions, resulting in undesirable clipping of
. speech signals.
An echo canceler, rather than inserting a loss in the return path, uses
the transmitted speech signal to generate a signal that is a replica of the
echo. Subtracting this signal from the actual echo signal cancels out the
echo, while allowing normal communications to continue undisturbed.
The echo canceler principle is illustrated by the simplified diagram in
Figure 6-20.

228

TABLE 6-2
LOSS OBJECTIVES FOR CONTROL
OF ECHO ON TRUNKS
Loss
(dB)

Trunk Type

Nominal

Maximum

Intertoll

VNL*

2.9 (high usage)
1.4 (final groups)

Toll connecting

VNL + 2.5

4.0

Intertandem

VNL

1.5 (balanced offices)
0.5 (unbalanced offices)

Interregional direct

VNL+6

8.9

*VNL = VNLF x I-way length in miles + 0.4 dB. VNLF, the via net loss factor, depends on
the type of facility used. For example, VNLF = 0.0015 for carrier facilities.

TABLE 6-3
VNL COMPONENT VERSUS LENGTH
(OPERATING ON ALL CARRIER FACILITIES)
Trunk Length
(Miles)

VNL*

0-165
166-365
366-565
566-765
766-965
966-1165
1166-1365
1366-1565
1566-1850
Any length with echo
suppressor or canceler

0.5
0.8
1.1
1.4
1.7
2.0

*VNL

=

0.0015 x I-way length in miles

(dB)

2.3
2.6
2.9
0.0

+ 0.4 dB.

ECHO CANCELER
SPEECH
SIGNAL

SPEAKER

/
_J

/

/

/
LISTENER

ECHO
SIGNAL

Figure 6·20. Echo canceler principle.

The current echo objective is for performance to be satisfactory on 99
percent of all telephone connections that encounter delay. The VNL
Plan, hybrid balance, and echo suppressor/canceler administrative rules
are currently being used in the analog portion of the Bell System network to meet this objective.
Message Circuit Noise
Message circuit noise is a complex signal, comes from a variety of sources,
and has a great influence on the transmission quality of the network.
Some of the components are induced power hum, or inductive interference (discussed below); thermal noise}5 central office battery noise (see
Section 12.2); and impulse noise that is generated by central office switch
contacts.
For purposes of transmission planning and design, message circuit
noise is defined as a weighted average of the noise level from a variety of
sources within a voice channel as measured by a 3A noise-measuring set
or equivalent, using a frequency weighting called C-message weighting.
Measurements are expressed in dBrnC, decibels above reference noise
(10- 12 watts), using C-message weighting. The noise meter is designed so
that a I-kHz tone having a power of -90 dBm 36 will read 0 dBrnC.
The setting of noise objectives is based on the grade-of-service concept, the same concept applied to loudness loss (a measure of perceived
speech volume) and echo impairments. Subjective tests were used in
which speech was evaluated in the presence of noise. Observers were

35 Random noise related to the operating te'mperature of electrical circuit elements.
36 The unit dBm is a logarithmic measure of power with respect to a reference power of 1
milliwatt.

230

231

Transmission

Chap. 6

asked to rate a series of simulated telephone calls having different combinations of noise level and acoustic-to-acoustic loudness loss. The rating
was in terms of five quality categories: excellent, good, fair, poor, and
unsatisfactory. The results of these tests are shown in Figure 6-21.
30~----____~----~-.~--~----r-~----~-r------~

25

iii'
~

III
III

0

...J

III
III
W

20

Z
C

;:)

0

...J

Z

0

i=
w

15

(J

z
z

0

(J

10

15

20

25

30

35

40

45

MESSAGE CIRCUIT NOISE (dBrnC)

Figure 6-21. Loss-noise grade of service.
Each curve represents
combinations of loudness loss and message circuit noise for which a
constant percentage of observers rated the transmission quality good or
better.

Subjective reaction varies with both the level of circuit noise appearing at the terminals of the telephone set and the loudness loss. For
example, when the message circuit noise level is less than about 25 dBrnC
and the connection loudness loss is greater than 7 to 8 dB, the percentage
of test subjects judging the performance as good or better depends more
on loudness loss than on the noise level. For somewhat higher noise levels, the grade-of-service contours show more of a one-to-one relationship
in dB between the effects of loudness loss and noise level (that is, both
are equally responsible for the result). This indicates that over this range,
subjective preference is roughly constant with a constant signal-to-noise
ratio. 37 Finally, when the loudness loss decreases below 7 to 8 dB, an
increasing percentage of test subjects finds the high received talker
volume itself increasingly objectionable, regardless of the circuit noise
level.
37 On the figure, lines with a slope of -1 would represent constant signal-to-noise ratio.

Network and Systems
Considerations

232

Part 2

To ensure high customer satisfaction in terms of a loss-noise grade of
service at a reasonable cost, separate noise objectives have been established for trunks and loops. For trunks, the performance objectives
recognize the fact that circuit noise on analog facilities accumulates with
distance, as shown by the slanted solid lines in Figure 6-22. Trunk noise
objectives are summarized in Table 6-4. Actual noise performance should
be' better than or equal to that shown. Consistent with these overall
objectives, separate allocations have been made for trunks on short-haul
and long-haul carrier systems (see Figure 6-22).
The message circuit noise objective for customer loops is stated in
terms of a maximum or upper limit. Changes in noise level below about
20 dBrnC measured at the station set were found to have a negligible
effect on the overall customer-to-customer connection performance grade
of service, whereas the grade of service deteriorated appreciably when
noise exceeded this value. A loop limit of 20 dBrnC, therefore, was
chosen for all customer loops, including loop carrier systems.
NOISE INCREASES
3 dB PER
DOUBLE DISTANCE

Q:"
..J
~

40

o
~

c:(

oc

en

~

I

37
34

I
I

31

GRADE OF SERVICE: ,
95% GOOD OR BETTER

UJ

>
i=

o

I

UJ
...,

!Xl

o

I

I
I

28~----e~

UJ
If)

(5

STANDARD
DEVIATION =4 dB
FOR ALL ILENGTHS

-,

I

I

I

1000

2000

4000

CARRIER

UJ

I

I

I

- .....--+-- LONG-HAUL

z
zc:(

I

:::IE

20

40

60

120

250

500

TRUNK LENGTH (ROUTE MILES)

Figure 6-22_ Message circuit noise objectives for trunks on analog
carrier systems. Solid lines for short-haul and long-haul carriers reflect
noise objectives for a grade of service of 95 percent good or better.
Transmission level points (TLPs) define relative signal or noise levels at
various points in a telephone connection. Section 8.6.1 contains more
information on TLPs.

Intelligible Crosstalk

Intelligible crosstalk is a single, unwanted and understandable speech
signal coupled from one message channel intQ another. Coupling of any
type of signal between channels is a concern, but this interference is

Chap. 6

Transmission

233

TABLE 6-4
MESSAGE CIRCUIT NOISE OBJECTIVES FOR
CUSTOMER-TO-CUSTOMER TRUNK CONNECTIONS
(EXCLUDING LOOPS)
of
Toll
Connection

Distance
Bands
(Miles)

Percent
Good or
Better

Average
Noise
Power
(dBrnC)

Maximum
Standard
Deviation
(dB)

Short
Medium
Long
Intercontinental

0-180
180-720
720-2900
>2900

99
97
95
90

19
23
27
30

6
5
4
3.5

Typ~

particularly objectionable because, in addition to impairing the quality of
the channel, it creates a sense of loss of privacy. Crosstalk is caused by
nonlinearities in circuits38 within frequency-division multiplex (FDM)
carrier systems and by electric and magnetic coupling between disturbing
and disturbed circuits within various transmission media. Methods used
to control crosstalk include shielding conductors, separating disturbing
and disturbed circuits, impedance balancing, and designing and maintaining systems to suppress nonlinearities.
Stringent objectives have been established to minimize the probability
that customers will encounter intelligible crosstalk. These objectives are
expressed in terms of a crosstalk index, which is the actual percent probability of receiving an audible, intelligible speech signal on a call. This
index depends on such factors as the number of disturbers, talker
volumes, how much disturbing circuits are used, the coupling between
disturbing and disturbed circuits, the listener's acuity in the presence of
noise, and the noise level at the listener's telephone set. Table 6-5 lists
typical objectives for trunks and loops. The objectives take into account
the fact that exposure to crosstalk is generally greater on the longer intertoll connections than on shorter local trunks.
Delay
Delay is the transmission time required for a talker's speech signal to
reach the listener. Subjective tests have shown that delay has no major
effect on speech transmission if customers are not anticipating it, and it is

38 For circuits with nonlinearities, the output signal amplitude is not directly proportional
to the input signal amplitude.

Network and Systems
Considerations

234

Part 2

TABLE 6-5
INTELLIGIBLE CROSSTALK OBJECTIVES
Trunk/Loop Type

Objective
Crosstalk Index
(%)

Intertoll trunks
Toll connecting trunks
Direct trunks
Tandem trunks
Intertandem trunks
Loops

1.0
0.5
0.5
0.5
0.5
0.1

kept to 600 milliseconds or less one way. Once customers become aware
of a long delay on a channel, however, they tolerate it less.
Terrestrial facilities within the United States may introduce about 20
milliseconds of I-way delay depending on the length and the velocity of
propagation of the transmission facility. Intercontinental connections
using submarine cable may have I-way delays of 100 milliseconds.
Geosynchronous satellite circuits introduce considerably larger delays (on
the order of 250 milliseconds one way); therefore, current delay objectives
restrict the use of satellite trunks to no more than one up-down link per
end-to-end connection.
Amplitude/Frequency Distortion
Amplitude / frequency distortion occurs when the relative magnitudes of
the different frequency "components of a signal are altered during
transmission over a channel. The graph on the left in Figure 6-23 illustrates the amplitude response of a voiceband channel under preferred
conditions. When conditions are significantly poorer than this, the
amplitude response may be distorted as suggested by the graph on the
right in Figure 6-23, where amplification is not the same for different frequencies across the channel bandwidth.
As mentioned in Section 6.1.1, an acceptable bandwidth for a voice
signal is from 200 Hz to 3.5 kHz. Although a consistent set of bandwidth
objectives for individual loops and trunks in the network has not been
established, the factors that tend to increase the effective voiceband channel bandwidth also help make the amplitude response within the band
more uniform. Amplitude distortion is controlled primarily by design
rules for loops and trunks, office c~bling limitations, and design requirements for carrier system channel units (see Section 9.4.3) and other multiplex arrangements.

LU

LU

I/)

I/)

Z

Z

oD.

oD.

LU

LU

I/)

I/)

a:

a:

200 Hz

3.5 kHz

200 Hz

FREQUENCY

Figure 6-23. Amplitude/frequency responses.
right, poor channel.

3.5 kHz

FREQUENCY

Left, preferred channel;

A single inband amplitude distortion objective for speech signals is
not available because of difficulties in expressing an objective that can be
allocated optimally to trunks and loops.
Carrier Frequency Shift
Carrier frequency shift affects speech and program signals by reducing
the naturalness of received signals. It results when carrier frequencies of
transmitting and receIvIng analog transmission terminals using
suppressed-carrier FDM (see Section 6.4.2) differ. Subjective tests have
shown that customer-to-customer frequency shift should be held to a
maximum of ±2 Hz to preserve a harmonic relationship that is satisfactory to discerning listeners. This objective has been selected as achievable in today's facilities, even for the longest connections through the
network.
Inductive Interference
Since the power and telephone industries serve the same customers, they
frequently share the same right-of-way (for example, the same utility
pole). Because of this proximity, telephone cables are exposed to electromagnetic fields created by the power system currents. Under ideal
conditions, this power influence results in equal voltages to ground on
the two conductors of a wire pair. If the power influence is high enough
and if the telephone circuit is not perfectly balanced, a voltage difference
between the conductors will be produced that can result in audible noise
levels. The standard telephone set is designed to have relatively low sensitivity at the fundamental frequency of power systems in the United
States (60 Hz). Distortion of the power system current waveform, however, can generate harmonics of 60 Hz of sufficient amplitude to cause
interference in the voice band. 39
39 As mentioned earlier in this section, the Bell System objective is for noise not to exceed
20 dBrnC on customer loops.

235

236

Network and Systems
Considerations

Part 2

The Bell System employs various methods to minimize the possibility
of inductive interference. These include using adequately balanced telephone cable with grounded sheaths, employing balanced telephone
equipment, and separating power and telephone circuits. Cooperative
efforts between the Bell System and the power industry, known as inductive coordination, may also be required to achieve satisfactory operation of
telephone equipment. When total compatibility is impractical or impossible to attain using these measures, it may become necessary to install
mitigation devices on telephone facilities. At present, these consist of a
variety of magnetic core devices,4o but they are used only as a last resort
because they are expensive, bulky, and incompatible with certain systems.
Overload
When a signal is transmitted at an amplitude higher than that normally
imposed on a telephone channel, it can produce intelligible crosstalk or
interference. This impairment is considered overload when all channels
in a transmission facility are affected. To control overload, a limitation is
set on the amount of power that may be applied to an individual channel. The objective for both system load and customer load is that the
per-channel long-term average power not exceed -16 dBmO.41 This objective considers signal level variations with time and channel activity.
6.6.2 DIGITAL DATA SIGNALS
Transmission of significant amounts of digital data over the PSTN began
about 1960 and is now a sizable portion of the traffic handled. Originally, data communications were carried only over analog transmission
facilities using data sets. However, as digital transmission facilities and
digital switching machines increase in number and connectivity, data signals can be transmitted in digital form more economically, generally at
higher data rates and with potential advantages in transmission quality as
well.
Digital data signals are generally transmitted in blocks containing
error-detecting codes so that, when errors occur, the data may be
retransmitted. Alternatively, the use of error-correcting redundant codes
permits some or all errors to be corrected at the receiving end of
transmission. Both techniques reduce or neutralize the effect of errorproducing impairments. Because loss and noise are controlled by objectives for speech, specific objectives for data are confined to impairments

40 Some of these devices are neutralizing transformers, drains, and chokes.
41 The unit dBmO is an expression of power level in decibels with reference to a power of
o milliwatt.
.

Chap. 6

Transmission

237

such as impulse noise. In addition, there are service objectives that
address the efficiency of data communications in terms of throughput.42
These service objectives focus on transmission performance, availability,
and maintainability.
For example, objectives for the Digital Data System (005)43 are:
The transmission performance objective is error-free transmission in
at least 99.5 percent of all I-second intervals. System performance is
measured in error-seconds, where an error-second is a I-second interval
during which one or more bit errors occur.
• The availability objective is that data circuits remain connected station
to station and error free 99.96 percent of the time as a long-term average. In terms of cumulative outage time, this objective equates to an
average down time of less than 3.5 hours per year.
• The maintainability objective is that no single outage exceed 2 hours.
This objective recognizes the impact of long outages on a customer's
business and the perishable nature of some data.
The rest of this section discusses major transmission impairments that
can affect data service over analog channels-impulse noise, linear distortion, nonlinear distortion, and incidental modulation.
Impulse Noise
Impulse noise consists of short-duration high-amplitude bursts (or spikes)
of noise energy, which are much greater than the normal peaks of message circuit noise on a channel. An impulse is considered to have
occurred if the noise voltage increases by at least 12 dB above message
circuit noise for no more than 10 milliseconds. If impulse noise occurs
often enough, it can obliterate large blocks of data and, in extreme cases,
render a channel unsuitable for data transmission. Impulse noise sources
include voltage transients generated by electromechanical switching systems, lightning, switching to protection channels,44 and various maintenance activities.
Objectives for impulse noise are stated in terms of the maximum
number of times noise above a particular threshold value occurs during a
prescribed measurement interval. The present customer-to-customer

42 The quantity of data a user can transfer during a given time interval.
43 Section 11.6.1 discusses DDS.
44 Broadband channels in carrier facilities used as spares. They can be switched into service
if normal working channels fail.

Network and Systems
Considerations

238

Part 2

objective for the switched network is for no more than fifteen impulse
noise counts in 15 minutes at a threshold set 5 dB below the received signal level for 50 percent of all connections in the Bell System.
Linear Distortion
Linear distortion is a signal-independent impairment associated with a
non ideal transmission channel characteristic. It is caused by transmission
networks that produce a nonideal inband amplitude and phase response.
To prevent the dispersion of pulses in a digital data stream, the channel
characteristic ideally should have a bandwidth as wide as possible, a flat
amplitude / frequency response, and a linear phase-versus-frequency45
response within the channel. Otherwise, intersymbol interference may
occur (the dispersion of pulses may cause a pulse to be detected where
none exists), resulting in unacceptable error rate performance. Figure 624 shows an ideal pulse stream versus a pulse stream where dispersion
has occurred.
PSTN objectives for linear distortion have been applied to loops and
special-services circuits conditioned for data transmission (that is, specially amplitude and phase equalized), for example, DATAPHONE service46 for data speeds above 300 bps.
In this case, the
amplitude / frequency distortion objective is that the loss at 2.8 kHz and at

o

o

o

o

o

IDEAL PULSE STREAM

DISPERSED PULSE STREAM

Figure 6-24. An ideal pulse stream versus pulse dispersion.

45 With a linear phase-versus-frequency response, all frequency components of the
transmitted signal experience the same delay.
46 Section 2.5.4 discusses DATAPHONE service.

Chap. 6

Transmission

239

400 Hz not exceed the loss at 1 kHz by more than 3 dB. Moreover, the
phase/frequency distortion objective permits no more than 100
microseconds delay difference between any two frequencies in the band
from 1 to 2.4 kHz.
A simple method of indirectly measuring the linear distortion of a
channel is the peak-to-average ratio (P / AR) meter method. To make a
P / AR measurement, a representative pulse train is applied at the
transmitting end of a channel, and the ratio of the pulse envelope peak to
the envelope full-wave average at the receiving end is detected. This
measures the pulse dispersion produced by the channel. Amplitude and
phase distortion in the transmission channel tend to disperse the energy
in each pulse, thereby reducing the peak-to-average ratio. The current
objective for customer-to-customer connections requires a P / AR value of
at least 48, on a scale of 0 to 100. A P / AR of 50 or more usually indicates
that intersymbol interference will be acceptable for 2.4 kbps data.
Although the P / AR method is simple to apply in the field, it has the
property of responding to linear distortion and other impairments that
produce pulse dispersion. It may be applied for assessing delay distortion
when that is. the predominant impairment on a channel.

Nonlinear Distortion
Nonlinear distortion (where the amplitude of the output signal from a
channel does not have a linear relationship to the amplitude of the input
signal) produces intermodulation products of the transmitted signal that
fall within the channel bandwidth. (See Bell Laboratories 1982, pp. 389401.) The amount of distortion may vary with time, even during a call.
It is most often introduced by such transmission components as multiplex
terminals, voice-frequency amplifiers, and station sets.
Nonlinear distortion within a channel is measured using a 4-tone test.
Two pairs of equal-level tones (856/863 Hz and 1374/1385 Hz) with a
composite power of -13 dBmO are applied to the transmitting end of the
channel, and second- and third-order products formed by nonlinearities
within the channel are measured at the receiving end. The -13 dBmO
power level corresponds to the maximum power limit permitted for
voiceband data averaged over a 3-second interval. This is roughly
equivalent to a long-term average speech signal power of -16 dBmO. The
overall customer-to-customer objective for nonlinear distortion is that
second- and third-order modulation products should be at least 27 and
32 dB, respectively, below the received composite signal power.

Incidental Modulation
Incidental modulation is any unwanted amplitude or phase modulation
appearing on a received data signal. Its effect on transmission is similar
to that of impulse noise. When the variations in amplitude or phase are

240

Network and ~ystems
Considerations

Part 2

small, the impairment is called jitter. Larger changes in amplitude and
phase are called gain hits and phase hits, respectively. Both jitter and hits
involve rapid amplitude and phase changes, which may be either
periodic or random.
Phase jitter, which perturbs the zero crossings of a data signal, tends
to be more serious than amplitude jitter and can often be traced to an ac
power hum on carrier frequencies and power supplies associated with
FDM terminal equipment. Synchronization and timing signals as well as
data signals are subject to phase jitter resulting in higher error rates. The
current customer-to-customer objective restricts the maximum peak-topeak phase modulation to 8 degrees for modulating frequencies between
4 and 300 Hz.
Gain and phase hits are produced in the network in several ways
including automatic protection switching of radio channels during fading
and manual switching of transmission facilities or multiplex equipment
from working to standby status for maintenance work. Switching a
transmission facility can cause a direct hit on the transmitted signal
because of the phase or attenuation difference between the working and
the standby facility. If the switch occurs within synchronizing equipment or on a facility carrying a synchronizing signal, a hit will affect the
synchronizing signal and, in turn, several data signals. Gain hits of 3 dB
or more begin to cause errors in data transmission. Similarly, phase hits
of 20 degrees or more are likely to cause errors. The current end-to-end
objectives are for no more than eight gain hits of 3 dB or more or eight
phase hits of 20 degrees or more in a IS-minute interval.

AUTHORS

A. O. Casadevall
w. C. Roesel

7
Switching

7.1 INTRODUCTION
The primary function of switching in a telecommunications network is to
interconnect the telephones or other telecommunications terminals of a
large number of customers as economically as possible. As shown in Section 3.1, centralized switching requires only one telephone and one telephone line per customer; the interconnection of n customers without
switching requires n-1 telephones per customer and a total of [n(n-1)]!2
wire pairs. Even with centralized switching, as n and the area served
continue to grow, points are reached where using two or more switching
systems interconnected by transmission facilities becomes the most
economic way to provide telecommunications access for all n customers.
The public switched telephone network (PSTN) extends the concept of a
network of centralized switching systems to a very large geographic area
through a hierarchical network of switching offices (see Section 4.2.1).
Although connection is the primary function of switching, other functions must also be provided, as discussed in Section 7.2. Some of these
functions are associated with the operation of the telecommunications
network and others are associated with customer services.
Interest in improved customer services has strongly influenced the
design of modern switching systems that use stored-program control, a concept that enables new features to be implemented through changes in
software rather than hardware. Section 7.4.3 describes stored-program
control. Chapter 10 describes the various switching systems, including
those that use stored-program control.
Traditionally, Bell Laboratories has been responsible for systems
engineering of switching systems for both network and customer applications on behalf of AT&T and the Bell operating companies. These
include systems developed by Bell Laboratories and manufactured by
Western Electric, as well as systems from other suppliers.
241

l"~{WUrK

242

anu

;:)y~H~lll:;

Considerations

Part 2

The rest of this chapter provides a basic understanding of some of the
concepts and considerations involved in the design of switching systems.

7.2 BASIC SWITCHING FUNCTIONS
Connection can best be understood by considering the two essential parts
of a switching system: the switching network and the control mechanism. The switching network consists of the individual switching devices
used to connect the communication paths. The control mechanism provides
the intelligence to operate the appropriate switching devices at the
proper time. (Sections 7.3 and 7.4 discuss switching networks and control
mechanisms, respectively.)
In addition to the basic function of connecting communication paths,
a switching system must be capable of receiving and sending network
control signals between itself and customer terminals and other switching
systems, performing functions required to administer and maintain the
system, and providing customer services. Each of these switching functions is described below. These additional functions are many and complex and must interact properly, compounding the work of switching system designers.

7.2.1 CONTROL

Control is the technique by which a switching system interprets and
responds to signals and directs the switching network. In the past, control was accomplished by logic circuits using relays and other electromechanical devices. Today, virtually all new switching systems
employ stored-program computer control. By changing and adding to the
stored program, the operation of the switching system can be modified
and extended.

7.2.2 SIGNALING

All modern switching involves the transfer of information (for example,
dialing and ringing) between users and the switching system and
between two switching systems. This is known as signaling and can be
thought of as a special form of data communication.
In early automatic switching systems, most signaling between systems
involved dc electrical signals. Later, as signaling distances increased,
single- and multiple-frequency tones were used. Recently, faster and
more versatile digital signaling over dedicated networks separate from
the voice network has been introduced. (Chapter 8 discusses the signals
required in the network and the various signaling techniques.)

Chap. 7

Switching

243

7.2.3 ADMINISTRATION AND MAINTENANCE
Today's switching systems provide separate features to ensure that the
switch operates reliably and efficiently. These features monitor, test,
record, and permit human control of service-affecting conditions of the
switching system. Examples include:

• network management, which enables traffic to be rerouted to avoid congested portions of the PSTN (see Section 5.6)

• traffic measurement, which provides indications of the traffic loads
being carried by various components of the switching system (see Section 5.7)

• billing, which allows recording of call-related information required to
charge properly for service (see Section 10.5)

• maintenance, which involves features that automatically detect, isolate,
and often locate system and component troubles to within several
plug-in circuit packs (see Section 13.3.3).
Today, much of the operating data are reported by the switching system
to operations systems that collect, analyze, filter, and summarize the data
for human use. Chapter 14 describes some related operations systems.

7.2.4 CUSTOMER SERVICES
In addition to connecting communication paths, modern switching systems also provide a variety of customer services. As described in
Chapter 2, customer services are enhancements to basic interconnection.
They include operator services, coin services, Custom Calling Services,
and special business features such as centrex service. Some examples of
switching system functions related to these services are:
• routing a call for a nonworking number to an intercept operator
• returning deposited coins at a coin telephone when the called party
does not answer
• routing a call to a line other than the one dialed (Call Forwarding)
• identifying the calling line for billing purposes on outgoing calls.

7.3 SWITCHING NETWORKS
The network is the portion of a switch that provides the connection
between communication channels (lines or trunks) terminated on the
system. Traditionally, switching networks are made up of connective
devices or circuits arranged in a structure that allows the simultaneous

244

Network and Systems
Considerations

Part 2

connection of many pairs of communication channels. This mode of
switching is known as circuit switching, denoting the dedication of circuits
to each connection for the duration of the call. Other forms of switching
are currently in use for data communications (see Section 11.6), but the
telephone networks of today are virtually all circuit switching.

7.3.1

CIRCUIT-SWITCHING NETWORK TYPES, APPLICATIONS,
AND TECHNOLOGIES

Two types of circuit-switching networks in use today are distinguished
by the manner in which the information passes through the network. In
space-division networks, the message paths are separated in space, as
described in Section 7.3.2. In time-division networks, the message paths
are separated in time, as described in Section 7.3.3.
As discussed in Chapter 4, switching systems are used in a variety of
applications throughout the network. For local service (class 5 operation), 2-wire switching is sufficient. In 2-wire switching, one message
path (the equivalent of two wires in a space-division switching network)
simultaneously carries the transmitted and received information. For toll
operation (classes 1, 2, 3, and 4), however, the need for more stringent
control of transmission impairments (such as echo) due to the longer distances involved generally requires 4-wire switching. In space-division
switching networks, 4-wire switching is provided over two message
paths, one for each direction of transmission. As will be seen in Section 7.3.3, time-division switching systems are inherently 4-wire since the
information flow in each direction is switched separately.
Three basic types of technology have been used to implement switching networks:
• The manually-operated switch where an operator places plug-ended
wires (cords) in jacks is the oldest.
• Electromechanical switches are either motor-driven, gross-motion
devices (such as rotary and panel switches), gross-motion devices
driven by electrical impulses (such as step-by-step switches), or electromagnetically operated, fine-motion switches (such as crossbar and
dry reed matrix switches). Gross-motion switches have inherent limitations in their operating speed; and they use common metal (for
example, copper) contacts to withstand the wear of the sliding or wiping motions, which may, in time, result in noisy transmission paths.
Also, the considerable wear on the contacts causes maintenance costs
to be high. In fine-motion switches, contact motion is essentially perpendicular to the contact surfaces, resulting in much less wear. Therefore, precious metal (for example, gold) can be used to improve
transmission quality.
(Section 10.2 describes electromechanical
switches in detail.)

Chap. 7

Switching

245

• Electronic switching elements are usually made from semiconductor
devices. The evolution of semiconductor technology has reduced the
size, power consumption, and cost of electronic switching elements
and, at the same time, increased their operating speeds, ruggedness,
and reliability. This has made the use of electronic networks in
switching systems practical. In particular, their operating speeds,
which are orders of magnitude faster than electromechanical switching elements, are necessary for their use in time-division switching
networks. These networks make the most efficient use of electronic
switching elements in terms of the number of e~ements required.
Because electronic switching devices tend to have less ideal transmission characteristics than metallic contacts, electronic switching networks that handle digitally encoded (for example, pulse-code modulation 1) signals are also easier to design than those that handle analog
signals. With the decreasing cost of digital coding circuits and the
increasing use of digital transmission systems, most new circuit
switching systems employ time-division networks to switch digitally
encoded speech and data.
7.3.2 SPACE-DIVISION NETWORKS

The most common switching network in use today in the Bell operating
companies is the space-division network. In space-division networks, a
physical, electrical, spatial link is established through the network to connect terminated lines and trunks. Space-division networks are constructed of stages of metallic devices or semiconductor network elements.
The electrical characteristics of the connecting links are generally suitable
for carrying message signals requiring wide bandwidth or considerable
power. The connecting path through the network is maintained for the
duration of the message, and either analog or digital information (within
the bandwidth of the network) may be passed.

Network Topology
The topology of a network describes the pattern in which the elements of
the network are interconnected to allow input terminals to have access to
output terminals. Most space-division networks involve concentration,
distribution, and expansion as stages in switching. They are designed to
make the required interconnections with little probability of blocking,
while minimizing the number of switch elements in the network. The
network must also be designed to be changed in size conveniently to
adjust to changing demands in terminals served and traffic carried.

1 Section 6.4.3 discusses pulse-code modulation.

246

Network and Systems
Considerations

Part 2

Figure 7-1 illustrates a 4-point rotary switch. By directing the wiper
(the moving rotary contact) to the proper position, any of the four customer lines can be connected to the rest of the network via the output
link. While one of the lines is in use, however, the others are blocked.
The usage of the four lines is concentrated on one output link. More paths
and less blocking can be provided by multiple appearances of the same
lines with access to two output links as shown in Figure 7-2. This allows
two simultaneous conversations, but at the cost of an added switch and
increased control complexity. (Some means must be provided to choose
an idle concentrator switch for the second customer and to ensure that
both switches do not inadvertently serve one customer line simultaneously.) With four inputs and two outputs, the concentration ratio is 2 to
1. An output link is in use, on the average, twice as much as an individual line.
Once the traffic has been concentrated, the switch must be able to distribute the traffic to the proper output terminals of the network. An
example of distribution is shown in Figure 7-3 where each input has
access to each output terminal. A network (or portion of one) that provides this function is known as a distribution network. In contrast to concentration networks, in which the ratio of inputs to outputs is greater
than one, distribution networks have a ratio of one.
The ratio of inputs to outputs in one portion or stage of a network
may also be less than one. This effect is the opposite of concentration
and is known as expansion (see Figure 7-4). Expansion serves the same
function for outputs that concentration serves for inputs. (It allows the
distribution paths to operate at higher occupancies than that of the outputs.) Even when the outputs are trunks, they may operate at lower
occupancies than the distribution paths. This may occur when the output
terminals consist of many small groups of trunks connecting to different
offices. It could also occur if groups of output terminals experience their
busy hour of traffic at different times of the day.
The control of space-division networks falls into two general
categories: progressive control and common control. In progressively
controlled networks, a call progresses sequentially through the stages of
switching. There is no capability to "look ahead," so the call may be
blocked at a later stage. With common control, all stages of the network
are examined simultaneously, and usually, a number of possible paths
between the end-points are examined. This reduces the possibility of
blocking in the network.
Progressively Controlled Networks
Setting up a path through a switching network with progressive control
can be demonstrated using Figure 7-4, which represents a step-by-step
network. First, the wiper of an idle switch in the concentration stage

CUSTOMER
LINES

n-----;;~---

OUTPUT LINK

Figure 7-1. A 4-point rotary switch.

CONCENTRATION

CUSTOMER
LINES

OUTPUT LINKS

Figure 7-2. Concentration stage.

CON CENTRA TION

DISTRIBUTION

OUTPUT
TERMINALS

Figure 7-3. Concentration and distribution stages.

CONCENTRA TION

DISTRIBUTION

EXPANSION

TRUNKS
TO
OTHER
OFFICES

Figure 7-4. Concentration, distribution, and expansion stages.

must be moved to the input of the calling customer. Then, the connecting switch in the distribution" stage must be moved to select the proper
path to the expansion stage, where again a one-out-of-four choice must be
made. Section 7.4.1 discusses control of such a network, in which the
paths are established one stage at a time. Although a step-by-step network was used in this example, progressive control has also been
em ployed in crossbar switching systems. "
Progressively controlled networks have two inherent disadvantages:
• The blind progression of the connection through the network may not
find an available complete path because all possible choices of connecting paths are not examined.
• They can be used to set up paths in one direction only. This implies
that each customer must have two terminations, one for originating
and one for terminating traffic.
Coordinate Networks
A coordinate switch consists of an array of contacts or crosspoints arranged
in a matrix through which several inputs can be independently connected to several outputs. (As mentioned earlier, rotary switches are
one-to-several or several-to-one devices.) Consequently, several commuunication paths can exist simultaneously in a coordinate switch.
Figure 7-5 represents a coordinate switch (typkally used in crossbar
and reed matrix switches). The XS located at the intersections of the horizontal and vertical lines represent the crosspoints, which may be metallic
248

249

Switching

Chap. 7

contacts, diodes, transistors, etc. In a coordinate network, generally only
one crosspoint on any row or column can be in operation at any time.
(Figure 7-6 shows the coordinate switch equivalents to rotary switch
networks.)

INPUTS

()

()

,... ,...

I'
-"I

I'

I'

.

,
,

,"
,,/

~

~

,/

"

,

II

q

"

"

JI'

()

,/

()

()

()

~

1/

'1'

'1'

,~

,~

'1'

,

()

Q

~

...

,,~

,~

~

I'

I'

...,

~

...

,,/
I'

"

"

"

,

,

,

,,, "

...

,

... ,

"

"

,"

~

... ~

II

~

,, "
r,

I'

'1'

'1'

,1/

,~

,r,

I'

I'

JI'

'1'

'I'

,
...

I'

"

""

,,~

'1'

"

,~

I'

II
I'

,
,

,~

I'

,I.-

,~

~

I'

I'

" I'

...

[,

[,

"

'1'

JI'

-

"

<)
,,/

'
,"

...

J'

,,

,

,'"

r,

"...,
,..

,

II

,, ""

-

"

...

-

"

o
U

T
P
U
T
S

,J')

-

"

,J')

Figure 7-5. A 9-by-9 coordinate switch.

Figure 7-7 shows a coordinate network made up of several stages of
coordinate switches. It provides full access between the nine inputs and
the nine outputs and requires only fifty-four crosspoints as opposed to
the eighty-one required in the single switch of Figure 7-5. In Figure 7-7,
in put A can be connected to output B through the single link connecting
switch 1 to switch 6. However, once this connection is made, no further
connections between inputs on switch 1 and outputs on switch 6 are possible until the connection between A and B is released.
By introducing three more switches (a third stage of switching), as
shown in Figure 7-8, three paths (I, II, III) are available for interconnecting input A to output B. Each path consists of two links (for example, 11
and 12)' This network requires eighty-one crosspoints, but it is still not
nonblocking in contrast to the single 9-by-9 coordinate switch shown in
Figure 7-5.
Although no crosspoints are saved in the example in Figure 7-8, in
general, where the switching networks must handle between hundreds
and tens of thousands of inputs and outputs, multiple stages of coordinate switches will produce a large savings in the total number of

COORDINATE SWITCH

ROTARY SWITCH

INPUTS

~~

tt

OOUTPUT

A

INPUTS

~

OUTPUTS

B

OUTPUTS

c

Figure 7-6. Rotary switches and equivalent coordinate switches. A,
concentrator; B, multipled concentrator; C, concentration and distribution.

crosspoints (and, hence, the cost) compared with a single, large coordinate switch. For example, a 3-stage network of lO-by-lO coordinate
switches would have relatively little blocking with a 70-percent savings
in crosspoints over a single-stage lOO-by-lOO coordinate switch. The
choice of switch size and the number of stages required depends on the
expected range and the size of the network. Typically, lO-by-lO coordinate switches in network switching systems are found in a 4- to 8-stage
coordinate network.

7.3.3 TIME-DIVISION NETWORKS
In space-division switching networks, each call has its own physical path
through the network. In time-division networks, simultaneous messages
are separated by assignment to separate time slots. Time-division networks can be divided into two types, depending on the form of the signals switched. Pulse-amplitude-modulated switching networks switch
250

A

4

OUTPUTS

INPUTS

2

B

3

LINKS

6

Figure 7-7. A 2-stage 9-by-9 switch using six 3-by-3 switches.

pulse-amplitude-modulated samples2 of the signals; digital time-division
switching networks switch digitally encoded samples of the signals.
The type and format of the digital encoding of voiceband signals is an
important factor in the design of a digital time-division switching network. When the proper format is chosen, such networks can provide a
direct interface with digital transmission facilities carrying digital timedivision multiplex (TDM) signals. 3 Generally, with space-division networks, TDM signals must be demultiplexed (separated), and the separated
pulse-code modulation (peM) signals must be converted to analog form
by digital-to-analog converters before passing through the switching network. Both conversions and converters can be eliminated when digital
transmission facilities connect to digital time-division switching systems
that have the proper format. The use of digital transmission is growing
rapidly, and time-division switching networks offer significant economies
in parts of the telecommunications network with a heavy concentration

2 Section 6.4.3 discusses pulse-amplitude modulation.
3 Section 6.5 discusses TDM.

251

A

7

4

INPUTS

OUTPUTS

2

8

5

9

6

3

Figure 7-8. A 3-stage, 9-by-9 coordinate network using nine 3-by-3 switches.

B

253

Switching

Chap. 7

of digital facilities. Even where analog facilities predominate, new timedivision switching systems are more economical than the space-division
systems they generally replace because of the new technology they
employ.
Principles of Operation
In Figure 7-9, customers 11 to In are talking to customers 0 1 to On
through a simple time-division switch. The talking paths are opened
and closed by switching devices or gates labeled Al to An and Bl to Bn.
The switch connects the desired pairs of /customers by controlling the
operation of the selected gates. To connect 11 to O 2, gates Al and B2 are
closed during the same "time slot," and a sample of the speech signal is
carried over the common path (or bus). Customers 13 and 0 3 may be connected by closing gates A3 and B3 during another time slot. As discussed
in Section 6.4.3, if the speech signals between 11 and O 2 are sampled at
least 8000 times per second, the resultant pulse-amplitude-modulated signal provides satisfactory speech transmission. The time slot for 11 and O 2
must, therefore, be repeated at that rate, as must the rest of the n time
slots.
In a digital time-division switch, the amplitude samples are encoded
into PCM form, and the resultant sequence of binary digits is switched.
Lines or trunks into a switching system from digital facilities present
signals from multiple calls in a time-division-rnultiplexed format. The
signals are separated and switched by a time-slot interchange (TSI).

11

12

13

@
@
@
•
•
•

A1
B1

A2

@itr

01

°2

A3
B3

InQ_?

•
•
•

•
•
•

An

•
•
•
On

BUS
ELECTRONIC
GATE

Figure 7-9. A simple time-division switch.

Network and Systems
Considerations

254

Part 2

As shown in Figure 7-10, the TSI can be thought of as having storage
locations (stores or buffers) associated with specific time slots in the incoming TDM bit stream and the outgoing multiplexed bit stream for each
direction of transmission. A transfer (read-write) operation can change the
order of the time slots by changing the order of the information from the
input to the output stores. The transfer operation is determined by routing requirements and directed by the call processor. The transfer instruction for a given time slot is set for the duration of a call on that time slot.
A mirror transfer is performed for the other direction of transmission
through the TSI. It is worth noting that an n-channel TSI is equivalent to
an n-by-n space-division switch since the n inputs have full nonblocking
access to the n outputs.
The figure shows some samples from four calls that are in time slots 1,
2, 3, and n of the input multiplexed bit stream being routed through a
TSI. The binary information from that line is directed to locations 1, 2, 3,
and n of the input store. The routings for these calls have established
that the calls should be moved to locations r, r+l, 2, and 1, respectively,
of the output store and, thus, time slots r +1, r, 2, and 1 of the output bit
stream.
In the simplest form, time-division switching can be accomplished by
a switch consisting of one TSI, with n customers attached to the input
side and the same n customers attached to the output side. A given customer's loop could be associated with a specified time slot (by means of

INPUT STORE

INPUT MULTIPLEXED
BIT STREAM

••• 3 2 1 n

3 2 1

ONE PER
TIME SLOT

TRANSFER
LOGIC

'------v"...---~
INPUT TIME
SLOTS

OUTPUT TIME
SLOTS

CALL PROCESSOR

Figure 7-10. A time-slot interchange.

Chap. 7

Switching

255

multiplex equipment such as a channel bank).4 Then, by transferring samples from input time slot 3, for example, to output time slot 2, customer 3
could talk to customer 2. Likewise, on the receive side of the switch,
time slot 2 would be transferred to time slot 3 to allow transmission in
the other direction. In most applications, as described in the following
section, the TSI is used as one of the primary building blocks of a timedivision switching network.
The TSI shown in Figure 7-10 is a nonblocking network component,
since any time slot has access to any other unused time slot, and both an
input and output store are not necessary if there is flexibility in the order
of reading from the input store directly to the output multiplexed bit
stream.
Network Architecture
In many practical switching applications, a number of multiplexed input
and output bit streams appear at the switching system, requiring the use
of multiple TSIs. 5 Since the switching network must be able to switch any
input time slot to any output time slot as determined by call routing, it
must be capable of connecting the outputs of one TSI to the inputs of
other TSIs. This leads to the other basic building block of time-division
switching networks-the time-multiplexed switch (TMS). The TMS
operates as a very high speed space-division coordinate switch whose
input-to-output paths can be changed for every time slot to rearrange the
interconnection of successive time slots of TSIs.
The TMS consists of high-speed electronic switching elements (gates)
at the crosspoints of an m-by-m switching matrix as shown in Figure 7-11.
Each of the m inputs and outputs is a multiplexed bit stream from a TSI.
The shaded gates indicate the input/output connections during one time
slot. When the clock signals the start of the next time slot, the controller
changes the connections by sending control pulses to the appropriate
gates. A TMS is essentially an m-by-m space switch with a third
dimension-time. Instead of leaving the network paths up for the duration of a call as in a space-division switching system, a TMS is changed
for each of the n time slots in one TDM frame. Conceptually, it can be
viewed as n m-by-m space switches as shown in Figure 7-12, where each
m-by-m matrix switches one of the n time slots from each of the m multiplexed bit streams.

4 See Section 9.4.3.
5 A TSI can be made to operate n times as fast as the incoming bit stream and thereby
multiplex and switch n lines of incoming bit streams. However, a limit is reached in the
speed with which TSIs can operate, and multiple TSIs are then required.

INPUT
MULTIPLEXERS

Figure 7-11. A time-multiplexed switch configuration shown during one
time slot. Input lines 1, 3, and m are connected to output lines 3, 1, and m,
respectively.

2

2

•
••

••
•

m

m

Figure 7-12. An analog equivalent of a time-multiplexed switch. The
figure shows n different m-by-m switch matrices, one for each time slot of
the input multiplexes.

257

Switching

Chap. 7

Figure 7-13 illustrates one way of operating a TMS in conjunction
with TSIs in a time-division network. Since time-division networks are
designated according to the sequence of time and space switching stages,
the network shown is a time-space-time (TST) network, as are most digital time-division switches in the Bell System. The TMS network may
actually employ several stages of switching, such as in a time-spacespace-time (TSST) network. The figure shows the connection of two communications channels, each of which is represented by a time slot at one
of the TSIs. In the example shown, the directions of transmission have
been separated prior to arriving at the TSI. Samples from channel A
arrive in time slot 3 on TSI A. Samples from channel B arrive in time
slot 1 on TSI B. Each TSI has a mirror image (the shaded squares in the
figure) that handles the reverse direction of transmission. To connect
channels A and B, the central control establishes a path between TSI A
and TSI B during time slot 25 of the TMS. TSI A transfers channel A
samples to TMS time slot buffer 25. During time slot 25, the TMS connects the output buffer of TSI A to the input buffer of TSI B. TSI B then
transfers the samples to channel B's time slot, time slot 1. The other
direction of the transmission path progresses similarly and simultaneously on the other (shaded) halves of the TSIs, using the same time slot
of the TMS.

BIDIRECTIONAL
MULTIPLEXES

NETWORK
CONFIGURATION
FOR A PARTICULAR
TIME SLOT
TSI B
SAMPLE CHANNEL ROUTING

Figure 7-13. A time-space-time (TST) digital network.

258

Network and Systems
Considerations

Part 2

Other arrangements of time (T) and space (S) stages are in current use
in digital time-division networks. These include STS, TSTS, and SSTTSS.
As in space-division switching, time-division networks may use concentration and expansion to reduce the size of the switching network
fabric. This concentration may be done prior to the TSI or may be done
digitally as part of a TSI.
The number of time slots provided for in a time-division network
depends on the speed of the technology employed. Typically, in a
voiceband network, there may be from 32 to 1024 time slots. The coded
information in digitized samples may be sent serially (typically, eight bits
per sample), in parallel, or in combinations. Extra bits are sometimes
added as they pass through the switch to check parity, to transmit supervision states, or to allow for timing adjustments.
Digital time-division networks are economically attractive because the
network fabric may be constructed with large-scale integrated circuit
components. Interfaces to digital transmission facilities are less costly
and avoid the transmission impairments introduced in digital-to-analog
and analog-to-digital converters that are required with analog spacedivision switches. Therefore, time-division networks promise better compatibility with the current trends in digital techniques being applied to
data and speech transmission.

7.4 CONTROL MECHANISMS
The control mechanism in a switching system must interpret and respond
to signals and operate the switching network. Techniques of control
have evolved from early electromechanical switches (which were driven
directly by dc pulses from telephones) to sophisticated computer systems.

7.4.1 DIRECT PROGRESSIVE CONTROL
In the earliest automatic switching systems, each switch has an associated
controlling mechanism. Step-by-step systems, for example, have a stage
of switching associated with each dialed digit. As the customer dials, successive stages of switching respond to the dial pulses that represent successive dialed digits and progressively select a path through the network;
hence, the term direct progressive control. The dialed number is not stored
in one place; rather, each dialed digit is represented by (in effect, stored
in) the position of the related switching device.
Direct progressive control systems are advantageous in that the
integration of switching network and control is economical for small- and
medium-size offices. (Even in 1983, there are thousands of these smaller
step-by-step offices in service in Bell and independent operating

Chap. 7

Switching

259

companies, and their replacement has only become significant in the last
few years.) However, there are disadvantages:
• Customer lines must be connected to switch terminals in strict
correspondence with directory numbers since no logic is provided for
translation. Telephones with similar numbers that receive many calls
will be subject to undue blocking. This may require a change in one
of the numbers and may adversely affect that customer.
• Alternate routing is not possible since dialed digits are "used up" by
the system instead of being stored for possible future reuse.
• The switching network is inefficient because it cannot look ahead to
see if it will be blocked along the way.
• New services are usually impossible or expensive to incorporate in a
direct progressively controlled switching system.
7.4.2 REGISTERS, TRANSLATORS, AND MARKERSCOMMON CONTROL
Switching designers alleviated some of the problems of direct control by
adding a register to the control of a switching network. When customers
take their telephones off-hook to originate a call, the control responds
immediately (before dialing can start) to set up a connection through a
first portion of the network from the originating line to an idle register
among a group of registers. The portion of the network used may be the
concentrator of the switching network or a special separate network
called a connector. In this register (or indirect) progressive system, the
dialed digits are collected in the register. Logic associated with the register then causes digits to be pulsed out to the rest of the switching network to make the proper connection. If blocking occurs, a second
attempt to find a connecting path through the switching network can be
made because the register still contains the dialed digits.
The registers are not dedicated to each customer on a per-line basis
but instead are connected after the concentration stage and, thus, are
common to many customers. Once the connection is established to the
calling line, the register is cleared of its collected digits and returned to
the idle state to serve another originating call. Thus, a few tens of registers can serve all the calls generated by thousands of customers. This
sharing of control registers in a progressive network is a limited form of
the common-control principle.
Translators convert one multidigit number to another. For example,
once a register has collected the dialed digits, it can search for and connect itself to an idle translator circuit. The translator then converts the
dialed number (or directory number) into another number that determines the location of the terminating connection (switch terminal) of the

260

Network and Systems
Considerations

Part 2

number being called. Thus, switching system design is freed from the
constraints of the decimal numbers associated with the rotary dial. Also,
by simple rewiring in the translator, a customer's terminating line connection can be moved without requiring a corresponding change in the
directory number.
Since a register can obtain the information it needs from an electromechanical translator in less than a second, one translator can serve
the calling traffic for a number of registers, further concentrating control
and progressing toward full common control.
Common control takes advantage of the fact that control functions are
not needed for the entire duration of a call but only for small portions of
the time. With common control, many users share the control equipment, thus reducing the amount of control equipment that must be provided. The common-control principle is further exploited in marker systems, where path selection and switch control functions are concentrated
into a wired logic unit called a marker.
The coordinate network in Figure 7-8 can be used to demonstrate
marker control of a space-division network. The marker has access to the
busy-idle condition of all the links in that network. If the control determines that terminal A is to be connected to terminal B, then it can examine the status of links I, II, and III for an idle path. Having found an idle
path, it will operate the associated switch crosspoint to make the connection, and the links will be made busy, or "marked," for the duration of
the call.
It should be noted that common-control techniques using a marker are
not limited to coordinate networks. With link access, a marker can also
be used to control the step-by-step network shown in Figure 7-4. In
actual practice, all four combinations of progressive or common control
with step-by-step or crossbar switches have been used in public telephone networks.
When markers are used to control switching networks, separate registers may still exist (see Figure 7-14), but instead of driving the switches
directly, they pass the digits to a marker. The marker makes translations
(for example, from directory number to terminal identification), tests
many possible paths simultaneously to select one, and causes the proper
switches to operate.
As shown in Figure 7-14, connectors efficiently access the functional
elements of common control to communicate control information from a
customer line to registers, markers, translators, and finally, to the network
to set up the call. This functional approach allows more flexibility of
action and ease of change. Modifications to such systems have resulted in
many new services and wider applications of a single system to serve as
local, tandem, and toll switches and as automatic call distributors. 6

6 Section 11.2.6 discusses automatic call distributors.

.4~

.4~

CUSTOMER {
LINES

SWITCHING
NETWORK

~~

.4.

.4~

.4~

I

}

~~

~~

CONNECTORS

J

I
REGISTERS

OTHER
OFFICES

~

CONNECTORS

~

TEST
~
MARKERS

CONNECTORS

TRANSLATORS

Figure 7-14. Marker-type common-control system.

Common-control functions originally used electromechanical switches
and relays. With the advent of solid-state electronics, some or all of the
common-control functions were implemented with electronic wired logic.
This allowed a further reduction in the number of control elements, saving floor space and money. Whereas a large switching system carrying
heavy traffic might require a dozen or more relay markers to control its
network, one electronic marker may handle that same workload, although
a second such marker might be included primarily to serve as a backup.

7.4.3 STORED-PROGRAM CONTROL
With stored-program control (SPC), specially designed processors that
execute software programs replace the wired-logic common control.
Since the addition of new customer services and special features, changes
in translations, and other modifications can be implemented as changes
in the program rather than as more complex changes in hardware, SPC
greatly enhances flexibility. Figure 7-15 illustrates the basic components
261

Network and Systems
Considerations

262

Part 2

of a typical SPC system with centralized control. Besides the switching
network, they include:
• one or more high-speed processors (central control) that interpret and
execute the instructions of the program
• a memory (for example, program store) that stores the program
• a memory (for example, call store) that is used as an erasable scratch
pad to record and accumulate data during call processing
• input signal devices (for example, scanners) through which the central
control receives information such as customer on-hook, off-hook, and
dialed digits
• output signal devices (for example, signal distributors) through which
the central control causes network switches to operate.

CUSTOMER
LINES

{

SCA NNER

}

SWITCHING
NETWORK

...

4

I

.4~

t-

ADDRESS

ADDRESS

...

.....

......

...

CALL
STORE

........

~r

CENTRAL
CONTROL

~,

/~
V

OTHER
OFFICES

SIG NAL
DISTR IBUTOR

....L

.....

....

PROGRAM
STORE

ADDRESS
STORED PROGRAM
AND TRANSLATIONS

CALL-PROCESSING
REGISTERS
MAINTENANCE
CENTER

Figure 7-15. A centralized stored-program control system.

In the particular system shown in Figure 7-15, the principle of common
control is fully applied. Processors perform all the functions necessary to
switch calls. Through time sharing, processors simultaneously handle
many calls in various stages of completion. The central control executes

Chap. 7

Switching

263

one function per call and then progresses to the same function on a
different call or to ,another function on the same or on a different call.
The processors are sufficiently faster than the network they control, so
that they are able to interleave the control functions for many different
calls without creating any delays noticeable by the customer.
Of the several types of SPC processor arrangements, the most common
are a full centralized control, independent multiprocessors that perform
load sharing, and functional multiprocessing arrangements in which
different processing functions are allocated to different processors.

Centralized Processing
The application of electronics allows a large system to be fully controlled
by a single high-speed processor. This has the advantage of greatly simplifying the interface circuits between the controller and the rest of the
switching system. It also presents a single point (the processor's
software) for introducing new features and services. It has the disadvantage that a complete control complex must be provided even for small
office sizes, and complete redundancy is often required to allow the
system to operate in the presence of a processor fault or while changes
are made to the controller.

Independent Multiprocessing
By using more than one processor, the central controls may be designed
to operate independently and share the workload. Load sharing allows
two (or more) central con troIs to handle more calls per unit time than
would be possible with a single (usually duplicated) unit. However, in
the event of a failure of one unit, the remaining units must carryon with
reduced system capacity. Under normal conditions, the two (or more)
processors work independently, each performing the full range of functions required for call processing. At least part of the memory (for example, the busy-idle status of lines) must be accessible by all processors in
order to avoid conflicting actions among the processors.
A major advantage of multiprocessing is that small offices require less
than the maximum number of processors, lowering the cost of control.
Increased capacity can be provided by adding processors. However, as
more processors are added, the call capacity of each processor decreases.
Conflicts on processor access to memory and peripheral equipment
increase with the number of processors. Unless the amount of sharing of
memory and peripheral equipment is carefully controlled, independent
multiprocessing rapidly reaches a practical limit with respect to office size
and would not be an economical approach for large switching offices.

264

Network and Systems
Considerations

Part 2

Functional Multiprocessing
The advent of low-cost microprocessors has encouraged the distribution
of the call-processing functions to a number of processors within a single
switching system. Functional multiprocessing involves the allocation of
different call-processing functions to different processors. As in load
sharing, the processors frequently communicate with one another either
directly or through common memory.
Functional multiprocessors may be arranged in a sequential, hierarchical, or hybrid structure. In sequential structures, as on an assembly line,
each processor is responsible for a portion of a call and hands the next
step to a succeeding processor. In hierarchical structures, as in a masterslave relationship, a master processor assigns tasks to subsidiary processors and maintains control of the system. Frequently, the subsidiary processors are dedicated to segments of the switching or signaling circuits.
As network and peripheral equipment modules are added, this control
capability is correspondingly enlarged. This enables processing capacity
to be added as the system size increases and avoids incurring the cost of
excess processor capacity. Most recent switching systems, such as the
5ESS switching equipment (discussed in Section 10.3.4), use functional
multi processing.

AUTHORS
C. E. Betta
w. R. Byrne
W. S. Hayward
L. G. Raymer

F. F. Taylor

8
Signaling and Interfaces

S.l INTRODUCTION
8.1.1 SIGNALING

Signaling is the process of transferring information between two parts of a
communications network to control the establishment of connections and
related operations. For example, a customer originating a simple call usually has three signals to send: call origination (receiver off-hook), call
address (dialing), and call termination (receiver on-hook). The signaling
considered here is normally carried on over a distance and is of two traditional application-related realms: customer-line signaling and
interoffice trunk signaling. Customer-line signaling refers to the interaction between the customer and the switching system that serves the customer. Interoffice trunk signaling refers to the exchange of call-handling
information between switching offices within the network. Certain applications, such as foreign exchange lines, may have some characteristics of
each of these signaling realms and therefore are discussed as a third
realm called special-services signaling.
Alerting stations on a
nonswitched private-line configuration is another signaling process in
this realm. A fourth signaling realm has been created by the need to
communicate directly among the computer-based systems that have been
introduced to support telephone company operations. These signals are
often not directly associated with a specific line or trunk but are carried
by dedicated data links. The operations systems are linked by the operations system network (see Section 15.5), which also links some operations
systems to switching and transmission systems.
The term signaling, as it is used in this chapter, does not include the
many and varied call-handling interactions between components within a
switching system or within the more complex forms of station
equipment.
265

266

Network and Systems
Considerations

Part 2

The amount of information sent in signaling is normally much less
than in the associated communications. As a result, network control signals have generally been carried by the same transmission channel as
message signals. Use of the voice channel when it is not occupied by
conversation, as in TOUCH-TONE dialing, or use of frequencies not occupied by voice are typical examples of how this is done.
In considering signaling, then, it is necessary to describe interfaces
between terminal equipment and transmission system, between transmission system and switching system, and between transmission system and
transmission system.
8.1.2 INTERFACES
An interface can be thought of as the common boundary or set of points
where two systems or pieces of equipment are joined. Interface
specifications are the technical requirements for mating equipments. An
interface device is any equipment used on one side of an interface to
ensure that the interface specification will be met.
The set of boundary requirements at an interface, to a degree, separate
responsibilities on the two sides of the interface. This allows each side
the flexibility of rearrangement and evolutionary introduction of new
equipment and services. The interface specification should be defined so
as not to impede technological progress and to minimize the need for
changes in the interface specification itself as new products and services
evolve.
Interfaces also provide a demarcation point from which testing can be
performed on the individual units being mated. If the interface is well
defined, the units can be designed and tested independently, ensuring
that they will work together as a total system. The interface specification
also makes it possible to conduct tests at the interface to locate the
sources of service impairment after equipment is put into service. Ideal
interfaces are not always achievable, and, as will become apparent, the
term interface sometimes is used in a broader sense than defined above.
The first five sections of this chapter discuss signaling as a principal
component of a telephone interface specification, including descriptions
of the signaling functions, applications, interfaces, and techniques. The
rest of the chapter discusses other forms of interfaces as they apply to the
telecommunications network.

8.2 SIGNALING FUNCTIONS
A typical telephone connection sequence from one line to another in the
same central office illustrates the primary functions of signaling. These
functions (and the corresponding actions by the customer where applicable) are listed below. Figure 8-1 is a schematic of this sequence including

CALLING
CUSTOMER

LINE

TRUNK

)

"
ON-HOOK
IDLE
OFF-HOOK

....

TERMINATING
CENTRAL OFFICE

ORIGINATING
CENTRAL OFFICE

DIAL TONE

......
......

""'"

(ADDRESS)

......

ON-HOOK

LINE

....

..........

""'"

......
......

""'"

..........

.

~

........

OFF-HOOK(CONNECT)

®

OFF-HOOK

"WINK"

ON-HOOK
ADDRESS

AUDIBLE RINGING TONE

~,

"

ON-HOOK

TIME

~

..........

,/

CD

........ CD

®

CALLED
CUSTOMER

...

@. . . 0.
........

OFF-HOOK (ANSWER)

AUDIBLE RINGING
TERMINATED

CD

------- - - - - - - - - - -

---4

RINGING
(20 Hz)

OFF-HOOK

......

(ANSWER)

RINGING
._

":'E~~T':D_ •

CONNECTED
..................... ~ •• • • (CUSTOMERS
.................
CONVERSATION ENSUES) •••• •
~

ON-HOOK

......

.....

......

....

""'"
DISCONNECT

ON-HOOK

......

Figure 8-1. Signaling on a typical completed call_

the associated switchhook states. 1 The circled numbers in the figure
correspond to the steps in the list.

1

1)

A request for service is initiated when the caller lifts the telephone
handset off the cradle or switchhook.

2)

The central office sends a dial tone to the caller to indicate that the
caller may begin dialing.

When both customers are served by the same central office, that office handles both the
originating and terminating central office functions shown in Figure 8-1 (steps 1 through
8 in the text).

267

Network and Systems
Considerations

268

Part 2

3)

The address of the called station is transmitted when the caller
dials the called number.

4)

If the called station is not busy, the central office alerts the called
party by sending a ringing signal to the called station.

S)

Feedback is provided to the originating station by the central
office:
a)

if the called station is not busy, the central office returns an
audible ringing tone to the caller; or,

b)

if the called station is busy, the central office sends a busy
signal to the caller (not shown on Figure 8-1); or,

c)

if the call cannot be completed through the central office, the
office sends a "reorder" message to the caller (not shown on
Figure 8-1).

6)

The called party indicates acceptance of the incoming call by lifting the telephone handset.

7)

The central office recognizes the acceptance and terminates the
ringing signal, removes the audible ringing signal, and establishes
a connection between calling and called stations.

8)

The connection is released when either party hangs up.

When the called customer is in a different central office, the following
interoffice trunk signaling functions are required:
9)

The originating office seizes an idle interoffice trunk, sends an offhook indication on the trunk, and requests a digit register at the
far end.

10)

The terminating office sends a "wink" (an off-hook followed by an
on-hook signal); this indicates a register-ready or start-dial status to
the originating central office to initiate the output of address digits.

11)

The originating office sends the address digits to the terminating
office.

In Figure 8-1, the terms on-hook and off-hook have been extended to the
interoffice trunk signaling junctions even though the switchhook does
not exist as a part of the trunk. The "wink" signal is needed because the
receiving register for the address signals is not permanently associated
with a trunk; it is called for and switched to the trunk in response to the
connect signal. Also, in interoffice calls, the originating office generates
dial tone while the terminating office generates the ringing signal and
the audible ringing tone.

Chap. 8

Signaling and Interfaces

269

The term supervisory is often used to refer to control functions that are
basically on-off, such as request for service, answer, alerting, and return
to idle. The term address refers to the telephone number of the called
party. The term information refers to audible tones and announcements
used to convey call-progress information to customers or operators.
The ways in which these signals are generated, transmitted, and
detected differ for each realm of signaling (for example, customer line,
interoffice trunk) and transmission facility used (for example, analog carrier, digital carrier).2 Sections 8.4 and 8.5 describe, respectively, signaling
interfaces and techniques used to implement signaling functions.
Most of the above signaling functions have been provided in some
form since the initiation of telephone service and are apparent to the customer. Other signaling functions have been introduced more recently
and are not apparent to the customer. Network management signals, for
instance, are sent from one office to another to control the gross flow of
traffic under unusual load situations, as described in Section 5.6. Other
signals may interrogate a remote data base to access special billing or
routing information for more flexible call handling. Also, a set of functional signals is provided for many electronic offices that are remotely
monitored and administered by computer-based operations systems. 3 In
general, with the widespread introduction of computer control and operation of the network, the list of functional signals can be expected to grow
rapidly.

8.3 FUNDAMENTAL CONSIDERATIONS
A fundamental objective of the telephone industry is to make operation
of the telephone as simple and universal as is practical. This has resulted
in a relatively small number of arrangements for customer-line signaling,
which are usually seen by the customer as highly standardized procedures. On the other hand, interoffice signaling is essentially a
machine-to-machine interaction and is, therefore, less constrained by consideration of human factors. Rather, the emphasis is on overall efficiency
and flexibility. Consequently, over the years, interoffice signaling has
been extensively affected by new transmission techniques and advances
in switching system design. This is reflected in the large variety of signaling arrangements in service.
To satisfy the objectives of uniformity for customer signaling and
flexibility for interoffice signaling, a high degree of independence has
been maintained between these two signaling realms. Facilitated by the
2 Sections 9.3 and 9.4 describe analog and digital carrier systems.
3 Section 14.3.2 describes one such operations system-the Switching Control Center
System.

270

Network and Systems
Considerations

Part 2

widespread use of common-control, switching systems in local central
offices provide an effective signaling buffer between lines and trunks.
Another basic signaling consideration is promptness or its inverse,
delay. The calling customer may experience three components of delay
in completing a call: dialing time, post dialing delay, and answer delay;
all involve signaling to some degree.
Dialing time is the time it takes for the calling customer to dial the
desired number: from the time the customer lifts the handset (or goes
off-hook) until the last digit is dialed. This delay is determined by the
central office delay (called dial-tone delay) in recognizing the customer's
request for service by providing the dial tone, the customer's reaction
time, and the speed with which the customer operates the rotary dial or
pushbuttons.
Postdialing delay is the elapsed time from the end of dialing to the start
of ringing (or other feedback such as a busy signal) at the called end.
Postdialing delay depends on many factors, including the number of
switched links in the connection, the types of interoffice signaling used,
switching system work times, and the traffic load.
Answer delay is the time from the beginning of ringing until the called
station answers. It is determined primarily by the called customer's
promptness in answering and, to a lesser degree, by the alerting method
used.
The ability to interconnect with systems designed by a number of
manufacturers has become more important recently. This requires greater
attention and effort with respect to national and international signaling
standards organizations, such as the Co mite Consultatif International
TeIegraphique et Telephonique (CCITT), so that interfaces are kept simple
and few in number.

8.4 SIGNALING SYSTEM APPLICATIONS AND
INTERFACES
For any signaling carried by a transmission facility, interfaces must be
provided for the interchange of signaling information between the facility and the source and between the facility and the destination. In some
cases, the interface is a well-defined demarcation point with specified
impedances and voltages. (The E&M lead interface described in Section
8.4.2 is an example of this type.) In other cases, the interface may be so
integrated into the circuit design that it is hard to define. In such cases,
the signaling information interchange requires careful coordination in
design and, to some degree, in engineering between the source, destination,4 and intervening facility.
4 Source and destination are used here to mean station equipment or a switching office.

Signaling and Interfaces

Chap. 8

271

This latter concept is best illustrated for a customer loop formed by a
metallic pair. Supervision is provided between the central office and the
station set by the presence or absence of direct current in the circuit
formed by the two wires of the loop, the switchhook contacts in the station set, and the battery (or other source of dc voltage) to which the loop
is connected at the central office as shown in Figure 8-2. (Other circuitry
needed for message transmission and address signaling is omitted from
the figure.)
CUSTOMER'S STATION
TERMINAL

OUTSIDE PLANT
FACILITY

CENTRAL OFFICE

CURRENTSENSING
CIRCUIT
TELEPHONE
SWITCHHOOK

DC

T

POWER
SUPPLY

STATION-LOOP
SUPERVISION

Figure 8-2. Schematic diagram of supervisory signaling interface
between a customer's station terminal and the central office.

A current-sensing device is useq. at the central office to detect the station set's on-hook or off-hook signal. A single standard interface does
not exist between the customer's line and the station set or between the
line and the central office because of the wide variation in length and
type of loop plant. s It would be prohibitively expensive to condition
many millions of customer's lines to end in a standard impedance and
signal level.
Until 1976, signaling in the Bell System was almost entirely on a perline or per-trunk basis; that is, the signaling information for a particular
channel was carried on the same channel as the voice or other message
information. Of the various signaling interfaces, only two are described
here: loop signaling and E&M lead signaling, as used for per-line and
per-trunk signaling.

5 AT&T 1983 and 1980a, respectively, contain definitions and some standards for the two
interfaces.

272

Network and Systems
Considerations

Part 2

Common-channel interoffice signaling, introduced in 1976, combines the
signaling information for a number of channels and transmits the information over a data link derived from a dedicated facility. For commonchannel signaling, the interface is primarily defined in terms of a set of
binary-coded messages between digital processors. The rate and form of
transmission, however, can vary. For example, the digital stream can be
transmitted via data modem over an analog (frequency-division multiplex) channel or more directly over a digital (time-division multiplex)
channel. The format used at this interface is often called a protocol (see
Section 8.8). The following paragraphs discuss interfaces for the
customer-line, interoffice trunk, and special-services realms (or areas of
applications), which are summarized in Table 8-1.
8.4.1 CUSTOMER-LINE SIGNALING
Most customer loops consist of a pair of wires between the central office
and the customer's station equipment. The percentage of loops implemented by carrier systems has, however, grown rapidly. Recent growth
has primarily been in digital carrier systems that have become economically attractive as a result of advances in technology. When the loop is
implemented on a carrier system, the customer's station equipment is
connected by a pair of wires to a carrier remote terminal (near the
customer's premises) rather than to the central office.
The signaling interface between a customer's terminal equipment and
either the central office or the carrier remote terminal has a standard format known as metallic loop signaling. This arrangement provides for continuous application of a dc voltage from a dc power supply (see Section
12.2) toward the station, in conjunction with a current-sensing device to
recognize the supervisory status of the station. The nominal value of the
dc voltage is usually 48 volts and is commonly called battery voltage, or
simply, battery.
Figure 8-2 shows the situation for metallic loops. Similarly, Figure 8-3
shows the situation for carrier-implemented loops. It should be noted
that loop signaling is also shown between the central office switching
system and the central office terminal. The carrier terminals convert the
signal information to any signaling method appropriate for the carrier.
(Sections 8.5.3 and 8.5.4, respectively, discuss signaling over analog and
digital carrier systems.)
On-hook, or idle, is indicated by no current flow, whereas off-hook, or
seizure, is indicated by the flow of current in the loop. For purposes of
supervision, the battery polarity and, hence, current direction are not critical. Normally, however, negative 6 battery (~48 volts) is provided on the

6 Positive voltages are generally avoided in outside plant cables because if there is any
moisture present, copper may be lost through electrolysis.

TABLE 8-1
SUMMARY OF SIGNALING SYSTEM
APPLICATIONS AND INTERFACES
Signaling
System
Application

Interface

Characteristics

Dc signaling
• Loop signaling basic station

Loop-start origination at station
Ringing from central office

f ______________________________________

~

Customer line

Dc signaling

• Loop signaling coin station

Loop-start or ground-start origination
at station
Ground paths may be used in addition
to the line for coin collection
and return
I-way call origination

• Loop-reversebattery

Directly applicable to metallic facilities
Both current and polarity are sensed

Interoffice trunk

Available to carrier facilities with
appropriate facility signaling system

2-way call origination
• E&M lead
Requires facility signaling system for
all applications
Standard station loop and trunk
arrangement as above

• Loop type
Special services

Ground-start format for private branch
exchange (PBX)-central office trunks
similar to that for coin stations
Automatic or ringdown for PBX nondial
tie trunks

E&M for PBX dial tie trunks
• E&M lead

E&M for carrier system channels in
special-services circuits

LOCAL
WIRE
PAIR

CARRIER
SYSTEM
REMOTE
TERMINAL

CARRIER
CHANNELS
FOR VOICE
AND
SIGNALING

CURRENTSENSING
CIRCUIT

-=-

TELEPHONE
SWITCHHOOK

CARRIER
SYSTEM
CENTRAL
OFFICE
TERMINAL

--~--

CENTRAL
OFFICE

I

------

1
CURRENTSENSING
CIRCUIT

..1

-=-

DC
POWER
SUPPLY

DC
PO WER

~SU PPLY
STATION-LOOP
SUPERVISION

LOOP SIGNALING

OOP SIGNALING

Figure 8-3. Schematic diagram of signaling interface
for loop carrier application.

ring conductor, with ground on the tip? This format provides simple 2state supervision from the customer toward the central office, with only
minor variations introduced by the special features of different switching
systems.
The customer is alerted from the central office by a ringing signal
indicating the presence of an incoming call. Standard single-party ringing consists of 2-second intervals of 20-hertz (Hz) energy applied
between the tip and ring conductors, followed by a 4-second quiet interval. Other ringing durations and intervals, including 1 second on, 2
seconds off, are used for multiparty lines. To provide for ring tripping,
which turns off ringing when the customer answers, a superimposed dc
voltage is generally used in association with a current detector. Following ring tripping, line potentials revert to the normal supervisory state.
For loops implemented on carrier systems, ringing and other signals from
the central office to the customer must be converted to a form suitable for
carrier transmission. As in the case of on-hook, off-hook supervisory signaling (Figure 8-3), the carrier system is essentially transparent as viewed
from either the station or central office end.
For 2-party and multiparty (4-party or more) service, the 20-Hz ringing voltage is applied to either the tip or ring conductor with respect to
ground and is superimposed on either a positive or negative dc ringtripping potential. In this way, up to four distinct signals are provided,
7 In a 2-wire pair, the two leads are often called tip (T) and ring (R) after the parts of a
standard telephone plug to which they connected in the days of manual switchboards.
Similarly, a third wire (if present) is called sleeve. In 4-wire transmission, the four leads
are called T, R, TI, and Rl.
SLEEVE....

.... RING

.: . -D'------I.....-..~

~-----...... '~
L--_ _ _ _ _........

274

TIP

Chap. 8

Signaling and Interfaces

275

which, with appropriate connection of ringers in the station sets, provide
full selective ringing of 2- or 4-party lines. With coded ringing durations, semiselective ringing of 8-party service can be provided. For 10party service, additional coding is used beyond the standard one long
and two short rings. By 1982, multiparty service was quite rare in the
Bell System.
A special case of loop supervisory signaling exists for coin stations, in
which additional control signals are required for coin collection and coin
return. There are two basic types: loop-start, associated with dial-tonefirst service, and ground-start, associated with coin-first service. Loopstart provides ordinary loop s,ignaling, with the addition of battery polarity reversal used for answer supervision and a positive or negative 130volt dc signal applied from ground to both the tip and ring conductors
for coin collection or coin return. For ground start, seizure of the loop is
initiated by applying ground to the ring conductor at the station in
response to the insertion of the proper coin(s). A detector circuit in the
central office recognizes this ground, applies dial tone, and establishes
conventional loop supervision. Coin collection and return functions are
then provided as described for loop supervision.
In addition to switchhook supervision, the customer must pass the
address code of the called party to the switching office. This address
information is communicated either as dial pulses or as tones from a
pushbutton telephone. Dial pulses consist of short on-hook pulses occurring as interruptions in the normal off-hook loop supervision current at a
10-pulse-per-second rate. The number of dial pulses in a sequence equals
the value of the digit, except for ten pulses, which equal the digit O. The
digits are separated by a relatively long off-hook interval, the length of
which depends on how fast the customer can rewind and release the dial.
The ratio of break (on-hook) interval to total pulse cycle interval is
frequently referred to as percent break and is typically between 58 and 62
percent. Twenty-pulse-per-second dialing is also used occasionally for
special applications. Figure 8-4 shows dial pulsing for the digit 3.
Tones from pushbutton telephones consist of pairs of selected frequencies corresponding to the digits to be transmitted. The frequencies
used are coded as shown in Figure 8-5.

J-----

OFF·HOOK

1+--

100 ms

--.I

'~----------------------.---------------------~/
DIGIT 3

Figure 8-4. Dial pulsing.

HIGH-FREQUENCY GROUP

1209 Hz

697H'

Il.
;:)

oa:

C)

o>
z

w

;:)

o

w

a:

II.

~

o..I

1336 Hz

G] [!]

m
1477 Hz

170H'm m m
852H'm m m
~1H,

GJ mEJ]

Figure 8-5. Tone-dialing frequency groups.

For TOUCH-TONE service, a pair of tones is generated when a button
is pressed. If the number 7 is pressed, for example, the 1209- and 852-Hz
frequencies are generated. Receivers at the central office recognize these
tones as representing the number 7. The tones have been carefully
selected to minimize harmonic interference and the probability that a
pair of high and low tones will be simulated by the human voice, thus
protecting network control signaling. For TOUCH-TONE service, signals
may be transmitted over any voice circuit (providing loss and noise limitations are not exceeded). They are therefore facility independent and
are frequently used for other applications. Upon receipt of address signals from TOUCH-TONE service, the tones are decoded to indicate the
called address to the central office.

8.4.2 INTEROFFICE TRUNK SIGNALING
In the Bell System, network operation with automatic switching has historically required that signaling for a call begin at the originating station
and follow the same path as the call. itself. The obvious reason for this
approach was to avoid the added costs of separate transmission channels
for signaling. However, this mode of operation introduces the very real
possibility of mutual interference between signaling and voice transmission. To minimize interference, the basic rule has been to keep signaling
276

Chap. 8

277

Signaling and Interfaces

and talking from overlapping in time as much as possible on a given
connection.
Because there was no separate signaling route or network, voice connections involving several trunks in a series were established by adding
one trunk at a time, starting at the central office that served the calling
customer and progressing toward the central office of the called station.
On such calls, most of the signaling activity takes place before the called
station is rung. During this time, the transmission path is available for
signaling. This signaling technique is referred to as per-trunk signaling.
As shown in Figure 8-6, interoffice signaling may, in principle, be provided on a per-trunk or common-channel basis. Per-trunk signaling is
widely used in the Bell System at present and is discussed first.
Common-channel interoffice signaling (CCIS), whose use is growing
rapidly, is discussed in subsequent sections.
TRUNKS

TRUNKS
OFFICE C

OFFICE 0

··•

I PROCESSOR I
I
I

CCIS
SIG

I
I

I PROCESSOR I
I

I

CCIS
SIG

CCIS SIG

COMMON-CHANNEL INTEROFFICE SIGNALING EQUIPMENT

SIG

PER-TRUNK SIGNALING EQUIPMENT

Figure 8-6. Interoffice signaling techniques.
bottom, common-channel interoffice signaling.

L

I

Top, per-trunk signaling;

Per-Trunk Signaling
There are two major per-circuit trunk signaling interfaces: loop-reversebattery (a form of loop signaling) and E&M leads. Loop-reverse-battery is
applicable to those trunks that require call origination at only one end.
These are called l-way trunks, although it should be recognized that h is

I'll etWOrK

ana ;:,ystems
Considerations

278

Part 2

only the call origination or trunk seizure that is one way. Basic signaling
functions are still required in both directions. An E&M lead interface also
may be used on I-way trunks. On 2-way trunks, that is, trunks with call
origination permissible at either end, an E&M lead interface is required.
Loop-reverse-battery, as used on metallic facility trunks, provides for
application of nominal -48 volt battery on the ring conductor and
ground on the tip at the terminating office end of the trunk. In addition,
the terminating office has a current-sensing device and provision for
polarity reversal (that is, -48 volt battery applied to the tip conductor).
At the originating office end, means are provided for closing the loop,
causing current to flow as an off-hook signal. A polarity-sensing device
recognizes the state of the terminating end (for example, idle, disconnect,
etc.). As described in Section 8.5, the loop-reverse-battery interface can
also be used with an appropriate signaling system for applications on carrier systems.
There are numerous specialized variations of loop-reverse-battery,
most of which have very limited applications and are not discussed here.
The E&M lead interface is a true interface as described in Section 8.l.
The basic (Type I) E&M lead interface provides for signaling from the
switching system toward the facility over the M-Iead in the form of
ground for on-hook and -48 volt battery for off-hook. From the
transmission facility toward the central office, on-hook corresponds to an
open E-Iead, whereas off-hook is ground. Table 8-2 summarizes the signaling states. Figure 8-7 shows an example of E&M signaling from
switching system A to switching system B with Type I interfaces. Switching system A indicates an off-hook status (-48 volts on the M-lead) by
controlling the relay contact at point 1 on the figure. The M-lead detector at point 2 interprets the status, and the signaling circuit transmits it
over the facility, closing the E-lead contact at point 3. The E-lead detector
at point 4 in switching system B interprets the status as off-hook.
As an aid to remembering the functions of E&M leads, the "mouth"
(M) and "ear" (E) analogy is useful because it is a reminder that the Mlead at one end of the facility drives the E-Iead at the other end, although

TABLE 8-2
BASIC E&M LEAD SIGNALING STATES

State

From
Switching System
to Facility
(M-Lead)

From
Facility to
Switching System
(E-Lead)

On-hook

Ground

Open

Off-hook

-48 volts

Ground

:

WEST

IV

~

•

EAST

I

INTERFACE

INTERFACE

~

~

FACILITY

,4

.,

FACILITY SIGNALING SYSTEM

RECEIVE

:

SEND

I

-48V
-48V

M
-48V

RELAY

GND

-48V
GND

CD

M

RELAY

GND

GND

GND

E

CD
RECEIVE

SEND

SWITCHING
SYSTEM A

SIGNALING
CIRCUIT

SIGNALING
CIRCUIT

Figure 8-7. E&M lead signaling between switching systems
with Type I interfaces.

SWITCHING
SYSTEM B

280

Network and Systems
Considerations

Part 2

that is not the source of the two terms. For example, when the switching
system grounds the M-Iead, signaling an on-hook state, the facility signaling conveys this to the signaling system at the other end of the facility using one of the techniques described in Section 8.5. The far-end signaling system will then provide an open condition on the E-Iead to
signal the next switching system.
With the introduction of electronic switching systems (see Section 10.3), the standard E&M lead interface was modified to avoid noise
problems associated with ground return paths. Two varieties of looped
(paired-lead) E&M interfaces exist. In some electronic switching system
applications, full-looped E&M leads are used. These consist of loop closures on paired leads, both to and from the facility interface. For other
electronic switching system applications, both ground and battery return
for the M-Iead are brought back to the facility interface as SG and SB
leads. An example of one of these arrangements is the Type II interface
shown in Figure 8-8.
Transmission facilities are often connected in tandem without an
intervening switching system. If at least one of the facilities at a point of
connection uses facility-dependent signaling, a through connection
involving these dc signals must be made. This may not be simply a
matter of connecting interface leads, because in the past, systems have
not always been designed this way. Even if both of the facility signaling
systems have E&M lead interfaces, it may be necessary to insert a signal
conversion unit between the interfaces enabling the E-Iead of one facility
to drive the M-Iead of the other (see Figure 8-9). If the Type II interface
is used, direct interconnection is possible without a conversion unit
because the contact closure that is the E-pair output is exactly what is
needed to activate the M-Iead pair detector (see Figure 8-10).
The loop-reverse-battery and E&M lead interfaces just described provide the supervisory interface for trunks. Destination codes are transmitted as either dial pulses applied through the same interfaces or multifrequency pulsesB applied to the voice path. Section 8.5 discusses these
and other signaling techniques.

Common-Channel Interoffice Signaling
New switching and transmission systems, as well as new features for
existing systems, are continually being developed, and the complexities
are such that providing intersystem compatibility is a major challenge.
Another challenge is to reduce postdialing delay, which is becoming
more important because of new ways in which customers are using the
network. For example, many calls involving computers at one or both

8 The frequencies differ from those used for customer address signaling.

V

WEST •

INTERFACE '-....

~I

1

FACILITY SIGNALING SYSTEM

E

~

I

:

~

FACILITY

I
I
I

EAST

INTERFACE

14
RECEIVE

•

+

I

I

:
SEND

I

I
I

-48V

1
I S8

GND

GND

-48V

------' ...L
S8

SEND

I

R

GND

1

t

I
I
I

I

RECEIVE

SG

1
I

I
SIGNALING
CIRCUIT

E

I

-48V

SWITCHING
SYSTEM A

M

SIGNALING
CIRCUIT

Figure 8-8. E&M lead signaling between switching systems
with Type II interfaces.

GND
SWITCHING
SYSTEM 8

I

V

INTERFACE

INTERFACE

SIGNAL CONVERSION UNIT

I

~

I

FACILITY A

I

FACILITY B

I

-48V

I
RECEIVE

I

I
I

M

E

SEND

-

-48V

GND

-48V

GND

GND

E

M

SEND

GND

-48V
GND
GND

Figure 8-9. E&M lead signaling between facilities with
Type I interfaces using a signal conversion unit.

ends have relatively short messages, and therefore, the call setup time
represents an appreciable part of the customer's total network time. Call
origination, dialing, and answering functions are readily mechanized in
these cases, so that these components of call setup time become small.
Still another concern is that traditional per-trunk signaling methods
are not easily adaptable to certain evolving needs. Examples of such
needs are (1) transfer of network management signals, (2) combining
different types of traffic on one trunk group and yet retaining their identity at the far end, (3) far-end make-busy of trunks for maintenance purposes, (4) return of busy signal from originating rather than terminating
office, so that the intermediate trunk(s) can immediately be made available to other calls, (5) increased transparency for the network (such as
removal of constraints imposed on customer data transmissions to prevent
harmful interactions with inband signaling equipment), (6) call tracing,
(7) elimination or improved handling of simultaneous seizure of both
ends of 2-way trunks, and (8) reduction of fraud.

282

INTERFACE

S8

-48V

SEND

4
M

RECEIVE
GND

I
I
I
SEND

E

I

•

--

GND
S8

R

I

-48V

I
I
I
I
Figure 8-10. E&M lead signaling between facilities with
Type II interfaces not using a signal conversion unit.

Since 1965, stored-program control switching systems have been used
in the network in rapidly increasing numbers. With processor-controlled
offices, a totally different signaling technique, which meets all the above
needs, is possible for interoffice trunks. In such arrangements, there is no
per-trunk signaling interface but rather a single data line (with backup)
containing time-multiplexed signaling information. For economic reasons, this data link will usually go to a signal transfer point rather than
provide one data link per trunk group. The data link will then serve
multiple trunk groups from the office. Where such systems are used, neither loop-reverse-battery nor E&M leads are required, and similarly, neither dial-pulse nor multifrequency addressing is used.
Section 8.5.5 gives a functional description of common-channel
interoffice signaling (eelS). In 1976, eelS became available for No. 4A
Electronic Translator Systems and 4ESS systems. Later in the growth of
eelS, lESS systems and Traffic Service Position Systems were equipped. 9
It is expected that eelS will be developed in other processor-controlled
offices in order to permit the widespread availability of new services.

9 Chapter 10 discusses these systems.

283

284

Network and Systems
Considerations

Part 2

The initial system used a 2.4-kilobits-per-second (kbps) data link. In
1981, a 4.8-kbps terminal became available. In the future, data rates up to
64 kbps and an improved protocol may be justified.

8.4.3 SPECIAL-SERVICES SIGNALING
Special-services circuits frequently require special signaling arrangements
because they have some characteristics of both customer lines and
interoffice trunks. To provide these services, the standard customer-loop
and trunk arrangements with various modifications have been used. It is
important to recognize that the wide variability of implementation
arrangements needed for special-services circuits represents a significant
challenge in standardization and consolidation of signaling system equipment. It is beyond the scope of this text to consider all possible arrangements; however, a few major types are described.
A major category of special-services circuits includes special-access
lines and trunks, such as foreign exchange lines, Wide Area Telecommunications Services lines, and private branch exchange (PBX)-central office
trunks. Short special-access lines normally use standard single-party customer loop-start signaling. Long special-access lines, such as some
foreign exchange lines, which use carrier transmission facilities, employ
an appropriate form of signaling such as inband single-frequency signaling, described in Section 8.5.2.
Special-access trunks may use standard loop signaling but frequently
employ a ground-start arrangement similar to that used in coin service.
Special-services ground-start signaling is used to minimize the probability
of simultaneous seizures on high-usage PBX trunks. This can occur on
such circuits because a PBX attendant may attempt to initiate an outgoing
call at the same time an incoming call is attempting to alert the attendant.
If. the incoming call connection occurs during the ringing silent interval,
the attendant may not be alerted to this fact for up to 4 seconds, and a
misdirected outgoing call could be placed. To prevent simultaneous
seizures, ground-start provides for the application of ground on the tip
lead toward the PBX as an initial contact or seizure signal even before the
application of ringing. The attendant can test for this seizure and, if
found, avoid placing an outgoing call on the busy trunk. Similarly, the
trunk will appear busy to the PBX, so it cannot be selected for outgoing
calls dialed from PBX extensions.
Tie trunks on PBX or key systems constitute another major class of
special-services circuits. Tie trunks may be either dial-repeating trunks
or nondial trunks.
Dial-repeating trunks are associated with those PBXs that are capable of
routing tie trunk calls without attendant assistance and that, therefore,
require dial-pulse address information. The E&M lead tie-trunk interface
used is virtually identical to that used on interoffice trunks and, hence,

Chap. 8

Signaling and Interfaces

285

passes dial-pulse address information. Although address signaling is normally dial-pulse, TOUCH-TONE dialing is also used. Multifrequency
pulsing is normally not available at PBX systems.
Nondial trunks are associated with attendant-completed calls in which
address information is transmitted verbally. Two modes of operation for
nondial trunks are automatic and ringdown. Automatic and ringdown nondial trunks provide only a supervision alerting function. In the case of
automatic operation, the signal to the interface at the originating end is a
loop closure. For ringdown, the originating-end signal is the application
of 20-Hz ringing by the operation of a ringing key by the attendant.
When 20-Hz ringing is expected at the interface at the terminating end
but there is an intervening facility that cannot pass the 20-Hz ringing signal, a combination of these types, known as automatic ringdown, is often
used. The originating-end interface is a loop closure that results in the
application of 20-Hz ringing at the receiving end. When the attendant
answers the call,. ringing is tripped.
Combinations of dial-repeating, automatic, and ring down signaling
are often used for different directions on a given tie trunk, depending on
customer needs and the capability limitations of the terminating PBXs.
Private lines are another major class of special-services circuits that use
many different types of signaling interfaces. For the simplest cases of
private-line service, no signaling functions need be provided by the network. For example, in some private-line data services, the data sets provide the alerting function. For many private-line services, however, a
separate alerting function is necessary. This is frequently provided by
automatic or ringdown operation. The signals used are very similar to
those just described for tie trunks, with automatic ringdown being the
most common.
Many private lines are multipoint; that is, they have from several to
many stations bridged together on a common private-line network. For
such applications, selective alerting is frequently desirable. For simple
multipoint networks with a minimum number of stations, code select
may be associated with ringdown operation, or it may be combined with
automatic operation in special arrangements to suit customer needs. In
such an arrangement, coded ringing (one, two, three, etc., rings) may be
used to alert a particular station. The ring count may be interpreted by
an attendant or more typically by a code select ringing circuit in the terminating station that, recognizing its code, triggers an alerting signal at
the station.
Services requiring more signaling features beyond simple alerting or
multipoint networks with many stations may use a selective signaling
system containing combinations of dial pulses or tones from TOUCHTONE telephones to provide the selective ringing and control functions.
Many other private-line signaling systems and interfaces are also in use.
Common-channel signaling has potential for switched special-services
signaling, but at present, data link and signal transfer point costs are an

286

Network and Systems
Considerations

Part 2

obstacle because of the small number of circuits that are served in each
application.

8.5 SIGNALING TECHNIQUES
To convey signaling information from one interface to another requires
signaling transmission techniques that are compatible with the facility
and the switching system involved. There are basically five major types
of facility signaling systems: direct current, inband tone, out-of-band
tone, digital, and common-channel interoffice. In addition, signaling
systems are classified as facility dependent or facility independent.
Facility-dependent signaling techniques include dc signaling, out-ofband tone signaling, and digital carrier signaling. Facility-independent
techniques are single-frequency (SF) inband tone signaling, multifrequency (MF) signaling, and dual-tone multifrequency (DTMF) signaling
(for TOUCH-TONE service). For services built up from a combination of
tandem facilities, there may be economic advantages in the use of
facility-independent signaling systems. Common-channel interoffice signaling (eCIS) is also truly facility independent, as it is based on a
separate data link carrying multiplexed signaling information for many
different, circuits. The CCIS data link may be over an entirely separate
facility, often following a different geographic route. The interface with
a switching system requires eCIS capability in that system.
The three modes- of signal transmission are continuous, spurt, and
compelled. Continuous supervision requires steady-state transmission and
reception of status on a per-channel basis and has no memory associated
with the signaling system. This mode is restricted to dedicated, nontime-shared facilities. Spurt signals are normally associated with shortduration address information but also may be used for supervision if the
signaling system or connecting circuit provides the necessary state
memory. Short-duration signals are particularly useful on time-shared
facilities such as those used for eCIS. The compelled mode is a highreliability system used for critical applications such as overseas signaling.
With compelled signaling, every state change is transmitted, and its
reception at the far end is acknowledged with an appropriate signal. Following acknowledgement, the signal is normally removed; hence, state
memory is required.
The following sections describe each of the facility signaling systems
mentioned here-direct current, inband, out-of-band, digital, and CCISin more detail, and Table 8-3 summarizes signaling implementation.
8.5.1 DC SIGNALING
For each of the major signaling system interfaces discussed in Section 8.4
(loop signaling, loop-reverse-battery, and E&M lead signaling), one or
more facility signaling systems exist for use on dc (metallic) facilities.

Chap. 8

Signaling and Interfaces

287

The existing metallic facility signaling systems with their interfaces came
into being through an interaction of desired performance, technical possibilities, and cost of both facility and switching system-an interaction too

TABLE 8-3
SIGNALING IMPLEMENTATION
SIGNALING FUNCTIONS
• Supervision
• Addressing
AREAS OF APPLICATION
• Customer lines
• Interoffice trunks
• Special services (includes PBX trunks and PBX tie trunks)
SIGNALING SYSTEMS INTERFACES
• Loop signaling (may be associated with station loops, trunks, or
special services)
• E&M lead signaling (may be associated with trunks or special services,
usually those requiring 2-way origination)
• CCIS
FACILITY SIGNALING SYSTEMS
Facility dependent (carry supervision, address, or both)
• Dc (design and engineering depend on resistance of signal path
conductors) - station loops, short trunks, short special- services
channels, and end segments on long channels
• Out-of-band (as on Nl carrier)
• Digital (as on Tl carrier)
Facility independent
• Inband tone
SF (carries supervision, address, or both)
MF (carries address)- interoffice
DTMF (carries address) - primarily station loop
• CCIS (carries both supervision and address) - interoffice
SIGNALING MODES
• Continuous
• Spurt
• Compelled

Network and Systems
Considerations

288

Part 2

complex to be treated here. In general, the main constraint is the dc
resistance of the signaling path, including that of the facility.
At the simplest end of the spectrum is customer-line, or station-loop,
signaling. This uses a wire pair connecting a pair of contacts at one
interface with a battery and current sensor at the other (as shown in Figure 8-2). Constrained by design and engineering rules to ensure proper
operation, this arrangement is a facility signaling system. With a typical
central office and station, it will operate with external conductor loop
resistances up to 1300 ohms, a value that permits satisfactory signaling
and transmission over a majority of loops. Table 8-4 illustrates signaling
range capability for different gauges of cable pair. In practice, the loop
may consist of two or more segments of different gauge for best economy
within the 1300-ohm limit.
When the station-loop resistance exceeds 1300 ohms, provision is
made for signaling range extension. This is used only in exceptional cases;
the great majority of loops do not require any such treatment. Range
extension normally involves an extra circuit for detection and regeneration of signals, but it may involve no more than a higher-than-normal
battery voltage at the central office. Range extension can be applied on
trunks but is very seldom required. (Section 9.2.1 discusses loop design
in more detail.)
For trunks, the loop-reverse-battery interface with dc signaling
transmission, like basic station-loop signaling, uses only the metallic conductor, but it can work over a much longer circuit. In common with

TABLE 8-4
LOOP LENGTH ATTAINABLE WITH FIXED WIRE GAUGE

Gauge

Ohms Per
Thousand Feet

Loop Length Limit'"
For 1300 Ohms
(Miles)

26 - fine wire least c~st per foot

83

3.0

24 -

52

4.7

22 -

32

7.7

19 - coarse wire greatest cost per foot

17

14.5

* On the longer circuits, load coils are required to improve speech quality. These are spaced
at 6000-foot intervals and have a resistance of 9 ohms. This makes the actual loop length
limit slightly smaller than shown in the table.

Chap. 8

Signaling and Interfaces

289

basic station-loop signaling, loop-reverse-battery uses a grounded voltage
supply at only one end (the terminating end of the trunk), thus avoiding
problems of earth potential differences between offices.
In station-loop signaling and loop-reverse-battery signaling over
metallic conductors, the relays that receive signals are generally considered part of the switching system, and so the facility signaling system
design for these applications naturally has fallen to people who design
switching systems. The design of range extension equipment for metallic
facilities is done primarily by transmission designers.

8.5.2 INBAND SIGNALING

Facilities not capable of passing dc signals must be provided with
different types of facility signaling systems. This is true of all carrier systems. For analog carrier systems, a number of inband 10 tone-signaling
techniques have evolved. Since these depend only on the existence of a
voice channel, they are facility independent and may be used over tandem facilities including metallic pairs and digital carrier.
Included in inband systems are SF signals, MF signals, and signals
from TOUCH-TONE service. SF signaling provides both supervision and,
where appropriate, dial-pulse address signaling and is compatible with
loop-reverse-battery and E&M lead signaling interfaces. All of these systems perform their functions without the need for special signaling range
extension or conversion equipment.
SF signaling provides two basic states: on-hook (normally 2.6
kilohertz [kHz] tone-on) and off-hook (normally tone-off). Both the application and the removal of tone are controlled from the input dc interface;
a received-tone detector provides an output signal to the dc interface.
There is, of course, a mating SF unit at the distant end of the channel. A
typical application is on a 4-wire carrier system with origination from
either end.
The use of the two supervisory states is dictated by application
requirements. A simple example is the use of SF to extend customer-line
signals over carrier. Toward the central office, tone-off is off-hook, or
seizure; and tone-on is on-hook, or idle. In the reverse direction, tone-off
is both idle and busy, with tone-on corresponding to the application of
ringing.
With SF signaling, time delays must be allocated for tone recognition,
to allow for distortion caused by facility noise or transmission degradation and to prevent false signaling by voice simulation. The latter is
referred to in signaling terminology as "talkoff" because a voice-produced

10 Within the voiceband, 200 Hz to 3.5 kHz, as described in Section 6.1.

290

Network and Systems
Considerations

Part 2

tone with the proper characteristics can simulate an on-hook condition,
causing the circuit to be disconnected (the SF receiver functions as if the
customer has hung up).
These factors normally necessitate sending a tone signal of minimum
duration (typically over 50 milliseconds) independent of the input pulse
duration. The detected signal must then be reconstructed to approximate
that of the original source by a process referred to as pulse correction. The
pulse-correction scheme used depends on the particular service requirements of the SF application.
Many signaling system interfaces can be used with SF signaling units.
In toll applications, the interface is typically E&M lead. The interface
also can be loop-reverse-battery, in which case, the SF unit appears transparent to a switching system. Where a customer line is quite long (as in
the case of foreign exchange or some other special-services circuit), an SF
unit can be used at the station, with a loop-signaling interface. On a
foreign exchange line, for example, the off-hook signal from the station
would remove the tone sent toward the switching office. Since specialservices circuits often use various combinations of metallic and carrier
facilities, optional signaling conversion circuitry with enough range capability for the majority of applications is provided in the SF unit for each
of several signaling interfaces.
As in customer-line signaling, dial-pulse trunk address signals consist
of interruptions of the off-hook supervisory state of the trunk; the rate of
interruption may be ten or twenty pulses per second. Dial-pulse signals
can be applied directly to both loop-reverse-battery and E&M lead interfaces. Multifrequency pulsing is very similar to TOUCH-TONE signaling
in that pulses of pairs of selected tones are used to represent digits. In
MF pulsing, the tones were not specifically chosen to protect against
simulation by voice, since voice transmission from the calling customer is
inhibited during MF pulsing. In addition to the actual address digits, a
keypulse tone pair is sent as a start-of-address signal to unlock the
terminating-end receiver, and a start signal is sent to indicate end-ofaddress (start call processing). These signals are sometimes faintly audible to the calling party.
8.5.3 OUT-OF-BAND SIGNALING

Analog carrier systems typically do not use the full 4-kHz voicefrequency channel slot because of channel separation filters in the channel modems. ll These filters sharply attenuate the input voice signal above
about 3.5 kHz. This has led to the development of out-of-band signaling

11 A contraction of the words modulator and demodulator signifying an equipment unit that
performs both of these functions.

Chap. 8

Signaling and Interfaces

291

systems using tones above this value but below 4 kHz. Such systems are
not widely used, as they are normally facility dependent and are not
applicable to tandem facilities without added hardware at the point of
interconnection. However, one out-of-band system is often used with the
Nl carrier system (see Section 9.3.3) that uses 3.7 kHz as the signaling
frequency. While out-of-band tone signaling has been considered for
other systems, including A6 channel banks, the universal applicability of
SF signaling has outweighed any possible economic advantages of outof-band signaling.

8.5.4 SIGNALING OVER DIGITAL FACILITIES

On most digital facilities, when transmission is provided by digital bit
stream, it is very convenient to designate one or two bits periodically as
signaling. The bit 1 or 0 then corresponds directly to a specific signaling
state such as on- or off-hook. Signaling bits may be assigned to every
encoded sample (8000 per second for pulse-code modulated voice)12 or at
a much lower rate consistent with the very low information transfer rate
associated with per-channel signaling. Typical sample rates of about 1
kbps (sample interval of 1 millisecond) are used to simplify terminal
equipment. Thus, for digital carrier, a signaling system not based on
tones is possible and is used extensively. This leads to considerably
simpler terminal equipment compared with SF units and, hence, to much
more economical arrangements. This low signaling cost, combined with
its inherent low transmission cost, has prompted the use of digital carrier
as an alternative to analog carrier with SF and also to wire facilities.
The manner in which the signaling information is digitally encoded is
now standard for facility terminal channel unit design. The channel unit
detects the per-channel direct current signaling input (for example, the
M-lead) and applies the necessary control to set the corresponding state
of the digital signaling bits. The digitally encoded signaling information
is then multiplexed (see Section 6.5) with the voice signals for transmission over the associated digital line. At the receiving terminal, the signaling information is recovered in the demultiplexing process. The
appropriate dc conditions are then applied toward the connecting circuit
(for example, the E-lead).
Since most digital signaling information is transmitted at a rate of at
least 1000 samples per second, the timing of the signaling information is
not appreciably disturbed. A state change at the source end is recognized
at the destination end with essentially no increase in delay caused by
sampling. For this reason, digital signaling is typically referred to as distortionless' compared with SF signaling, which typically distorts or
12 Section 6.4.3 discusses pulse-code modulation.

Network and Systems
Considerations

292

Part 2

changes the timing of short-duration signals. Since it is essentially free
from distortion, the digital system has the potential for providing superior performance, but it is, of course, facility dependent. Digital signals,
therefore, cannot be extended to other facilities such as wire or analog
carrier without conversion at the signaling interface.
Some stored-program control switching systems, including all timedivision switching systems, can provide an internal interface to digital
signaling without a digital-to-direct current conversion.

8.5.5 SIGNALING OVER SEPARATE FACILITIES COMMON-CHANNEL INTEROFFICE SIGNALING
The principle of the CCIS system is to transmit all of the signaling information pertaining to a group of trunks over a channel separate from the
communication channels. As shown in Figure 8-11, one data link can be
provided per trunk group, but it has been found to be much more
economical to use a data link in common for many trunk groups and

SIGNALING LINKS
OFFICE A

OFFICE B
TRUNKS

SIGNALING LINKS
SIGNAL
TRANSFER
POINT

SIGNAL
TRANSFER
POINT

SIGNALING
LINKS

TOLL
OFFICE A

SIGNALING
LINKS

TRUNKS
(16 TRUNKS PER BAND)

TOLL
OFFICE B

Figure 8-11. CCIS operating configurations. Top, associated mode;
bottom, nonassociated mode. In the nonassociated mode, signaling for A-B
trunks is by way of the signal transfer point. Both radio and cable paths
may be provided for signaling links with only one path in operation at a time
(the other is switched as needed) to provide redundancy with diversity.

Chap. 8

Signaling and Interfaces

293

"switch" the messages at a signal transfer point (STP). The STP is primarily concerned with the address of each message rather than with its
content.
eels permits a great reduction in call setup time, not only by its
inherent speed but also by signaling the next office in the route before an
office has finished switching. Other inherent advantages are flexibility
and low cost where the volume of signaling is substantial. These advantages stem from the fact that the signaling function is fully disassociated
from the voice path and is handled in a highly concentrated form. By
the same token, eels could introduce new problems unless special precautions are taken. For example, since signaling does not take place on
the talking path, the signaling process does not inherently verify the
transmission quality of the trunk. Therefore, it would be possible to
route a call over a defective trunk. To avoid this problem, a test of the
voke path is made part of the procedure for setting up each connection.
Another concern is the possibility of simultaneously disabling all of the
trunks (perhaps 4500) served by a eels link in the event of failure of that
link. Duplication of STPs with diversity of facilities (for example, facilities connecting the same points over different geographical routes)
provides the required reliability.
eCls represents the first major Bell System application of packet switching.13 Messages are transmitted as signal units of twenty-eight bits,
including eight bits for error detection. Twelve signal units are linked
into a block, or packet, for purposes of further error detection and
retransmission. With the STP architecture, each group of sixteen trunks
in toll office A (in Figure 8-11) having the same destination is given a
unique numerical assignment (called a band number) on the CCIS link to
the STP. The STP is arranged to transfer messages from the incoming
band and link to an associated outgoing band and link in order to reach
the distant office (toll office B) and the other end of the group of sixteen
trunks. The present CCIS protocol (see Section 8.8) was designed to be
optimal for medium-speed data links. This protocol, called CCITT Signaling System No.6, or System No.6, is used internationally and domestically.
The domestic version varies somewhat from the international version.
The initial domestic application employed 2.4-kbps data links. This speed
was later increased to 4.8 kbps for increased capacity, whereas the international application operates at 2.4 kbps. The domestic version of System
No.6 provides for generalized communications between computers not
just the relaying of supervisory and address signals related to trunks. For
example, in Automated Calling Card Service, customers at most coin telephones first dial 0 plus a destination number that is sent via the CCIS
network to a data base. After receiving a prompt, customers dial their

13 Section 5.8 discusses packet switching.

294

Network and Systems
Considerations

Part 2

calling card number (a billing number, often the home telephone number
plus a personal identification number), which is also sent via eels to the
data base for validation. A favorable response then initiates the call and
permits subsequent billing of the call to the billing number. No telephone operator is involved.
Another protocol, called eelTT Signaling System No.7, has been
specified to be optimum for high-speed digital data links (64 kbps). It
uses signal units of variable lengths, a functional structure of levels, and
a more straight-forward method of message addressing. System No.7
will be introduced into parts of the domestic telecommunications network in the near future.
In Section 8.4.2 on Per-Trunk Signaling, there is little discussion of
two topics that become more important with the introduction of eels:
reliability and availability. Reliability is the degree of assurance that the
received signal is the same as the transmitted signal; that is, it has not
been altered by electrical dropout or interference. Availability is the measure of continuity of signaling service between two points. In per-trunk
signaling, the signaling implementation is conservatively designed and
uses the same path or channel as the call to which it relates. The dc and
ac signals used with per-trunk signaling are relatively immune to transient electrical interference. As a result, if the path is adequate for voice,
it will usually transmit the signaling information properly. With eelS,
however, the capabilities of the channel are exploited to achieve a higher
information transmission rate. Thus, transient electrical interference can
cause occasional bit errors. To minimize the effect of such errors, redundant information is added to the outgoing message in the form of errorchecking codes. A parity code in which the parity bit is 0 for odd data
and 1 for even data is an example of simple error-detection coding. The
receiver can then detect most errors and call for retransmission of the
message segments.
The status of eels in 1983 is that it is spreading rapidly in toll applications. Projections of load due to trunk-related signaling and for generalized communications indicate the need for higher capacity links and
STPs in mid-1985. To meet this need, a new protocol optimized for 64kbps digital links will be adopted. A new STP architecture is also
planned for much higher capacity.
eels is of less value in the local network because of the signaling
already provided with digital facilities (see Section 8.5.4). However, the
value of features made possible by eels may motivate growth in local
areas as well.

8.6 INTERNAL INTERFACES
An end-to-end connection between customers on a network such as the
public switched telephone network (PSTN) requires a number of internal
connections among pieces of transmission, switching, and signaling

Chap. 8

Signaling and Interfaces

295

equipment. As indicated in Section 8.1.2, specification of interfaces for
various equipment boundaries
• ensures the compatible interconnection of equipment on each side of
the interface,
• separates design responsibilities for equipment on each side of the
interface, and
• permits independent design and testing on each side of the interface.
Besides ensuring compatible operation initially, well-designed interfaces usually facilitate the introduction of new equipment and services as
technology evolves. The interface specification permits new equipment
to be designed and integrated into the network without requiring
redesign of mating equipment. Stability is a desirable objective for an
interface, and some internal interfaces have been stable for many years.
However, achievement of cost savings and other benefits offered by new
technology sometimes requires changing traditional interfaces or creating
entirely new ones.
The next three sections describe three important internal interfaces:
• the 4-wire analog carrier interface, which has provided a stable
boundary between existing and new transmission and signaling
equipment for many years
• the DSX-1 digital system cross-connect, which is a relatively new
interface developed in response to the rapid growth of digital facilities
in the network
• the digital carrier trunk (OCT), which is replacing a traditional interface between the lESS system and T-carrier facilities.

8.6.1 THE 4-WIRE ANALOG CARRIER INTERFACE
Figure 8-12 is a simplified representation of a connection between two 2wire switching offices (offices A and C).14 In the figure, an analog voice
channel is carried from switching office A by a cable pair to building B,
which contains only transmission equipment. From building B.. the channel proceeds to switching office C over a 4-wire analog carrier system.
(Office C illustrates another common situation-switching and transmission terminal equipment housed in the same building.) The terminating
sets in building B and office C provide the hybrid function described in
Section 6.6.1; the analog channel banks (see Section 9.3.5) perform the

14 The connection shown in the figure is one of many possible configurations. Different
equipment and interfaces would be involved if the transmission and/or switching
systems were digital. The interfaces described in Sections 8.6.2 and 8.6.3 illustrate some
of these differences.

296

Network and Systems
Considerations

Part 2

required modulation and multiplexing functions; and T and R pads provide the desired loss between the 2-wire switching systems and the channel banks.
A classic example of an established, successful, pervasive interface is
the 4-wire transmission interface between the single-frequency (SF) signaling unit (which performs the function of conversion between dc and
SF signaling) and the A-type channel bank in the analog carrier transmission system. Two such internal interfaces are shown in Figure 8-12-one
at point X and one at point Y. Each interface consists of four wires,
designated T, R, T1, and R1, that carry message signals (for example,
voice, data) and SF network control signals (for example, supervisory,
address). The T and R leads carry signals from the SF signaling unit to
the modulator input port of a channel unit in a channel bank. The T1
and R1 leads carry signals from the demodulator output port of a channel
unit in a channel bank to the SF signaling unit.
The 4-wire interface is clean and simple, requiring only specification
of the impedance in either direction (nominally 600 ohms), the power
level in the transmit and receive directions (in terms of transmission
level points, described below), and balanced pairs. IS
Transmission Level Points
The transmission level point (TLP) concept is convenient for relating signal
or noise levels at various points in a connection. As part of an interface
specification, it facilitates the independent design and test of equipment
by defining the signal amplitude or power level at the input and output
terminals of the equipment.
A TLP does not define power in absolute terms but in relative terms.
Specifically, the TLP at any point in a circuit is defined as the ratio, in
decibels (dBs), of the power of a signal at that point to the power of the
signal at a reference point. It is customary to consider the outgoing 2wire class 5 switching system as the O-dB TLP reference point.
The transmission losses and gains incurred during transmission from
office A to office C are plotted on the upper scale in Figure 8-12; the
losses and gains incurred during transmission from office C to office A are
plotted on the lower scale. The portion of the upper scale beginning at
office A and ending at building B plots the losses (in dB) from the 2-wire
switching system in office A (O-dB TLP) to the 4-wire analog carrier interface at building B (-16 dB TLP). Analog carrier systems (which include

15 A balanced pair consists of two wires (tip and ring) that are electrically alike and
symmetrical with respect to ground. This circumstance has the advantage that noise and
interference currents induced equally in both wires flow in the same longitudinal
direction along the pair and are canceled out using a transformer-coupled load
impedance.

TLP
IdB)

TLP
IdB)

WEST TO EAST TRANSMISSION LEVEL OIAGRAM

+10 - - - -

------ -

---

o t---------....:=-l_-- - - - ---------1""--__+-__-.1

-- -

--- -- -

-10r------------------------ -----20 r- -

-

- -

-

-

-

- - -

-

-- - -

- -

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

SWITCH T

--L...L.!-.

~--.... T

r=t8==

I

2-WIRE
CABLE
PAIR

PAD

,....-.l..-...&-....,

B.5-dB LOSS

TRUNK ~~--+-+--f 4-WIRE
CIRCUIT
R
TERM SET
0.5-dB LOSS Hf-:-::--:--:-::-+-+-I3.5-dB LOSS
13.5-dB
~1.5-dB
LOSS
LOSS)

-~---c....J--~-

I

PARD
SWITCHING OFFICE A

SF
S
I
G
N
A
L
I

+-t1
Rl
r--

I

"6

-

-

-

-

- -

-

-

-

-

BUllDIN~

-

-

-

-

+7 dB TLP,\

~t----------__t~.....-t

DEMOD
OUT

--- --- -

- - +10

0

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -10

..

AH B
A
N N
NE K
L
T1....

U
N
I
T

GN

MOO IN
C

~

-

----- --------- ---------

--

(,. -16 dB TLP

2-WIRE

-

4-WIRE ANALOG
CARRIER SYSTEM
123 dB GAIN)

........
....

-

-

-

-- - -

-

-

T1

tt* '-

OEMOO
SF
OUT ~
C
S
H
I
A B
G U
NAN N
N N
A I
E K
L T
L
T
~
MOD IN

'7 dB TlP

-

-j
r-

PAO

II~

-16 dB TlP

-

-

-

-

-

-20

2-WIRE
T

T SWITCH

4-WIRE r-- TRUNK L2-L....L..TERM SET R
CIRCUIT I R L-...J
3.5-dB LOSS r-- O.5-dB LOSS ~ -l..--

I

~12-dB
LTOSS

G

- -

I

~
5-dB LOSS

-

I

L-...J

PAD t - - -............

__

!-+-__+----J

-+___

SWITCHING OFFICE C
+10 r- -

- -. -

- - - -

-

-

-

-

- -

-- -

-

-

---

-- - -

O---------~~----------10

f-

-

-

-

-20

>- -

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-------------- ------

-

- -

-

-

-

-

-

-

-

-

-

-

-

-

-

-

x

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

EAST TO WEST TRANSMISSION LEVEL DIAGRAM

-

-

-

-

-

-

- I-- -

y

Figure 8-12. Circuit layout of a 2-wlre/4-wire carrier-derived channel
showing transmission levels and internal interfaces.

-

-

-

-

-

-

-

-

-

-

-- -

- - +10

---~;;;;;..;;;;;+=_=-r--___1

o

-

-

-

-

-

-

-

-

-

-

-

-

-10

-

-

-

-

-

-

-

-

-

-

-

-

-20

Chap. 8

Signaling and Interfaces

299

the channel banks) are designed to provide 23 dB of gain; therefore, a
O-dBm 16 test tone applied at the O-dB TLP will emerge from the carrier
system OEMOO OUT port in office C at a power level equal to +7 dBm.
It should be noted that the absolute power of a particular signal that
exists at any point along this circuit depends upon the service involved.
For example, the standard power for a data signal is -13 dBmO, which
implies a data signal power equal to -13 dBm at the O-dB TLP. Such a
data signal would have an absolute power of -29 dBm at the MOD IN
(-16 dB TLP) port and -6 dBm at the OEMOO OUT (+7 dB TLP) port of
the 4-wire interface.

8.6.2 THE DSX-l DIGITAL SYSTEM CROSS-CONNECT INTERFACE
The evolving Bell System digital transmission network is composed of a
great variety of digital signals, multiplexers, and transmission facilities
that operate at different levels or bit rates. Some of the digital signals
originate as voiceband analog signals, others as data. The initial multiplexing of the analog signals takes place in channel banks as one of
several functions that include the translation from analog to digital format. The result is the formation of the digital signal 1 (OSl) level signal
(1.544 megabits per second [Mbps]), which forms the basic building block
of the digital time-division multiplex hierarchy.17 The OSl signal may
result from the multiplexing of lower speed digital signals.
The OSX-1 cross-connect is a relatively new internal interface that
provides a convenient central point for cross-connecting, rearranging,
patching, and testing digital equipment and facilities at the OSl level.
Figure 8-13 depicts the extensive variety of OS! signal sources, multiplexers connecting to higher levels, and transmission facilities served by
the OSX-1 cross-connect. For example, at the top left of the figure, a signal at the OSl level is formed by the time-division multiplexing of
twenty-four voiceband signals, each of which has been converted to digital form by pulse-code modulation. The modulation and multiplex functions are performed by the digital channel banks (01, DID, or 03) that
output the OSl signal to the interface at the OSX-1 cross-connect. (Some
sources produce more than one OSl signal. For example, the 02 channel
bank produces four OSl signals as shown in the figure.) At the other
side of the OSX-1 interface, the OSl signal may be connected directly to
another terminal or to a transmission system such as the T1 carrier system
(see Section 9.4.2) or combined through multiplex equipment to produce
a digital signal at a higher level (bit rate).

16 A logarithmic measure of power with respect to a reference power of 1 milliwatt.
17 Section 9.4.3 describes the digital time-division multiplex hierarchy. The various levels
or bit rates in the hierarchy are identified by a DSn (where n = 1, 2, 3, etc.) designation.

SOURCES

r

NUMBER OF DS1 SIGNALS
DSX-1
CROSS-CONNECT

....-------, 1
VOICEBAND
ANALOG SIGNALS

24

TO HIGHER
LEVEL
CROSS-CONNECTS

1

VOICEBAND
ANALOG SIGNALS

96

VOICEBAND
ANALOG SIGNALS

48

MULTIPLEXERS

VOICE SIGNALS
1·12

r--*--,
DSX-1C
(3.152 Mbps)

*
DSX-2
(6.312 Mbp8)

DATA SIGNALS
1-12 (56 kbp8)
DATA SIGNALS
(56 kbp8)

DSX-3
(44.736 Mbp8)

23

DATA SIGNALS
(9.6 kbp8)

DSX-3
(44.736 Mbp8)

L ______

.J

VOICE SIGNALS
48

FACILITIES

DS120
120 VOICE
OR DATA
CHANNELS

{
DS120

24 SWITCHED VOICE
OR DATA CHANNELS

30 VOICEBAND
CHANNELS

TWO LMX GROUPS
(EACH CONTAINING
12 VOICE
CHANNELS)

* THE TtWB BANK ACCEPTS A TOTAL OF 24 SIGNALS COMPRISING FROM
DACS
DCM
DCT
DDS
DFI
DIF
DT
OS
TASI E

1 TO 12 VOICE SIGNALS AND FROM 1 TO 12 DATA SIGNALS.

DIGITAL ACCESS AND CROSS-CONNECT SYSTEM
DIGITAL CARRIER MODULE FOR DMS·10 SYSTEM
DIGITAL CARRIER TRUNK FOR 1 ESS SYSTEM
DIGITAL DATA SYSTEM
DIGITAL FACILITY INTERFACE FOR 5ESS SYSTEM (15 DF18-1 INTERFACE MODULE)
DIGITAL INTERFACE FRAME FOR 4ESS SYSTEM
DIGROUP TERMINAL FOR 4ESS SYSTEM
OUTSTATE AREA
TIME ASSIGNMENT SPEECH INTERPOLATION SYSTEM. VERSION E

Figure 8-13. DSX-1 (1.544-Mbps) digital system cross-connect. the
figure shows a representative, but not exhaustive, collection of equipment
and facilities served by the DSX-1 cross-connect at the end of 1982.

Chap. 8

Signaling and Interfaces

301

The DSX-1 interface specifications summarized in Table 8-5 describe
the signal characteristics required of all DS1 signals appearing at this
interface. As can be seen from the table, the DSX-1 interface requires a
more complex specification than the 4-wire analog carrier interface
described earlier. In addition to the parameters shown in Table 8-5, there
is a requirement on the signal pulse shape at the interface 18 and on the
maximum jitter impairment that connecting terminals or facilities can
tolerate. Unlike the 4-wire carrier interface, the transmission level point
is the same for both directions of transmission at the DSX-1 interface.
TABLE 8-5
DSX-l INTERFACE SPECIFICATIONS*
Parameter

Requirement

Line rate

1.544 Mbps ± 130 pulses per minute

Line code

Bipolar with at least 12.5% average l's density
and no more than 15 consecutive zeroes

Test load impedance

100 ohms resistive

Pulse amplitude

3 volts ± 20%

Pulse shape

Rectangular (50% duty cycle) fitting template
for an isolated pulse

Signal power

12.4 to 18.0 dBm (771 to 773 kHz)
<-16.6 to -11.0 dBm (1543 to 1545 kHz)

Pulse power balance

Difference in power between positive and
negative pulses <0.5 dB

"Bell Laboratories 1982 contains a detailed discussion of the elements in this table.

8.6.3 THE DIGITAL CARRIER TRUNK INTERFACE

The use of electronic switching systems has created new, more efficient
transmission and switching internal interfaces in terms of space, cost, and
required maintenance. These interfaces have phased out several traditional internal interfaces. Examples of the new interfaces are listed in
Table 8-6.
18 AT&T 1978 contains the specification for the pulse shape.

Network and Systems
Considerations

302

Part 2

TABLE 8-6
ELECTRONIC SWITCHING SYSTEM
TRANSMISSION /SWITCHING
INTERNAL INTERFACES
System

Internal Interface

1/IAESS

Digital carrier trunk

2/2BESS

Direct interface with T-carrier

3ESS

Direct interface with T-carrier

4ESS

Digroup terminal (DT)
Digital interface frame (DIF)

5ESS

Interface module (1M)
(composed of 15 digital facility
in terfaces [D Fls])

DMS-I0

Digital carrier module (DCM)
Subscriber carrier module (SCM)
Office carrier module (OCM)

The following paragraphs illustrate how departure from a traditional
design, which segregates transmission and switching functions, toward
an integrated approach eliminates some previous boundaries between the
two functions. They compare the interface associated with the new digital carrier trunk (DCT) channel bank and the traditional interface used
with a D4 channel bank in a 1/ lAESS system.
The D4 channel bank interface illustrated at the top of Figure 8-14
represents the traditional separation of the 04 channel bank and the
trunk circuit (described below) associated with the switching system. The
04 channel bank is composed of a common, or control, equipment portion and a set of up to forty-eight individual channel units. Each channel
unit handles a single trunk or special-services line. The 04 bank
prepares analog signals from the switching system for transmission over
digital facilities through the OSX-l interface described in Section 8.6.2.
A trunk circuit connects the transmission facility to a specific terminal
on the switching network at which the trunk terminates. Trunk circuits
provide the interfaces that permit signaling over the trunk. Leads for
each trunk are required to exchange signaling and supervisory information between the trunk circuit and the channel bank through the intermediate distributing frame (IOF). The 1/ lAESS system central control

04 CHANNEL BANK
CHANNEl.
UNIT

SWITCHING
NETWORK

--,

I
I
I

INTERMEDIATE
DISTRIBUTING
FRAME

TRUNK
DISTRIBUTING
FRAME

CHANNEL
UNIT
2

~

I
I

•
•
•

I

I

I

COMMON
EaUIPMENT

T-CARRIER
FACILITY

l-

I

CHANNEL
UNIT
47

-~I
I

DSX-1
CROSSCONNECT

I
I

CHANNEL
UNIT

_.J

48

SIGNAL
DISTRIBUTOR
AND SCANNER

OCT CHANNEL BANK

--,

SWITCHING
NETWORK

I

I

I

CHANNEL
UNIT
2

••

•

~

I

I

I
I
I

I
I

T-CARRIER
FACILITY

COMMON
EaUIPMENT

I

CHANNEL
UNIT
47

~I

DSX-1
CROSSCONNECT

I
I

CHANNEL
UNIT

I
_..1

48

I

I
I

PERIPHERAL
BUS

PERIPHERAL
UNIT
CONTROLLER

I

J

Figure 8-14. Traditional and modern interfaces between 1/A1ESS and Tcarrier transmission facilities. Top, traditional interface; bottom, modern
interface.

Chap. 8

Signaling and Interfaces

305

interfaces with the trunk circuit directly via a signal distributor and
scanner.
The DCT combines certain T-carrier transmission functions and electronic switching system control functions to provide a more economical
interface between the 1/ lAESS system and T-carrier facilities. As shown
at the bottom of Figure 8-14, the DCT is a single, integrated channel bank
frame that replaces the D4 channel bank, the trunk circuit, the signal distributor and scanner, the intermediate distributing frame, and the pertrunk leads. The DCT channel unit combines the D4 digital channel unit
and the switching system trunk circuit functions. The exchange of signaling, supervisory, and maintenance information with the switching system central control is accomplished through the digroup19 controller and
the peripheral unit controller of the DCT frame. The peripheral unit
controller includes a duplicated WE 20 8000 microprocessor.

8.7 INTERFACES FOR INTERCONNECTION
8.7.1 INTERCONNECTION

Interconnection, as discussed here, is the direct electrical connection to the
nationwide telephone network of (1) user-premises terminal equipment
and communication systems or (2) the facilities of other common carriers
(OCCs). It represents the relationship established for individual telecommunications services including not only the physical and electrical
characteristics but also the maintenance and administrative procedures
that have been agreed upon by the participants. Certain long-standing
relationships, such as those between the Bell System and other telephone
companies, may fit the basic concept of interconnection but are not
included here.
A few definitions may be helpful in understanding the concept of
interconnection.
• Nationwide telephone network refers to the combined telecommunications facilities networks of the various telephone companies
throughout the United States, including the Bell System. The Bell System network refers to that part of the nationwide network owned and
operated by Bell companies.
• Facilities is the telecommunications physical plant. The Bell System
network as a whole is referred to as the facilities network (to distinguish
it from traffic networks)}1 and a single channel provided by the facilities network is referred to as a facility.
19 A digroup consists of twenty-four vo~ce channels.
20 Trademark of Western Electric Co.
21 Chapter 4 discusses the facilities network and traffic networks.

306

Network and Systems
Considerations

Part 2

• Terminal equipment is any separately housed equipment unit (a telephone set, for example) or group of equipment units treated as an
entity (for example, a PBX and its on-premises stations). Such equipment is located on user premises and is interconnected with a Bell
System network facility.
• User-premises communications system refers to those cases where equipment that is part of a separate communication system (for example,
one end of a microwave facility) is interconnected with a Bell System
facility.
• Other common carrier is a telecommunications common carrier, other
than the Bell System, authorized to provide a variety of private-line
services such as facsimile, measured voice, and wideband data as well
as packet-switched digital data. The Federal Communications Commission (FCC) refers to these carriers as domestic satellite carriers,
miscellaneous common carriers, and specialized common carriers. Many
of them rely on Bell System local facilities to provide access to their
facilities.
• Separate ownership is usually, but not necessarily, implied in the concept of interconnection. Separate ownership is involved where the
terminal equipment interconnected with Bell System facilities is provided by the user. However, in those cases where the user leases the
terminal equipment from the Bell System, the FCC has ruled that the
Bell System must meet the same cr:iteria for interconnection as is
required of user-provided terminal equipment. Thus, the FCC rules
governing network protection (see Section 8.7.5) apply equally in both
cases. In any case, from the Bell System viewpoint, interconnection
always involves the Bell System as at least one of the interconnection
participants.
8.7.2 INTERFACES

In the interconnection environment, an interface is the boundary between
two telecommunications capabilities. The physical connection may be
either a jack and plug or a terminal strip for matching leads that are common to both telecommunications capabilities. Compatible electrical
characteristics (such as impedances and signal power levels) are essential
not only for proper operation but also to ensure that there is no interference with normal Bell System network functions. The interface may also
be a boundary for maintenance and administrative functions.
Interfaces for interconnection are referred to as either network interfaces or demarcation points. Where user-provided terminal equipment is
interconnected with the Bell System facilities network, the interface is
referred to as a network interface. Interfaces for interconnecting acc facilities to the Bell System facilities network are also network interfaces but

Chap. 8

Signaling and Interfaces

307

are commonly referred to as demarcation points. The two types of interfaces differ in two respects: (1) the physical and electrical characteristics
are not necessarily the same for a given service and (2) related functions
are handled differently.
Network interfaces tend to have standardized electrical characteristics,
pin arrangements, and protocols, particularly where FCC registration of
terminal equipment is involved (see Section 8.7.5). With such arrangements, the maintenance and administrative functions can be minimized
to ensure design and operating flexibility.
Demarcation points tend to be negotiable, since testing activities
extend across the interface and a more interactive approach is used for
circuit layout. But there is a trend toward standardized DCC interfaces.

8.7.3 INTERCONNECTION ENVIRONMENT
Bell System As Sole Provider of Service
The interconnection environment in those cases where a Bell System network facility is the sole provider of the telecommunications service is
shown in Figure 8-15.
In such cases, the network service is defined only between network
interfaces (NIs), and the user of the service is traditionally referred to as
the customer. Thus, the terminal equipment is located on the customer
premises.
Usually the NI is identified by some characteristic of the service with
which it is associated. For example, a 2-wire (tip and ring) interface with
loop-start signaling for interconnecting a telephone set to a PBX offpremises station (DPS) line is referred to as an DPS interface (station
end). In those cases where digital transmission is involved, the NI may
be identified by the bit-stream characteristics of the digital signal, for
example, the DS1 interface described in Section 8.6.2. In some cases, an

I

~

BELL SYSTEM
SERVICE
BELL SYSTEM SWITCHED
OR PRIVATE-LINE FACILITY ~

'~

__________________________- J

CUSTOMER PREMISES

EEl

EQUIPMENT-TO-EQUIPMENT INTERFACE

NCTE

NETWORK CHANNEL-TERMINATING EQUIPMENT (LIGHTNING
PROTECTOR WITH OR WITHOUT OTHER CHANNEL-TERMINATING
FUNCTIONS)

NI

NETWORK INTERFACE

TE

TERMINAL EQUIPMENT

I
~

:/ .r:l // r:l

N~~
~------------------~/
CUSTOMER PREMISES

Figure 8-15. Interconnection environment-Bell System services only.

N etworl< and

308

~ystems

Considerations

Part 2

NI does not have a service identifier. For example, the 2-wire loop-start
NI mentioned above is referred to simply as a network interface where it
is used for interconnecting a telephone set with a local central office line
for access to anyone of several Bell System PSTN services.
The equipment-to-equipment (EEl) interface shown in Figure 8-15 is
an important part of the interconnection environment but is not an interface for interconnection since it does not involve a Bell System network
facility as one of the interconnection participants. There is growing
interest in developing standards for those interfaces that are applicable to
both EEls and Nls to facilitate portability of terminal equipment. Achieving such commonality among interfaces for interconnecting terminal
equipment with public switched data networks and other digital services22 is an objective of the current efforts in both domestic and international standards activities.
The network channel-terminating equipment (NCTE) shown in Figure
8-15 is part of the network facility. It may provide lightning protection
for the terminal equipment in those cases where the facility involved is
exposed to lightning. It may also include other functions such as loopback for remote testing on a 4-wire line, facility loss equalization, and
digital signal regeneration. 23 The NCTE is not a network protection
device; however, it may include inherent network protection such as
hazardous voltage limiting (for example, diodes) and isolation from longitudinal imbalance (for example, transformers) and signal power contro1. 24

DCC Service Using Bell System and DCC Facilities

The interconnection environment also includes those cases where an
acc service is provided by a combination of Bell System and acc network facilities, as shown in Figure 8-16. In such cases, the Bell System
provides the connecting facility between the user premises and the acc
facility. The user is referred to as a patron instead of a customer since the
acc has the overall service responsibility, and the interface between
NCTE and acc channel equipment (aCE) is referred to as a demarcation
point (DP). The equipment configuration on the patron premises in Figure 8-16 is the same as previously described for those cases where a Bell
System facility is the sole service provider (EEl is not shown), except that
aCE may also be required in certain cases. Where such equipment is
included, both a network interface and a demarcation point are required
on the patron premises as indicated in Figure 8-16. Typically, aCE on
22 Section 8.8 discusses data communications services.
23 The FCC has not ruled on which functions may be included in the NCTE.
24 Section 8.7.5 discusses inherent network protection.

... 1

OCC SERVICE

BELL SYSTEM

CC
OCE

CONNECTING
FACILITY

"

DP

PATRON PREMISES

~

/

OCC PREMISES

OCC SERVICE

BELL SYSTEM
OCE

PRIVATE-LINE
FACILITY"

DP

CONNECTING
FACILITy

OCC PREMISES

- - - . AS, ABOVE

I

r::-L~~
U
~~. 0
~~ U

BELL SYSTEM
JIll"'" CONNECTING FACILITY

,

NI

DP

/

PATRON PREMISES

DP

DEMARCATION POINT

NCTE

NETWORK CHANNEL-TERMINATING EQUIPMENT

NI

NETWORK INTERFACE

OCC

OTHER COMMON CARRIER

OCE

OCC CHANNEL EQUIPMENT

TE

TERMINAL EQUIPMENT

Figure 8-16. Interconnection environment-OCC service with Bell
System connection facilities. Top, no aCE required; bottom, equipment
configuration when aCE is required.

PATRON PREMISES

l""CI.VVVJ.A ClllU.

310

.:Jyi:lll.Cl1l~

Considerations

Part 2

patron premises might provide a single-frequency signaling function.
Otherwise, OCE is located on OCC premises.
The Bell System may also be the sole facility provider between the following premises:
• OCC premises and OCC premises
• patron premises and patron premises
• OCC premises and patron premises
• patron premises and a Bell System central office building
• OCC premises and a Bell System central office building.
The configuration for such cases is the same as that shown in Figure
8-15 except that an NI is a DP, customer premises becomes patron premises, and in the latter cases, one of the premises is replaced by a central
office building.
Typically, the services provided might include a 2-wire control facility
between two OCC premises or between two patron premises for transmitting dc opens and closures for purposes such as status indication and
telemetering, a 2-wire control facility between an OCC premises and a
Bell System central office for echo suppressor disabling, or a video cable
facility between two OCC premises or between two patron premises.
8.7.4 EVOLUTION OF INTERFACES FOR INTERCONNECTION

The Bell System has been an active participant in identifying and standardizing network interfaces for many years. Prior to the 1968 FCC Carterfone Decision (see Section 17.2.7), which permitted direct electrical
connection of user-provided terminal equipment to Bell System and other
telephone company facilities networks, much effort was directed toward
standardizing interfaces for connecting customer-provided business
machines to Bell System data equipment. Through such activities in the
Electronics Industries Association (EIA), standards like the EIA RS-232-C
and RS-366 were developed.
Direct electrical connection of customer-provided terminal equipment
to the Bell System facilities network required a host of new interfaces to
accommodate the existing Bell System services. At first, this need was
filled by connecting arrangement service, an adjunct to existing services
for those cases where network protection was required. A family of new
interfaces was developed for connecting arrangement service, each interface being defined by a network protection device. Later, the FCC ruled
that such protection devices could not be required and that, instead, terminal equipment was to provide the network protection. Such protection
is required for all services except those classified as inherently protected
(see Section 8.7.5).

Chap. 8

Signaling and Interfaces

311

Interconnection between acc facilities and the Bell System facilities
network evolved as a result of FCC decisions dating as far back as 1949.
At that time, the FCC began licensing private microwave systems. These
systems were not interconnected with the Bell System facilities network
at first, but in 1969, the FCC permitted private microwave companies to
compete with the Bell System in the sale of telecommunications services.
In order for these companies (common carriers) to provide end-to-end
service, they were permitted to interconnect with local Bell System network facilities. At first, the interfaces were specifically defined to isolate
each carrier's area of responsibility with minimum maintenance, administrative, and design interaction. However, as a result of informal negotiations sponsored by the FCC, a more interactive means of supporting acc
interfaces was agreed upon (see FCC 1975). The new concept extended
maintenance activities across the demarcation point and provided a more
interactive design approach. Thus, even though standardization is
becoming more prevalent, demarcation points tend to be negotiable.

8.7.5 NETWORK PROTECTION

Protection of the network is required by the FCC to ensure that terminal
equipment on user premises does not cause physical damage to the Bell
System and other facilities networks, injure employees, or impair service
to other users (for example, usefulness to a third party). At first, when
the Carterfone Decision went into effect, network protection requirements (for example, electrical limitations) were specified in appropriate
tariffs, and other protection criteria were provided in various technical
references published by AT&T. Network protection was ensured where
possible by the network protection devices for connecting arrangement
service. Initially, the FCC permitted these devices to be interposed
between the terminal equipment and the facilities network.
The FCC also instituted proceedings to determine the conditions for
interconnection and what rules, if any, should be adopted for a more
suitable long-term solution to the network protection issue. As a result of
these proceedings, the FCC identified specific network "harms" that could
result from uncontrolled interconnection of terminal equipment and
ruled that network protection must be provided in all cases. These
proceedings led to the development of a comprehensive program, as
embodied in the FCC Registration Program, for ensuring that such protection is always provided.
FCC Registration Program
The FCC Registration Program, instituted in 1975, replaced all previous
programs for the direct electrical connection of terminal equipment to the
facilities network and created a need to define and standardize a new

312

NetworK and ~ystems
Considerations

Part 2

class of network interfaces. Each new interface is rigidly defined at connection points normally provided between equipment units (or between
an equipment unit and the network facility) on a customer's premises.
Network protection is ensured by requiring manufacturers of terminal
equipment to demonstrate compliance with FCC Part 68 Rules and Regulations. These rules establish a well-defined set of requirements for
interconnection.
In this regard, terminal equipment manufacturers must submit documented proof of compliance with the Part 68 requirements for the particular service(s) to which the terminal equipment will be connected. The
FCC then issues a registration number that must be affixed to each unit of
terminal equipment connected. For each type of service, all terminal
equipment connected to a network interface must be registered after a
specific date, known as the "register-only" date. Terminal equipment connected prior to that date does not have to be registered unless
modifications are made that affect compliance with the Part 68 requirements (see FCC 1977/1980).
Initially, the FCC Part 68 Rules and Regulations covered terminal
equipment, with certain exceptions, for interconnection with the
switched message network by a 2-wire interface with loop-start, groundstart, or reverse-battery signaling. The exceptions were PBXs, key telephone systems, main-station telephones, coin telephones, and terminal
equipment connected to private and multiparty lines. Subsequently, standard interfaces were identified for 4-wire station-line service and certain
private-line services including PBX message registration, PBX off-premises
station lines, PBX tie trunks, and PBX automatic identified outward dialing. The FCC Part 68 Rules and Regulations were amended to include
these services and main-station telephones. Standard interfaces have
since been identified for many other existing services including 2-point
private lines, multipoint private lines, local area data channels, and new
data services such as those prOVIded by DS1 facilities. However, to date,
the FCC Part 68 Rules and Regulations have not been amended to include
these services.

FCC Part 68 Rules and Regulations
FCC Part 68 Rules and Regulations specifies the requirements that equipment manufacturers must meet in order to be able to market their product for connection to the network (see FCC 1977/1980). Provisions are
made for the means of connection, notification of the telephone company,
verification requirements and procedures, labeling requirements, and
specific physical and electrical requirements and limitations. The physical and electrical requirements and limitations include the following:
• environmental simulation: vibration, temperature, humidity, physical
shock, metallic voltage surges, and longitudinal voltage surges

Chap. 8

Signaling and Interfaces

313

• leakage current and hazardous voltage limitations to control electrical
hazards to telephone company equipment and personnel
• signal power and longitudinal imbalance limitations to control
crosstalk and interference on other telecommunications channels
• on-hook impedance limitations to control pretrip (cessation of ringing
before the called station actually rings) and to ensure the integrity of
certain network maintenance procedures
• other requirements and limitations to ensure the integrity of network
billing.

Inherent Network Protection
Inherent network protection refers to those cases where protection from
network "harms" is an inherent part of certain private-line services. In
such cases, hazardous voltage limiting and longitudinal imbalance
protection are incidental to the NCTE functions. Signal power compliance may be ensured either by telephone company procedures normally
associated with the service or by a limiting function otherwise provided
by the NCTE. For services in this category, terminal equipment compliance with Part 68 Rules and Regulations is not required. Private-line services presently designated as inherently protected include a variety of
types such as program audio, video, Washington Area Metropolitan
Channels, ISO-baud (code elements per second) telegraph, and others.
Certain digital services, for example, DATAPHONE digital service, are
also considered inherently protected. However, in those cases where a
digitally encoded signal level can be converted to an analog signal level
and the service can be connected to an analog network service, inherent
protection cannot be ensured. As currently proposed, signal power
requirements for such services would be included in the Part 68 Rules
and Regulations.

8.7.6 TYPICAL NETWORK INTERFACES FOR INTERCONNECTION
Two examples of existing network interfaces for interconnection are summarized in Table 8-7. One, referred to as a Digital Data System (DDS)
interface, interconnects terminal equipment with a DDS facility. The
other, referred to as a network interface, interconnects voice-frequency terminal equipment, such as a telephone set, with a switched message network facility.
The DDS interface is a 6-wire interface. Voltages used for status indication on two of the six leads conform to EIA RS-232-C. The physical

TABLE 8-7
TYPICAL NETWORK INTERFACES
DDS Interface

Network
Interface

Type of service

Private line

Switched

Signals
transmi tted

Digital

Analog, voiceband

Physical
interface

6-wire

2-wire

Connector

Jack and plug

Jack and plug

Network
protection

Inherent*

FCC registration

NCTE

CSU

Protector block

Bipolar violations and
control code

Dc, loop-start

Not required

-9dBm

Control
signaling
Transmit signal
power limit
Signal
characteristics:

Not specified

Service rates

56 kbps; sub rates of 2.4,
4.8, and 9.6 kbps

Code

Bi polar return to zero

Pulse shape

Rectangular, 50-percent
duty factor

Timing

Encoded in received bit
stream

* Hazardous voltage, longitudinal balance, and signal power protection in channel service
unit (CSU).

Chap. 8

Signaling and Interfaces

315

connection is a IS-pin jack and plug (with only six pins used) with leads
designated as follows:
DTI,DRI
DT,DR
SI, GND

Data transmit
Data receive
Status indication

These and other interface requirements are specified in the Bell System technical reference for the service. The NCTE, in this case, is a Bell
System channel service unit (CSU) that is part of the DDS facility. The
CSU provides several network functions. In the transmit direction, it
restores the bipolar signals from the terminal equipment to a level suitable for application to the DDS facility. In the receive direction, it provides facility loss equalization, establishes a reference voltage level, and
converts incoming signals to uniform bipolar pulses. In addition, the
CSU provides loopback for the 4-wire line to facilitate the isolation of
troubles from a remote test center.
Other DDS interface items listed in Table 8-7 are DDS characteristics
that are part of the interface description required to distinguish it from
other 6-wire interfaces. Control signaling, accomplished by control codes
in the bit stream together with bipolar violations of the pulses, establishes the idle, out-of-service, and test (that is, loopback) conditions. A
transmit signal power limit is not necessary since the signal power is limited by the CSU. This, together with the fact that the CSU incidentally
provides hazardous voltage and longitudinal imbalance protection,
qualifies the service as inherently protected. The last item in the table,
signal characteristics, lists some of the other DDS characteristics that distinguish this interface from other network interfaces.
The network interface for interconnection with the switched message
network is typically a 2-wire interface for voiceband analog terminal
equipment. The physical connection is a modular jack and plug with
leads deSignated T (tip) and R (ring). The NCTE for this case is typically
a lightning protector and may, in some cases, include a maintenance terminating function. Terminal equipment connected to this interface must
be registered.
Other items in Table 8-7 for -the switched message network interface
define the interface as unique to services with loop-start control signaling. Loop-start establishes the idle, off-hook, and dialing conditions. The
-9 dBm signal power limit is also a part of the interface definition since
it signifies a need for signal power control in the connected terminal
equipment. It is the highest 3-second average signal power that can be
applied to the network facility by the terminal equipment without potential for network "harm." In this case, potential network "harm" is crosstalk

316

Network and Systems
Considerations

Part 2

to a third party as a result of carrier overload. The limit does not apply
to live voice signals or to network control signaling.

8.8 DATA COMMUNICATIONS INTERFACES AND
PROTOCOLS
Data communications is concerned with the transfer of information in
digital form between users. Information must be transferred in a manner
that preserves its meaning; the communications mechanism must provide
undistorted information transfer.
There is more to data communications than just the physical movement of bits between devices. The diverse nature of the information
being transferred places varying requirements on the communications
mechanism (for example, low transmission delay, high quality, low cost,
etc.). In addition, transferring information between source and destination may involve traversing several different types of communications
networks (for example, the direct distance dialing network, a packetswitching network, and a local area network). Thus, data communications generally involves a complex mechanism requiring precise rules to
govern the transfer of information. These rules and their syntax (format)
are called protocols.
A protocol may be thought of as a language used to interact with the
communications mechanism to establish, carry out, and terminate communications. Since, in general, the data communications process could
involve several systems and communications networks, the rules necessary to control the communications can become quite complicated. If
only one protocol were defined to control all aspects of the communications, it would be extremely complex. Rather than define a single
unwieldy protocol, the approach taken was to reduce the complexity to
manageable proportions by structuring the information transfer process
into smaller pieces. Extensive national and international effort has
recently resulted in a model called the Reference Model for Open Systems
Interconnection (051).25 The model partitions the functions required for
communications into seven hierarchical groups called layers (see Figure
8-17).
Each layer (or group of functions) provides capabilities, called layer
services, to the next higher layer by building on (enhancing) the layer service provided by the layer below. A protocol associated with each layer
controls the communications activities (functions) between entities26 in the
25 See ISO 1983 and CCITT 1983 for a discussion of this reference model.
26 A logical element in a given layer, within a system, which executes the layer protocols.
The logical element may be hardware, software, or a combination of both.

LAYER PROTOCOLS
APPLICATION

....oL

.......

6

PRESENTATION

...
...

APPLICATION

....oL

.......

PRESENTATION

5

SESSION

........

........

SESSION

4

TRANSPORT

.........

....

TRANSPORT

3

NETWORK

L...t

---..

..

NETWORK

2

DATA LINK

....oL

.......

...

DATA LINK

PHYSICAL

....
.....

...

PHYSICAL

7

I

.....
.....

....

.....

PHYSICAL MEDIA

SYSTEM A

J
SYSTEM B

Figure 8-17. Reference model for open systems interconnection.

layer residing in different systems (see Figure 8-17). Thus, the complex
task of information transfer has been broken down into seven smaller
tasks (layers) each of which has a protocol for controlling its activities.
The protocols of each layer are independent of one another; that is, when
a layer uses capabilities or services provided by the layer below, it does
not know or care how the capability was generated.
One of the major advantCiges of defining the model as a hierarchical
set of seven layers is that the end-users (users of the application layer,
which may be application programs, human beings, etc.) do not need to
be concerned with the details of the underlying communications mechanism. They need only be concerned with the information to be transferred
and the application at hand.
Another advantage and primary goal of the model effort is that by
standardizing the protocols for each layer, systems and networks built
independently but in conformance with the protocols can operate with
one another; this provides a potentially very large community of "open
systems."

8.8.1 LAYERS OF THE REFERENCE MODEL
A useful way to view the model is to consider the distinction between
the scope of the lower three layers and the scope of the upper four
layers. The lower three layers provide all of the capabilities required for
sending data over a network connection between end systems; that is,
they provide the capabilities to establish, carry out, and terminate connectionsthrough one or more networks. The upper four layers of protocol

317

Network and Systems
Considerations

318

Part 2

provide all of the end system-to-end system signaling for data transfer.
These protocols are carried transparently between end systems by the
lower three layers and are only interpreted by the end systems. (See
Figure 8-18.)

END_USER{ :

TO
END-USER

5

4

3
NETWORK
SPECIFIC

:

{

APPLICATION

........

PRESENTATION

.........

SESSION

....
.....

TRANSPORT
NETWORK

....
....
.........

DATA LINK

....
.....

PHYSICAL

....

I
END SYSTEM

A

....

......

.....
...

...

DATA LINK

.........
.........

PHYSICAL

....

NETWORK

.....

I I
COMMUNICATIONS
NETWORK

......
.....
.....

PRESENTATION

....

..

TRANSPORT

...

NETWORK

.......

DATA LINK

...

PHYSICAL

..

APPLICATION

SESSION

I
END SYSTEM

B

Figure 8-18. Protocol architecture.

The following paragraphs briefly summarize the functions of each
layer. It should be noted that each layer contains the layer-specific procedures for establishing, carrying out, and terminating its own activities
in addition to the specific functions mentioned below.
• The application layer provides protocols specific to particular applications (for example, airline reservations, credit checking, purchase orders) for the transfer of information meaningful to communications
users. These protocols are realized by building application-specific
protocols on top of the basic capabilities, called common application
capabilities, of the application layer. All of the services available to
users from the OSI environment are requested via the application
layer.
• The presentation layer is responsible for representing the information
to be transferred between applications in a manner that preserves its
meaning (semantics) while resolving differences in its representation
(syntax). Examples of different syntaxes include American Standard
Code for Information Interexchange (ASCII) versus Extended BinaryCoded Decimal Interchange Code (EBCDIC), and screen-oriented data
from a cathode-ray tube (CRT) terminal versus line-oriented data.

Chap. 8

Signaling and Interfaces

319

• The session layer is responsible for establishing, maintaining, and terminating an association between its users (presentation entities). It
also negotiates the appropriate dialog discipline (for example, which
user has the right to send at a given time) and has capabilities for synchronizing the activities of its users.
• The transport layer provides for the control of data transfer between
end systems, independently of how the connection between end systems is established. The transport layer may provide for end
system-to-end system acknowledgement, splitting of a transport connection onto many network connections, multiplexing many transport
connections onto one network connection, and the capability to detect
and, in some cases, correct errors in the data transferred between end
systems.
• The network layer provides the capabilities and procedures required
to control network connections (for example, set up, maintain, terminate) between end systems containing transport layer entities. The
network layer isolates its users (transport entities) from the specifics of
particular networks supporting the transfer of data.
• The data link layer enhances the basic capability of transferring bits
so as to provide high-quality ("error free") transmission of data over a
physical connection.
• The physical layer provides for the transmission of a bit stream over a
transmission channel in some physical communications medium.

8.8.2 INTERFACES

Standard data communications interfaces have been defined to facilitate
communications between pieces of equipment. At the point of demarcation between two pieces of equipment, two attributes of the interface are
defined: the physical media aspects (for example, electrical and mechanical) and the protocols.
An example of an important data communications interface is that
between data terminal equipment (DTE)-for example, a teletypewriter,
CRT, or host computer-and data circuit-terminating equipment (DCE)for example, a data set. This interface has been standardized in the
United States by the EIA and internationally by the CCITT and the International Organization for Standardization (ISO). The most widely
deployed DTE/DCE interface in the United States is EIA RS-232-C (see
EIA 1969). Other important interfaces include EIA RS-449 (see EIA
1977/1980) and CCITT Recommendation V.35 (see CCITT 1981a). These

Network and Systems
Considerations

320

Part 2

interfaces pertain to the physical layer of the model and to the physical
media (see Figure 8-19). Table 8-8 gives a listing of the interfaces provided by the Bell System data sets and Digital Data System (DDS) data
service units (DSUs).
The advent of digital network services has produced a new set of
interfaces at the network service boundary. Table 8-9 provides an illustrative listing of Bell System digital services and their respective interfaces. For packet-switching networks, CCITT Recommendation X.2S (see
CCITT 1981b) covers the DTE/DCE interface for packet-mode operation,
and Recommendation X.7S (see CCITT 1981c) covers the network-tonetwork interface. These interfaces consist of the physical media aspects
plus the protocols at the lower three layers of the model (see Figure
8_20).27 The physical layer and the physical media aspects can be, and
typically are, those described in the previous paragraph.
Another example is Bell System Protocol Specification BX.2S (see
AT&T 1980b) that was developed by the Bell System for use in the operations systems network (see Section 15.5). BX.25 fully specifies the actions
of the DTE in a way that permits either connection to a network that supports CCITT X.25 or direct connection to another BX.25 DTE. BX.2S now

....

APPLICATION

........

PRESENTATION

........

.......

5

SESSION

........

...

4

TRANSPORT

3

NETWORK

7

6

2

...

~

.....

....

I

PHYSICAL

....

I

.. I

I

"'1

I

DTE A

I
PHYSICAL

I...
I'"

... ,
....J

PHYSICAL

I l

I I

DCE A

DCE B

DTE/DCE
INTERFACE
\~------~ r ________~I

V

CUSTOMER PREMISES
A

DCE
DTE

.
..
...

DATA LINK

I

PRESENTATION

...

....

.oIIIIL

",.

APPLICA TION

I

L...

I ....

:

SESSION

.....

TRANSPORT

.....

DATA LINK

I

NETWORK

PHYSICAL

I
DTE B

DTE/DCE
INTERFACE

\~------~

V

r - - - - - - - - - J1

CUSTOMER PREMISES
B

DATA CIRCUIT-TERMINATING EQUIPMENT
DATA TERMINAL EQUIPMENT

Figure 8-19. Application of OSI model to modem interfaces.

27 As indicated in the description of the Basic Packet-Switching Service in Section 11.6.2,
the protocols in X.25 are referred to as the physical (or bit), link, and packet levels. They
correspond to the physical, data link, and network layers, respectively, of the OSI model.

TABLE 8-8
ILLUSTRATIVE LISTING OF
BELL SYSTEM DATA SETS AND DSUs*

Code

Technical
Reference
(PUB No.)

Interface

Voiceband Data Sets

103}
108F
108G
113C
1130

41106
41215
41215
41106
41106

EIA
EIA
EIA
EIA
EIA

RS-232-C
RS-232-C
RS-232-C
RS-232-C
RS-232-C

201C
202S
202T
208A
208B
209A
212A

41216
41212
41212
41209
41211
41213
41214

EIA
EIA
EIA
EIA
EIA
EIA
EIA

RS-232-C
RS-232-C
RS-232-C
RS-232-C
RS-232-C
RS-232-C
RS-232-C

407C

41409

EIA RS-232-C

2024A
2048A
2048C
2096A
2096C

41910
41910
41910
41910
41910

EIA
EIA
EIA
EIA
EIA

303
306

41302
41304

RS-449t
RS-449t
RS-449t
RS-449t
RS-449t

Wide band Data Sets

Coaxial cable
CCITT V.35

DDS DSUs

500B
500B

41450
41450

"Table 11-1 contains additional informatiol}.
t Adaptor available for RS-232-C.

EIA RS-232-C
CCITT V.35

TABLE 8-9
ILLUSTRATIVE LISTING OF BELL SYSTEM
DIGITAL SERVICES

Service

Technical
Reference
(PUB No.)

DDS

Use

Interface

Steed
( ps)

Operation

Type

41021

Private
line

6-wire

2.4,4.8,
9.6,56k

Full
duplex

Sync

Point to
point

DDS

41022

Private
line

6-wire

2.4,4.8,
9.6,56k

Full
duplex

Sync

Multipoint

DSI

41451

Private
line

9-wire

1.544M
1.344M

Full
duplex

Sync

Point to
point

CSDC

61310

Circuit
switched

8-wire

9.6,56k

Full
duplex

Sync

Alternate
voice/data

BPSS

54101

Packet
switched

6-wire

9.6,56k

Full
duplex

Sync

CCITT X.25

Remarks

CSDC ... Circuit-switched digital capability (see Sections 2.5.1 and 11.6).
BPSS ,.. Basic Packet-Switching Service (see Sections 2.5 and 11.6.2).

.....
....

~

....

APPLICATION

6 PRESENTATION ~

-

....
....

~

SESSION

5

~

4

TRANSPORT
t--------i-

3

~ I
NETWORK
t--------i""" I
~
DATA LINK
t--------i"""
PHYSICAL

,

I

I

I

..

I

"

r----J/

I

l

\~

..

PHYSICAL

_ _ _~

~--------~V~-----J

X.75
NETWORK/NETWORK
INTERFACE

DATA LINK

j4-f-+

PHYSICAL

L------..r,-----'

"
/

NET~RK 2

I

/

t--T_R_AN_S_P_O_RT--t

I ... t - - - - - - - - i

,,:, ,
NETW'<{RK 1

X.25
DTE/DCE
INTERFACE

I4-i-+

SESSION

..
NETWORK
... t - - - - - - - t

DATA LINK

...

\~________~

DCE
DTE

NETWORK ...
I ...
"""

ort....!..... DATA LINK ...
I...... I . . . . . . .
PHYSICAL ~ PHYSICAL
PHYSICAL ~
..

DTE A

\

~

...

m

'-----r-----'

I

NETWORK

APPLICA TION
PRESENTATION

\

DTE B

X.25
DTE/DCE
INTERFACE
/
~----~V~---~

CUSTOMER PREMISES

CUSTOMER PREMISES

A

B

DATA CIRCUIT-TERMINATING EQUIPMENT
DATA TERMINAL EQUIPMENT

Figure 8-20. Relationship of X.25 and X.75 to the OSI reference model.

322

Chap. 8

Signaling and Interfaces

323

has four layers. The lower three layers correspond to the DTE side of the
X.25 DTE/DCE interface. Currently, the uppermost layer of BX.25 combines the functions of layers 4, 5, and 6 of the model. This was done for
efficiency and expediency; the operations systems network needs at these
layers are relatively simple, and the standards committees had not completed the precise definition of these layers.

AUTHORS

P. D. Bartoli
H. V. Bertine
R. A. Fuller
W. S. Hayward
W. C. Roesel
W. R. Starrett
F. E. Weber

PART THREE
NETWORK AND
COSTOMER·SERVICES SYSTEMS

Building on the broad introduction to the Bell System in Part One and
the telecommunications fundamentals in Part Two, Part Three examines
network and customer-services systems and associated equipment. The
intent is not to catalog all systems, but to present advantages and disadvantages of selected systems relative to applications, describe system
operation on a functional level, and discuss the evolution and current
trends in each major area.
Chapters 9 and 10, respectively, describe many of the transmission
and switching systems in use in the Bell System. Chapter 10 also covers
billing systems, which are closely related to switching systems. Chapter
11 discusses the customer-services equipment and systems that provide
many of the services described in Chapter 2. Chapter 12 highlights three
types of common systems associated with the network and customerservices systems presented in Chapters 9 through 11. These types are
power systems, distributing frames, and equipment-building systems.
They are included here because, while they are not components of the
network and customer-services systems, they provide essential supporting
functions.

325

9
Transmission Systems

9.1 INTRODUCTION
This chapter describes many of the transmission systems now in use
throughout the Bell System network. 1 Section 9.1 identifies the major
applications for these transmission systems, namely, loops and the
interoffice networks that include metropolitan, outs tate, and long-haul
intercity. Since the characteristics of each of these areas are unique,
specific system features and capabilities are designed to match them. 2
There are three basic types of transmission systems: voice frequency,
analog carrier, and digital carrier. Section 9.2 covers voice-frequency
transmission systems, which are found extensively in the loop and metropolitan interoffice areas of the telephone plant. Sections 9.3 and 9.4
describe, respectively, analog and digital carrier transmission systems,
which have found application in the metropolitan, outstate, and longhaul areas.

9.1.1 AREAS OF APPLICATION
The Loop Plant
The connection between the telephone customer and the serving central
office is called a loop. Individual loops leave the serving central office
building in a large main feeder cable via an underground conduit system
that provides physical protection for the cable. Branching off from the
main feeder cable are smaller branch feeder cables and then still smaller

1 For a more thorough treatment of transmission system design, see Bell Laboratories 1982.
2 Sections 4.2 and 4.4 contain related information on the structure of the traffic and facilities
networks. It should be noted that Chapter 4 uses rural rather than outstate.

327

Network and Customer-Services
Systems

328

Part 3

distribution cables that extend the loop eventually to a point near each
individual customer's premises. The cables may continue in conduit to
the distribution points, or they may be carried via telephone poles or
buried directly (without conduit) in the ground. A drop wire or buried
service wire extends the loop to the living unit or office building and terminates at the station protector. Inside wiring completes the loop connection to the customer's station set.
The loop area is generally characterized by changing population density and extensive rearrangements. The loop plant must be readily accessible to accommodate these changes. The major access points in the loop
area are the serving area interface in the outside plant, which serves as a
rearrangeable cross-connect point between the feeder and distribution
cables, and the main distributing frame (see Section 12.3) in the central
office building, where the loops are then cross-connected to the local
switching system or to a dedicated interoffice circuit.
Loops must be capable of transmitting 2-way voice-frequency speech
or data signals. In addition, they must handle signals from rotary and
TOUCH-TONE dialing and supervisory signals peculiar to the switching
machine in the serving central office. The equipment used on loop facilities must be able to transmit and to withstand the voltage used to ring
the bell on the station set. The loop must also provide ample direct
current to permit proper operation of the station set. Some of these
design considerations are discussed later in this chapter.
As indicated by Figure 9-1, loops are generally quite short. The
median length is about 1.7 miles, or 9 kilofeet (kft), and 95 percent of all

100

L

90
80

en

a.

g~

....IZ

70

11.<

60

Wa:

50

o~

"w
<~

~a:

Zo
W:J:
~en

W

I

1

/

I

30

~

a.
20

0

,L

I(

40

10

".".-- r-

/
o

J

5

10

15

20

25

30

35

40

45

WORKING LOOP LENGTH (kft), L

Figure 9-1. Distribution of loop length (1973).

50

Chap. 9

Transmission Systems

329

loops are shorter than 5.2 miles (27.5 kft). As the distance from the central office increases, so does transmission loss. The primary design measure used to reduce the loss is the limitation of maximum loop resistance
by the application of larger gauge wire pairs. Figure 9-2 shows the distribution of different gauge cables as a function of loop length. For loops
longer than 18 kft (about 15 percent of all loops), loading and carrier
techniques are used, as described in Section 9.2.1.
100

........

90 ~

\

Z

~~

80

0<
e.::E:
oa:
ow
wI~a:
::)0
<::E:

70

Ci)z

::&1-

60
50

~CI)

CI)

we.
<0
1-...1
Za:
wo
~I&.
w
e.

~
~

\ / r\

~~ 22'

--

GAUGE \

,

\ \

40

~o

1

30

GAUGE "\
I

o

10

19

GAUGE

\

~\

20

GAUGE

1\

\

10 .... 26\

o

,

\

~

II

20

24
-

30

40

" -.........

50

60

70

80

90

100

DISTANCE FROM CENTRAL OFFICE (kft), L

Figure 9-2. Average percentage gauge composition
of loops versus distance from central office (1973).

Because of the need for flexibility to accommodate future assignments,
a long-time telephone company practice has been to provide for multiple
appearances of the same loop pair at several distribution points. These
bridged taps make reassignments of loops easier but influence the
transmission characteristics, so that rules have to be established to limit
their length to 6 kft or less.
The Metropolitan Interoffice Plant
A metropolitan, or metro, area is served by a number of closely spaced
central offices interconnected by interoffice transmission facilities. Two
general types of circuits are provided by these interoffice facilities:
trunks, which are shared by users of the public switched telephone network (PSTN), and special-services circuits, which are dedicated to those
customers who have a long-term need to reach specified distant locations.

330

Network and Customer-Services
Systems

Part 3

Special-services circuits utilize nearly half of the interoffice facilities.
Both trunks and special-services circuits may be bundled together on the
same cable or carrier facility as it leaves the central office (see Figure 3-2).
These facilities provide paths to nearby offices, where the circuits terminate or pass through to their ultimate destination. A sequence of spans
(subpaths), which carries a circuit from its origin to its destination, is
designated a route. The average metro span is about 6 miles long. Route
cross sections become thinner as the distance between offices increases,
but economies of scale are achieved by concentrating circuits into spans
and thus eliminating the number of geographic paths between pairs of
offices. (Table 4-6 presents data on route cross sections and the distance
between switching offices.)
Of some ten million interoffice circuits in the Bell System, about twothirds are in the metro area. The very short metro area circuits use
voice-frequency (VF) transmission, while the rest use digital carrier facilities. Transmission loss compensation is often required for the VF facilities in the form of inductive loading on many of the shorter VF circuits,
and VF repeaters and loading on the longer ones. For digital transmission systems, the metro area requires low per-circuit terminal costs
because the distances are so short (and line-haul costs correspondingly so
low) that terminal costs dominate. The economic break-even length
between vOice-frequency transmission on paired cable and digital carrier
has been dropping sharply in recent years, and in fact, for trunks connecting to 4ESS switching equipment, the break-even distance is 0 miles.

The Outstate Interoffice Plant
The outstate, or rural, area covers small cities and towns and remote
suburban as well as sparsely settled areas. The networks are usually simple and tree-like with low connectivity between central offices and few
alternate paths available for route diversity.
In the outstate area, trunk connections between class 5 and class 4
offices predominate. Since the central offices are widely spaced in the
outstate area, transmission systems tend to be longer than in the metro
area. Figure 9-3 plots the distribution of outstate area system lengths.
Interoffice cross sections (number of circuits) are also small by metropolitan standards: N -carrier routes average about fifty voice circuits.
Annual circuit additions also tend to be small, which has a distinct
impact on the type of facilities provided in outstate areas.
Currently, outs tate networks consist predominantly of N-carrier and a
small but growing number of TI carrier systems (see Section 9.4.2) provided with protection switching to improve reliability. This protection is
critical in an outstate area because alternate routes may not be provided,

100 r - - -........""T"""---,..---.,.---~=--'--_

80

t------#--~IIf__----+'~--+---_+_--_f

60

t--f---.-_+_---+----"~-+----_+_--__f

40

~~-_+_---+---+----_+_--__f

20

~~-_+_---+---+----_+_--__f

o

~

o

__

~

__

40

~

___

80

~

__

120

~

__

160

~

200

SYSTEM LENGTH (MILES)

Figure 9-3. Distribution of system length
in the outstate area.

many small offices are unattended, and communities could easily become
completely isolated.
The Long-Haul Interoffice Plant
The long-haul network overlays the metro and outstate networks and
interconnects their toll offices. For this reason, the long-haul network is
often referred to as the intertoll network. Traffic is highly concentrated
and brought to a limited number of large switching offices. High-density
intertoll trunks tie together class 4, 3, 2, and 1 offices throughout the
United States. Cross sections in all but the extremities of the long-haul
network tend to be large.
The discussion and accompanying facility map (Figure 4-11) in Section 4.4.2 indicate that multiple routes exist between metro areas. The
heavy cross sections in the network generally have facility diversity and
provide high reliability as well. (For example, parallel microwave radio
and coaxial cable systems frequently are provided along the same route.)
Intertoll trunks are typically from 100 to 1000 miles long, with an
average length of about 400 miles. Only 3 percent of the intertoll trunks
are longer than 2000 miles, but because of their extreme length, they
account for 25 percent of the total intertoll mileage.

331

Network and Customer-Services
Systems

332

Part 3

Transmission loss objectives for the long-haul area are stringent. For
example, an intertoll trunk of average length (400 miles) must meet a via
net loss (see Section 6.6.1) objective of about 1 decibel (dB). Thus, costly
high-quality transmission systems are generally required. On the other
hand, the economics require low per-circuit-mile costs. This is a motivating force for wider system bandwidths so that large numbers of circuits
may be combined on the system. This has resulted in a succession of systems that provide increasing amounts of capacity at decreasing percircuit-mile costs. This is conceptually illustrated in Figure 9-4, where
segments A through F represent cost versus circuit capacity for systems
suitable for different network applications in terms of circuit capacity.
Only 21 percent of all circuits in the Bell System are long haul, but
because they are so much longer, long-haul circuits represent the major
percentage of the circuit mileage in the total trunk plant. At the end of
1980, the total number of circuit miles in the Bell System amounted to
1.03 billion. Transmission systems in the long-haul area account for over
75 percent of this total.

III
.J

i
....
S
()
a::

(j

a::

III
D.

....
en
0

()

2

III

....

en
>
en
III

>

~

c(

.J
III

a::

10

I

I

I

100

1000

10,000

100,000

SYSTEM CAPACITY (CIRCUITS)

Figure 9-4. Economy of scale in the long-haul area.

9.1.2 INVESTMENT BY AREA

In 1980, the Bell System's total investment in transmission facilities
amounted to nearly $60 billion. Almost 60 percent of this investment
was in the loop plant. The remaining 40 percent was invested in the
interoffice plant. The breakdown was approximately: metro area, 26 percent; outstate area, 5 percent; and long-haul area, 9 percent.

Transmission Systems

Chap. 9

333

9.1.3 GENERAL TRANSMISSION SYSTEM TYPES
This section describes the basic characteristics of VF and carrier transmission systems. Details on specific systems are provided later in this
chapter.

Voice-Frequency Transmission Systems
The medium most often used for VF transmission is a twisted wire pair in
a multipair cable. The physical design of the cable (wire gauge, type of
insulation, twist lengths) determines the transmission properties of the
wire line. These properties include the attenuation, phase shift, and
characteristic impedance of the transmission line. Relating these to the
electrical properties of the pairs (resistance, inductance, conductance, and
capacitance) is beyond the scope of this book. (Section 6.3 presents a
general description of media.)
As Figure 9-5 indicates, VF transmission is limited to very short distances because of the high attenuation per mile of untreated cable pairs.
To meet a 3-dB loss objective, for example, applications are limited to less
than 3 miles even with 19-9auge cable.
To reduce the VF attenuation in cable pairs, lumped inductances
called load coils are placed periodically along the cable. The most common loading system used in the VF plant is designated H88. 3 Figure 9-6
10

9
8

iii'

~

7

z

0

i=

6

w

5

«
~
z

l-

I-

«
N

4

0

3

:r
6

~

2

CIRCUIT LENGTH (MILES)

Figure 9-5. 1000-Hz attenuation versus length
for non loaded cables.

3 This corresponds to load coils of 88 millihenries, each spaced 6000 feet apart.

10
9
8

iii'
~

7

z

0
j::

6

e(
;:)

z

w

lI-

5

..

4

0
0

3

e(

;:
~
2

0
0

2

3

4

5

6

7

8

9

10

CIRCUIT LENGTH (MILES)

Figure 9-6. 100Q-Hz attenuation versus length
for H88-loaded cables.

shows the attenuation at 1000 hertz (Hz) for H88-10aded cables. Comparison with Figure 9-5 shows that loaded circuits one and one-half to
three times the length (depending on wire gauge) of non loaded circuits
will have the same attenuation at 1000 Hz. Loading, however, produces
attenuation that depends on frequency. As Figure 9-7 shows, a loaded
cable behaves like a low-pass filter with a cutoff frequency of about
3000 Hz.
Loading introduces a number of other disadvantages. For example,
compared to nonloaded pairs, it reduces the velocity of propagation by as
much as a factor of 3. As a result, the absolute transmission delay4 is
increased. In addition, loading introduces considerable delay distortion
and some attenuation distortion in band that requires equalization on
longer VF lines.
To obtain additional range on non loaded and loaded wire pairs, VF
repeaters are necessary. In 2-wire operation, maximum repeater gains are
limited to about 12 dB because cable characteristics change with temperature, terminations vary due to switching, and crosstalk becomes a problem. The maximum number of tandem repeaters is usually limited to two
to avoid singing or other oscillations. For 22-gauge H88-10aded cables,
this limits the range to under 35 miles, assuming a 3-dB loss requirement.

4 Section 6.6 discusses transmission delay and other impairments mentioned throughout
this chapter.

334

65
60
55
50

iii
~

IJ)
IJ)

0

45
40

...J

z

35

~

a::

30

IJ)

25

0

w

~

20
15
10
5
0

2

3

4

5

6

FREQUENCY (kHz)

Figure 9-7. Effect of loading on 54-kft of
24-gauge cable.

To obtain additional range, 4-wire VF transmission may be employed.
Four-wire transmission is inherently more stable than 2-wire, since there
is only one singing path extending around the circuit from one end to
the other. Repeater gains are then limited by nonlinear distortion, echo,
and crosstalk considerations. In practical applications, 4-wire VF systems
are limited to the length at which their cost per circuit is less than that of
a competing carrier system.

Carrier Transmission Systems
Carrier systems, both analog and digital, are generally more economical
than VF transmission where distances are longer and where cross sections
are large. The lower line-haul cost of carrier transmission is the result of
the increased utilization of the transmission media achieved by combining (multiplexing) a large number of message signals into a composite
signal. This saving is at the expense of an increase in the cost of the
required multiplex terminal equipment.
In addition to the economic advantage, carrier transmission also offers
a number of performance advantages. The major one is a dramatic
improvement in the velocity of propagation at carrier frequencies. This
results in a very low absolute delay for carrier signals and a resultant
advantage with respect to echo control. Combined with better control of

335

Network and Customer-Services
Systems

336

Part 3

impedances, this permits long-distance circuits to operate at lower losses
while maintaining excellent stability.
Analog Carrier Systems. Analog carrier systems use repeaters that are
designed to compensate (equalize) for the loss characteristics of preceding
cable sections and reproduce at their outputs a linearly scaled version of
the cable section input signal. In practice, analog repeaters are not perfectly linear, nor are they noise free. Therefore, noise and nonlinear distortion accumulate with distance and determine to a large degree the performance achievable.
An analog carrier system is not limited in terms of the types of signals
that may be transmitted. Speech, data, video, and supervisory signals
may be combined, provided only that interference requirements are compatible with system performance and bandwidth. Speech signals are particularly suitable to analog carrier systems as compared to digital carrier
systems because system bandwidth is used more efficiently (see Section 6.4.3). However, techniques for placing a mix of different signals
(data, speech, and video) on an analog carrier system usually result in a
lower capacity (fewer channels) as compared to a digital system.
Over long distances, where line-haul costs are a greater factor than
terminal costs, analog transmission has traditionally proven more
economical than digital transmission, as illustrated in Figure 9-8. This
situation is changing, however (see the next section).
One im portan t factor in the economic comparison between analog and
digital carrier transmission is whether the switching system associated
with the facility is analog or digital. When an analog facility terminates
on a digital switch, additional terminal equipment is required to convert

• VF SYSTEMS COST
-

!!

S

. ~

~
~

t;
o

•••••••••••
~

: _---?-4~----._--- ---.~
~

_ -

DIGITAL CARRIER
."......... SYSTEMS COST

- -

-

ANALOG CARRIER
SYSTEMS COST

~

..

.~

(.)

~
~

.". . . . . ~

~

",..

~

~

........./

VF

DIGITAL CARRIER SYSTEMS
.....
.......

3 TO 5

ECONOMICAL

ANALOG CARRIER SYSTEMS

... .......
.... ~

ECONOMICAL

300 TO 500
DISTANCE (MILES)

Figure 9-8. Relative economics of transmission systems.

Chap. 9

Transmission Systems

337

the signal to the proper format. This increases the distance at which analog systems "prove-in" economically to a distance above that indicated in
Figure 9-8.
Digital Carrier Systems. Digital carrier systems are characterized by
discrete signals, regenerative repeaters, and time-division multiplex
signals.
Converting signals, whether they be speech, data, or video, into digital form permits them to be intermixed and treated identically within a
common transmission facility. Pulse-code modulation, described in Section 6.4.3, is used to convert speech signals into digital form. The major
digital impairment is quantizing noise, which occurs only in the terminals and can be controlled by assigning a sufficient number of binary
digits to represent the encoded samples.
The digital signal format is naturally well-suited for data signals. This
"match" between the data signal and digital carrier transmission is very
efficient. For example, T1-carrier can carry 24 times 64 kilobits per
second (kbps), or 1536 kbps of information; while twenty-four analog
voice channels can carry 24 times 9.6 kbps, or 230 kbps of information.
Digital carrier systems regenerate the signal at each repeater location.
Line noise and interference in the media have very little effect on the
signal since they do not accumulate. As a result, the required signal-tonoise ratio is lower for digital systems than for analog systems. Relatively noisy media, such as multipair cables where crosstalk is a limiting
factor, are better suited to digital carrier than analog carrier. On the
other hand, in high-quality media, such as coaxial cable, the ruggedness
property of the digital signal is of no real advantage. In congested radio
media, however, the digital signal is able to tolerate external interference
to a much greater degree than analog transmission.
Digital multiplex terminal equipment is able to exploit the latest
advances in low-cost, high-speed integrated circuit technology. Digital
devices continue to become faster, smaller, longer lived, more reliable,
and less power consuming.
As terminal costs decrease, the prove-in range for digital carrier systems will increase over that shown in Figure 9-8. The increase will result
from both a shorter prove-in distance with respect to VF systems and a
longer prove-in distance with respect to analog carrier systems on paired
cable.

9.2 VOICE-FREQUENCY TRANSMISSION
9.2.1 LOOP AREA APPLICATIONS
Several VF system design methods have been used in the loop plant.
They have evolved as a result of continuing efforts to reduce the cost of
the loop cable plant while maintaining satisfactory transmission and

Network and Customer-Services
Systems

338

Part 3

signaling performance on an overall statistical basis. Table 9-1 summarizes these design methods, and they are described below.
The resistance design method controls transmission loss by limiting
loop cable resistance and requiring loading on loops 18 kft or longer.
The rules include limitations on the maximum allowable lengths of
bridged taps and tolerances on load-coil spacing. By employing heavier
gauge cable pairs where needed, loops designed with the resistance
design procedure meet the needs of message telephone service generally
without the need of electronic equipment for gain (amplification), equalization, or network control signaling.
Unigauge design has as its objective the use of finer 26-gauge cable
rather than coarse gauge, thereby reducing the amount of copper
required. This design uses 26-gauge pairs exclusively to a range of 30 kft,
beyond which heavier gauge may be added to the loop. Loops longer
than 24 kft are loaded, and range extenders, when needed for message
traffic, are shared rather than permanently connected to each loop. These
devices, which extend signaling as well as transmission range, are
switched in by the local switching system. Longer loops for specialservices circuits must, of course, be treated on a dedicated basis.
Long-route design is intended to serve the small fraction of telephone
customers in rural areas who are located beyond the range covered by
resistance or unigauge design. A specific combination of range extension
and fixed gain is prescribed so that loop loss is limited. Loop lengths of
as much as 210 kft (about 40 miles) can be achieved using 19-9auge, H88loaded cable.
The concentrated range extension with gain (CREG) design method
offers a uniform and flexible approach that is compatible with resistance
design and long-route design methods. This method enables the
increased utilization of finer gauge cables in the loop plant through the
use of switched range extension shared by several customers. The application of CREG design rules results in lower loop losses than with other

TABLE 9-1
COMPARISON OF VF LOOP DESIGN METHODS

Resistance
range (ohms)
H88 loading (kft)
Gauging

Resistance

Unigauge

Long-Route

CREG

0-1300

0-2500

1300-3600

0-2800

>18

>24

All

>15

Mixed

26 gauge

Mixed
(to 30 kft)

Mixed

Chap. 9

Transmission Systems

339

designs over a considerable range of conditions. Because of the improved
overall transmission performance of CREG loops as compared with unigauge loops, in addition to administrative and design advantages, CREG
design has superseded unigauge design.
While these methods describe the design of the existing loop plant,
new rules for loop design became effective in 1983. In almost all cases,
digital loop carrier and revised resistance design will be used for future
relief and extensions of the loop plant. The principal application characteristics for these approaches are
• Revised resistance design will be used for loops that are 1500 ohms or
less and 24 kft or less.
Loops 18 kft or less s must be designed to 1300 ohms maximum and
must be nonloaded.
Loops greater than 18 kft but less than or equal to 24 kft,S may be
designed to 1500 ohms maximum and require H88 loading.
• Digital loop carrier will be used for loops longer than 24 kft.

9.2.2 METROPOLITAN INTEROFFICE APPLICATIONS
In the metro area, VF trunks and special-services circuits use multipair
cables similar to those used for loop applications. With the average
length of a metro circuit about 6.5 miles, the loss incurred requires
between 2 and 3 dB of gain.
One approach to providing gain and cable loss equalization in 2-wire
VF circuits is by use of the negative impedance E6 repeater. Some limitations of the E6 repeater, however, are:
• The maximum repeater gain must be reduced to provide adequate
margin against singing, crosstalk, and overload.
In many applications, additional signaling equipment is required to
compensate for a reduction in dc signaling range and dial-pulse-delay
distortion introduced by line build-out (LBO) networks,6 transformer
windings, and the capacitance of the amplifier.
• A fixed attenuation-versus-frequency characteristic, which cannot be
adjusted to match specific cable layouts, limits the maximum number
of repeaters on a circuit to one or two.

S Total length including bridged tap.
6 Amplifiers (repeaters) in a cable transmission system may be designed to compensate for
distortion of a specific length of cable. When the length of cable between amplifiers is
less than that for which the amplifier is designed, one or more line build-out networks are
used to bring the distortion to approximately the design level.

340

Network and Lustomer-:;ervlces
Systems

Part 3

The more modern VF equipment concept is one of consolidation,
using the metallic facility terminal (MFT). In the MFT, required
transmission and signaling functions are provided by separate transmission and signaling plug-in units, which mount side by side in a factorywired bay. When wired to the main distributing frame, the appropriate
plug-in units automatically provide all functions needed, using a
minimum number of cross-connections. The transmission unit provides
gain, level adjustment, impedance matching, amplitude equalization, precision balancing, and access to the signaling unit. The signaling unit performs such functions as circuit range extension, dial-pulse correction,
ringing regeneration, supervision regeneration, signal conversion, toll
diversion, and dc control of the customer's station set. MFT installation
is efficient since options are selected by switches on the plug-in units,
and periodic testing is eliminated through the use of reliable, stable
solid-state circuitry. The next section discusses 4-wire designs.

9.2.3 OUTST ATE INTEROFFICE APPLICATIONS
In the outstate area, the median length of a VF circuit is nearly twice that
found in the metro area. The via net loss design, discussed in Section 6.6.1, controls the loss and echo of these longer VF circuits and
requires 4-wire operation. Amplification is provided by separate gain
units for each direction of transmission. (There are some 4-wire designs
in the metro area, and this discussion applies there as well.)
The older equipment design is the V4, consisting of two major types:
the 44V 4, a 4-wire repeater, and the 24V 4, used where a 4-wire circuit terminates at a 2-wire switching office or at a customer's premises. The MFT
design for 4-wire facilities is the modern counterpart to the V4. Figure
9-9 illustrates the 44- and the 24-type MFT repeater configurations.
Two MFT shelf arrangements are provided, each having spaces for
twelve plug-in units. When both signaling and transmission treatments
are needed, one shelf assembly provides for six circuits using six doublemodule signaling and transmission plug-in units. When only transmission treatment is required, another shelf assembly is used, which mounts
twelve single transmission plug-ins.

9.3 ANALOG CARRIER TRANSMISSION
9.3.1 LOOP AREA APPLICATIONS
In the loop area, continued growth in demand, the trend from multiparty
to single-party lines, and the rising cost of copper pairs have provided
impetus for development of pair-gain loop transmission systems that
increase the utilization of loop cables by enabling more than one customer to share each physical wire pair. Analog carrier systems for loop applications fall into two basic categories: single-channel and multichannel.

CENTRAL
OFFICE
TERMINAL
OR
4-WIRE
LINE

r----'
~

[> :

----------~I~

>~I----------

I

I

I

I

----------~!~

I-

o
:

o<
..J

10,000

AVERAGE OVERSEAS
TRAFFIC GROWTH

5000

W
Z
Z

<
o

l:
W

"~

1000

w

500

(/)

I

::::E

I

o
i=
z
<
..J

I-

~
Z
<
II:

100
50

/iTJf: ~'~8
S:

___

SF (845 CHANNELS)
(1968)

CHANNELS,

(1963)

I-

S8 (48 CHANNELS)
(1956)
1956 1960

1965

1970

1975

1980 1983

YEAR OF INSTALLATION

Figure 9-14. Growth in transatlantic traffic and the
succession of undersea coaxial systems.

same. Repeater spacing is determined so as to optimize the received
signal-to-noise ratio. Signal levels are set well above thermal noise and
well below that which would produce excessive nonlinear distortion
products. Differences between overland and undersea systems are a
result of differences in the environment and the increased reliability
required in undersea applications.
Since undersea repairs are very time:-consuming and expensive, these
systems must be extremely reliable. In fact, the reliability objective is 100
times more demanding than that of an overland system, requiring fewer
than four service interruptions caused by component failures during a
20-year period. To achieve this, the number of components in an undersea system is held to an absolute minimum, and each component is specially manufactured and screened. Control of parameters is also undertaken as the cable is laid. For example, with SC, the 2-way repeaters are
actually spliced into the cable (every 5.1 nautical miles) during the cablelaying process, and measurements are made continuously as the cable and
repeaters are laid. Every 100 to 150 nautical miles, an equalizer (adjustable only when the cable is laid) is placed. At intervals of 700 nautical
miles, equalizers that can be controlled from shore locations are installed.
These are needed to compensate for changes in transmission characteristics due to cable aging.

351

352

Network and Customer-Services
Systems

Part 3

Long-Haul Analog Microwave Radio Systems
During the last three decades, long-haul analog microwave radio systems
operating in the 4- and 6-GHz common-carrier bands have provided an
ever-increasing percentage of the total Bell System circuit mileage. By
the end of 1980, TO and TH microwave systems accounted for over 57
percent of the total 1.03 billion circuit-miles; L-carrier coaxial systems
provided only 19 percent of this total.
TO systems operate in the 3.7- to 4.2-GHz band, while TH and AR6A
systems operate in the 5.925- to 6.425-GHz band. These systems have
been designed to satisfy signal-to-noise and reliability objectives for
long-distance communications paths up to 4000 miles long.
4-GHz Long-Haul Microwave Systems. The first TO microwave radio
system was introduced in 1950. In its initial form, TO-2 carried 480
voiceband channels on each of five 2-way, 20-MHz RF channels providing a route capacity of 2400 VF channels. A sixth 2-way RF channel was
used for protection. Over the years, numerous improvements have been
introduced that have greatly increased the route capacity and, in turn,
significantly lowered the cost per circuit-mile. There has also been a
significant increase in system reliability. Some major improvements
include: the introduction of the horn reflector antenna and its associated
circular waveguide system (1955) which was introduced to permit the
addition of a 6-GHz system on the same route and which also permitted
both vertically and horizontally polarized signals to be transmitted
(1959); the introduction of solid-state components starting in 1967, which
improved system reliability; and the increase in the power output of the
microwave transmitter from 0.5 to 5.0 watts (1973). The resulting route
capacity is now 19,800 VF channels. Table 9-5 shows the steps in the
evolution.
A portion of a typical system layout of a TO microwave system is ill ustrated in Figure 9-15. Repeater stations are usually placed between 20
and 30 miles apart, depending upon line-of-sight path clearance, tower,
height, fade margin, interference, and economic considerations. Unlike
analog coaxial systems, in which doubling the repeater spacing doubles
the amount of loss, doubling the distance between radio repeater stations
increases the loss by only 6 dB.
At each intermediate repeater station there are four antennas. One
antenna is used for receiving and another for transmitting in each direction. Received RF channel signals are separated by waveguide networks,
applied to intermediate frequency (IF) heterodyne radio receivers9 and
9 A heterodyne receiver "mixes" a single-frequency signal with the incoming RF channel
signal to produce a difference frequency, in this case, the IF of 60 to 80 MHz.
Amplification and certain signal processing can be done more efficiently at this frequency.
In the transmitter, similar mixing of frequencies produces the RF channel signal to be
transmitted.

Chap. 9

353

Transmission Systems

TABLE 9-5
4-GHZ LONG-HAUL ANALOG MICROWAVE RADIO SYSTEMS
(3.7 TO 4.2 GHZ)
Characteristic

TD Radio'"

Service date

1950

1959

1967

1968

1973

1979

VF channels
per RF channel

480

600

900

1200

1500

1800

Radio channel
assignments

12

24

24

24

24

24

Tota12-way
RF channels

6

12

12

12

12

12

1 by 5

Two
1 by 5

2 by 10

2 by 10

1 by 11

1 by 11

Working
VF channels

2400

6000

9000

12,000

16,500

19,800

Transmit
power (watts)

0.5

0.5

1.0

2.0

2.0/5.0

5.0

Protection
switching

"Denotes TD-2 radio, introduced in 1950, TD-3 in 1966, TD-3A in 1970, and TD-3D in 1973.

transmitters, then recombined by waveguide networks, and transmitted to
the next intermediate repeater station. Up to ten radio hops constitute a
protection switching section. Within a switching section, a standby protection radio channel may be automatically substituted for any working
radio channel in the event of a deep-frequency selective fade.
At a main terminal repeater station, frequency modulation (FM)
transmitters modulate baseband signals, consisting of frequency-division
multiplex (FOM) voice channels and data or video signals, to the IF band
extending from 60 to 80 MHz. Frequency modulation is used in TO radio
systems because FM signals are insensitive to the gain nonlinearity
(compression) introduced by the typical IF and microwave amplifiers. FM
receivers demodulate the 70-MHz IF signals to recover the baseband
signals.
6-GHz Long-Haul Microwave Systems. The need for additional route
capacity and the savings possible if 6- and II-GHz routes could be added
to existing 4-GHz (TO-2) routes led to the introduction of the horn

INDIVIDUAL
TELEPHONE
CHANNELS

..

"
,.H.,

i i

==1. u J
II

roiL,
FROM
ADDITIONAL
TERMINALS
AND
TRANSMITTERS

--~

I

--l,r J

"

,.JL,

==~

I

",r J

__ I
IF
8CHIO MHz

"

JL ,

- - ,L__ JI

II

II
II

r'" L,
I

L i

::~

'---

" ::==

rJ L,
I

" ::==

r"L,
I

",r"
II

FROM
ADDITIONAL
TRANSMITTERS

TO
ADDITIONAL
RECEIVERS

rJ L,

;:==

IL __ 01

I

J
L"r
II

L,r~

L, r'

"

ron,

,..--

IF
6O-BO MHz

,JI,

:-:::

I

"Hoi

"

1

pH.,
::~
I

.."

...

"

roiL.,
=:~
I
L __ "

BASEBAND
OUTPUTS

TELEVISION

O~~~~~:G +--

MAIN TERMINAL REPEATER STATION
FM
IF
RCVR
TERM
TRMTR

~--------------------------------~I
INTERMEDIATE REPEATER STATION

FREQUENCY MODULATION
INTERMEDIATE FREQUENCY
RECEIVER
TERMINAL
TRANSMITTER

Figure 9-15. Typical TO microwave radio system showing
main and intermediate repeater stations.

Transmission Systems

Chap. 9

355

reflector antenna and its associated circular waveguide in 1955. The 6GHz TH-l system was first placed in service in 1961 as such an overbuild.
When TH-l was first "overbuilt" on a TD-2 route, six additional working
2-way RF channels were provided, each having a VF capacity of 1800 VF
channels. As a result, the route capacity was increased from 6000 VF
channels using TD-2 alone to 16,800 VF channels using both systems.
Table 9-6 shows the characteristics of TH-l and TH-3 systems as well as
the features of the newest 6-GHz microwave radio system, the AR6A (discussed in later paragraphs). All three radio systems use the 500-MHz
common-carrier band extending from 5.925 to 6.425 GHz and provide sixteen radio channel assignments. TH-l and TH-3 are both FM systems
like TD-2 and TD-3. TH-3 is capable of a signal-to-noise ratio 4 dB better
than that of TH-l.
A combined TH-3/TD-2D route has a working capacity of 32,400 VF
channels. Even so, some of these routes may become filled by the early
1980s if the present growth in long-distance calling continues. Since

TABLE 9-6
6-GHZ LONG-HAUL ANALOG MICROWAVE RADIO SYSTEMS
(5.925 TO 6.425 GHZ)
TH-l

TH-3

AR6A

Service date

1961

1970

1981

VF channels per
RF channel

1800

1800*

6000

Radio channel
assignments

16

16

16

Total2-way
RF channels

8

8

8

Protection
switching

2 by 6

1 by 7

1 by 7

Working
VF channels

10,800

12,600

42,000

Baseband
modulation

FM

FM

SSBAMt

Characteristic

*See footnote 10.
tSingle-sideband amplitude modulation.

356

Network and Customer-Services
Systems

Part 3

building new long-haul routes in parallel with existing ones is very
expensive and often difficult because of RF interference, a new, larger
capacity, "overbuild" microwave radio system has been developed. Introduced in 1981, the AR6A system puts 6000 VF channels on a single RF
channel, compared to 1800 in TH-3.1O As a result, the combined capacity
of an AR6A/TD route is 61,800 VF channels. Because AR6A operates at
6 GHz, the system can use unoccupied TH radio channels on existing
TH/TD routes. Also, existing TH radio channels may be replaced with
AR6A. The capacity of the 6-GHz band is thus increased from 12,600 to
42,000 VF channels.
A 4000-mile microwave radio system has about 150 radio repeater stations spaced 25 to 30 miles apart. About 100 of these stations will simply
be heterodyne repeaters, consisting of a transmitter and a receiver. The
remaining 50 are called main radio repeater stations, or simply, main stations.
In the case of AR6A, those main stations are equipped with a new onefor-seven microprocessor-controlled protection switching system. Typically, about ten of the main stations are also terminals where baseband
signals enter and leave the system.
The AR6A is the first long-haul microwave radio system to use singlesideband amplitude modulation (SSBAM) rather than frequency modulation. In the past, SSB was not used in the long-haul radio network
because available microwave power amplifier tubes produced unacceptable levels of nonlinear distortion. Two technological improvements
have made SSB feasible: an ultralinear traveling-wave power amplifier
tube (TWT) and a distortion correction circuit. The latter actually adds a
controlled amount of distortion to the signal before it reaches the TWT so
as to cancel out the distortion generated in the TWT and modulator circuits themselves.
Mastergroupll translators, designed for AR6A and L5E, assemble five
600-circuit mastergroups into a tightly packed 3000-circuit multimastergroup signal. Multimastergroup translators combine two of these signals
plus five pilot tones to form a 6000-circuit IF modulating signal. Pilot
signals control receiver gain and dynamic equalization as well as synchronize the frequency of the receiving multimastergroup translators
along the radio route.
Satellite Communications Systems
A satellite communications system is a long-haul microwave radio system
consisting of earth stations and orbiting satellite microwave repeaters.
Between any two earth stations, regardless of route length, there are only
10 A new TH-3 design, available in 1983, provides 2400 VF channels per RF channel or
16,800 total VF channels.
11 Mastergroups and other levels of the analog FDM hierarchy are discussed in
Section 9.3.5.

Transmission Systems

Chap. 9

357

two radio hops as distinguished from terrestrial microwave radio systems
that usually consist of several hops between radio terminals. By the end
of 1982, about 6 percent of the long-haul radio circuit mileage in the Bell
System was carried over domestic satellite systems.
In 1962, the Bell System's Telstar demonstrated the feasibility of satellite communications systems for telephony. The Telstar satellite had a
low elliptical orbit, inclined 45 degrees to the earth's equator, with a rotation period of 158 minutes. Today most satellites are placed in high
(22,300 miles) circular orbits with zero inclination. This gives them a
period equal to that of the earth's rotation and has the advantage that the
satellite is geostationary; that is, it appears to be stationary from a point
on the earth.
Telstar provided a route capacity of only twelve voice channels.
COMSTAR, a modern communications satellite designed and built to
AT&T specifications for joint domestic service by AT&T and General Telephone and Electronics Satellite Corporation (GSAT), provides a route
capacity of 18,000 VF channels. By 1986, the Telstar 3 satellite will provide a capacity of 21,600 VF channels.
Figure 9-16 shows the orbital positions of existing domestic satellites
as of April 1981. Longitudes between 115 degrees west and 135 degrees
west are highly desirable satellite locations since those farther east cannot
serve Alaska, and those farther west cannot serve Puerto Rico.
In addition to domestic systems, there is also an international satellite
system called INTELSAT, under the direction of the International

1360

130

~125

a..

120",

II~

SATCOM-1

127" /
COMSTAR-D4

115

W~~::';'2 1 !
1190
SATCOM-2

ANIKA2IA3

'~ 105.

~
109"

ANIK-8

_,I i t
100

0;;

85

87"COMSTAR-D3

.,·WESTAR-3

• '8 ~950COMSTAR-D1/D2

1040
ANIKAl

~

'\
100SBS-1

990
WESTAR·l

Figure 9-16. Orbit locations of domestic satellites (April 1, 1981).

Network and Customer-Services
Systems

358

Part 3

Telecommunications Satellite Consortium. This system, which began in
1965 with INTELSAT I (Early Bird), currently involves over eighty-eight
earth stations in sixty-four countries and satellites positioned over the
Indian Ocean, the Pacific Ocean, and the Atlantic Ocean. The geostationary satellites view over 40 percent of the earth's surface. The Atlantic
Ocean satellites carryover 60 percent of the global traffic in the system.
Satellite communications differ from terrestrial communications in
several ways. Since there is only one intermediate repeater (the satellite),
there is no accumulation of multiple-hop repeater impairments. Multipath selective fading, which affects terrestrial systems, is not a problem in
satellite systems. However, atmospheric water vapor absorption is a
problem. Each satellite hop is 1000 times longer than a terrestrial hop,
and this results in a very great path loss and long transmission delay.
Satellite systems require larger ground antennas than terrestrial systems
to receive the considerably weaker signals. To minimize system outages
due to component failures, extremely reliable components must be used
in satellite repeaters, together with design redundancy, thorough testing
rou tines, and protection -swi tching arrangements.
The frequency bands used by satellite communications systems are
allocated by the Comite Consultatif International Telegraphique et
Telephonique (CCITT), the Co mite Consultatif International des Radiocommunications (CCIR), and-for those systems operating in the United
States-the FCC. Table 9-7 indicates that three frequency bands are available for satellite systems.
In the Telstar 3 system, six transponders provide coverage of the continental United States (CONUS) only. The down-link outputs of six transponders are switchable from CONUS to Alaska; six are switchable from
CONUS to Hawaii; and six are switchable from CONUS to Puerto Rico.
Separate frequencies for the two directions of transmission reduce
interference. By using the lower frequency band for down-link transmission, the path loss is somewhat reduced. This, in turn, reduces the

TABLE 9-7
FREQUENCY BANDS FOR SATELLITE
COMMUNICATIONS SYSTEMS
Frequency
Bands
(GHz)

Va-Link
GHz)

6/4
14/12
29/19

5.925-6.425
14.0-14.5
27.5-31.0

Down-Link
(GHz)

3.7-4.2
11.7-12.2
17.7-21.2

Bandwidth
(GHz)

500
500
3500

Chap. 9

Transmission Systems

359

transmitted power required from the satellite, where power is limited.
An additional consideration is that up-link 6-GHz signal leakage is less
likely to interfere with 4-GHz terrestrial radio systems located near an
earth station antenna. Most of today's satellite systems use the 6/4-GHz
band pair, which is commonly used for long-haul terrestrial systems. The
higher frequency bands permit more gain with the same size antennas
and do not interfere with terrestrial systems; however, propagation loss
due to rain is increased. The 14/12-GHz band pair is now being
developed, while the 291 19-GHz band pair is more experimental.
The earth station antenna is very large (about 30 meters in diameter),
thereby providing high signal gain and narrow-beam radio propagation.
Although the satellite is in a geostationary orbit, automatic tracking is
provided by means of a telemetry and control system to account for any
small position deviations and to maximize the received signal. This system also performs a number of "station keeping" functions such as keeping the satellite at its assigned longitude and inclination by the use of
small gas jets and adjusting the gain of the satellite receiver to balance
up 1down-link transmission.
The satellite is stabilized to maintain a fixed relation to the earth's axis
and to eliminate tumbling. This permits a moderate-gain satellite
antenna to be used. Solar cells and nickel-cadmium batteries are the primary power sources for the satellite equipment .. The batteries are needed
during solar eclipses.
Because satellite power is limited, FM/FDM modulation is used, since
this provides a favorable tradeoff of power for bandwidth. In addition,
an FM signal is not subject to amplitude nonlinearities. Therefore, the
satellite amplifiers can be operated very close to saturation and thereby
provide maximum power output.
Of the many design considerations that are unique to satellite systems,
only six are mentioned below: (1) sun transit outage, (2) satellite eclipse,
(3) rain effects, (4) transmission delay, (5) noise, and (6) interference
coordination.
The geometry of the sun transit outage phenomenon is illustrated in
Figure 9-17. As shown, the sun appears directly behind the satellite, and
emissions from the sun fall upon the earth station antenna, causing a
large increase in satellite circuit noise. This phenomenon takes place
during the spring and fall equinoxes and persists a few minutes each day
for a period of about six days. The resultant increase in noise is avoided
by temporarily switching to a protection satellite located at a different
longitude, using a separate earth station tracking antenna.
A second orbital phenomenon occurs when the satellite is opposite to
that shown in Figure 9-17. During two 46-day periods each year, the
earth's shadow causes the satellite to experience an eclipse of the sun.
The duration of the daily eclipse varies, lasting up to about 1 hour. During this time, the satellite is deprived of solar energy and must rely 01'1

SATELLITE
SUN

Figure 9·17. Geometry of sun transit outage.

battery power. ,In addition, this exposes the satellite and its circuitry to
large tern perature changes.
Rain causes three major impairments to a microwave satellite signal:
attenuation, thermal noise, and depolarization. Raindrops scatter and
absorb microwave energy, resulting in rain fades that, although small in
the 6/ 4-GHz band pair, increase with frequency and are very serious in
the higher 14/12 and 29/19-GHz band pairs. Rain is also a dissipative
medium at microwave frequencies and radiates thermal noise. The combined effect is a decrease in system signal-to-noise ratio. With earth station site diversity protection, system outages due to rain fades are held to
a minimum. Modern 6/ 4-GHz satellite systems, such as COMST AR and
Telstar 3, increase their channel capacity by using the same frequency
with different polarizations for two channels. These systems are
degraded by rain depolarization of the signal. The nonspherical shape of
the raindrops converts the signal's linear polarization to elliptical polarization and thereby impairs the ability of the antenna to discriminate
between the two channels. This impairment is more dominant than rain
attenuation in the 6/4-GHz band pair, and unlike attenuation, its effect
diminishes with higher frequency.
The 2S0-millisecond, I-way transmission delay from one earth station
up to the satellite and down to another earth station inhibits voice communications slightly, but it is a serious problem for data transmission.
Although data sets are now available with modified protocols to operate
over satellite circuits, many earlier data terminals would experience
difficulties. The round-trip delay of 500 milliseconds on a satellite circuit
would cause very annoying talker echo were it not for the use of echo
cancelers. Prior to the invention of these devices, the practice was to use

360

Chap. 9

Transmission Systems

361

a satellite facility in one direction and a terrestrial facility in the other
direction of transmission.
Satellite transmission is impaired by noise from the earth itself. Thermal radiation from the warm earth extends across the entire beamwidth
(main lobe) of the satellite antenna and is the dominant source of satellite
antenna noise. The earth station is also significantly affected by noise
from the earth. Since the earth station antenna is pointed at least several
degrees above the horizontal, this noise is not in the main lobe of the
earth station antenna but in the back and side lobes.
Interference sources include other communications satellites and terrestrial microwave systems. To control intersatellite interference, satellites currently operating in the 6/4-GHz band pair are placed no closer
together than 4 to 5 degrees. 12 Figure 9-18 illustrates the basic interference paths between a satellite and a 4-GHz terrestrial microwave system.
Since the 4-GHz down-link signal can interfere with a terrestrial system
as well as other satellite systems, the FCC restricts the transmitted satellite signal power to no more than 5 watts. To control 4-GHz terrestrial
interference into earth station antennas, site interference studies are conducted prior to locating an earth station. A large percentage of potential
sites are rejected based on these ground exposure studies. There is also

4-GHzINTERFERENCE

-----L------~L-

__

~

4-GHz TERRESTRIAL RADIO SYSTEM

EARTH
STATION

RADIO
RELAY
STATION

RADIO
RELAY
STATION

EARTH
STATION

Figure 9-18. Interference paths between satellite
communication systems and 4-GHz terrestrial radio systems.

12 Minimum satellite spacings are determined by the FCC and international agreement. In
1983, the FCC proposed a minimum spacing of 2 degrees for satellites using this
frequency band.

362

Network and Customer-Services
Systems

Part 3

possible up-link interference into a 6-GHz terrestrial system. This is controlled by limiting earth station antenna elevation angles to no less than
5 degrees.
9.3.5 ANALOG FREQUENCY-DIVISION MULTIPLEX TERMINALS
Section 6.5 introduced the concept of multiplexing. As discussed there,
analog carrier systems use FDM to combine large numbers of VF (0- to 4kHz) channels for efficient use of wideband systems. Each VF channel
occupies a specific 4-kHz portion of the broadband transmission media.
FDM terminals modulate, filter, and combine voiceband signals to produce a stack of VF channels across the allowable broadband frequency
spectrum.
The FDM plan and the various multiplex terminals in use in the Bell
System today are illustrated in Figure 9-19. A large number of modulation steps are required to produce the twelve levels shown beyond voice
frequency. The basic group, for example, is the first level in the multiplex
hierarchy. Twelve VF channels are combined and occupy the frequency
spectrum from 60 to 108 kHz, forming the output of the channel bank and
serving as the input to the group bank equipment. Six of the remaining
levels in the FDM hierarchy are system levels; that is, the frequency band
at each of these levels corresponds to one or more broadband transmission systems. These are indicated by heavy horizontal lines in Figure
9-19 with the system name(s) at the left.
A total of five basic frequency translations accomplished by SSBAM
are required to place VF channels in higher frequency levels and in
larger groupings throughout the hierarchy. These translations are shown
in the figure and are as follows:
• VF to basic group
• basic group to basic supergroup
• basic supergroup to basic mastergroup
• basic mastergroup to system level or basic jumbogroup or multimastergroup spectra
• basic jumbogroup or multimastergroup spectra to system level.
As new transmission systems have been developed, the FDM hierarchy has evolved, and certain features have become standard for the
family of multiplex terminals required. These include:
• 4-wire transmission throughout
• SSBAM for maximum bandwidth utilization
• VF input and output levels of -16 dB and +7 dB

HIERARCHY
LEVEL

VF
CHANNEL
CAPACITY
13200

L5E
10800
L5
6000
AR6A

MGT-B

4800

4200
MGT-B
BASIC
3600
JUMBOGROUP
(L4)
3000
MGT-B
1800
TH/L3
1200
TO

BASIC
600
MASTERGROUP ______________~~~----&-&-~--~---L--~--~~--&---~--~--~--------INTERMEDIATE
MULTIPLEX LEVELS

SUPERGROUP BANK

SYSTEM LEVELS
BASIC
SUPERGROUP

60
DS1
(24 VF 1.544 Mbps)
LT-1
CONNECTOR

GROUP BANK

BASIC
GROUP

60-108 kHz

12

CHANNEL BANK

DIRECT FORMED
SUPERGROUP

VOICE
FREQUENCY

Figure 9-19. Frequency-division multiplex plan.

• modulating and demodulating circuits combined into 2-way units
called modems 13
• all carrier frequencies precise, stable, and multiples of 4 kHz
13 These units should not be confused with the modems (data sets) used on customers'
premises to convert digital information to analog form for transmission.

363

Network and Customer-Services
Systems

364

Part 3

• pilot-controlled regulation used at the group, supergroup, and mastergroup levels.
A-Type Channel Bank
The first frequency translation performed in the FDM hierarchy places
twelve VF channels into the basic group band from 60 to 108 kHz as
SSBAM signals spaced 4 kHz apart. A block diagram illustrating the
operation of the AS channel is shown in Figure 9-20. Each of twelve
voiceband signals modulates one of twelve carriers spaced 4 kHz apart,
beginning at 64 kHz and extending to 108 kHz. The double-sideband
VOICE CHANNEL
SIDEBAND
ORIENTATION

ElEl

60 64 68

L_

~

BASIC GROUP BAND

1-72

80 84

76

88 92

96 100 104 108

IIIIIIIIIIIII
I~I~I+I~I~I~I~I~I~I~I~I~I
12

11

10

9

8

7

6

5

4

3

108

EL

0

4

CHANNEL
MODEMS

Ek
104

112

B

104

108

CH 1
TO GROUP BANK

104

EL

0

108 kHz

4

Ek
108

100

.R

100

104

CH 2
64

R

0

104 kHz

4

Ek
60

68

R
60

64

CH 12

64 kHz

BPF
HPF

BANDPASS FILTER
HIGH-PASS FILTER

Figure 9-20. A5 channel bank frequencies.

2

FREQUENCIES
(kHz)

CHANNEL
NUMBER

Chap. 9

Transmission Systems

365

signal at the output of the modulator is passed through a high-pass filter
to suppress VF energy and then through a quartz crystal bandpass filter
that passes only the lower sideband. Twelve translated channels are combined to form the composite basic group signal. In the reverse direction,
identical filters select the channels that, in turn, are translated to VF by
twelve demodulators.
Over the past 45 years, there have been six generations of A-type
channel banks performing this function. The A1 through A4 channel
banks used vacuum-tube technology and have been superseded by the AS
and A6 channel banks, which use transistors and hybrid integrated circuitry, respectively.
The AS channel bank was a major improvement over previous
vacuum-tube designs. It was smaller; consumed less power; and had
improved frequency response, gain stability, reliability, and maintenance
access. The A6 channel bank, the current design, uses monolithic quartz
crystal filters and thin-film hybrid integrated circuitry for further reductions in cost and size.
A single-frequency alarm pilot may be transmitted along with the
basic group as an option in either the AS or A6 channel bank to actuate a
carrier group alarm feature. The purpose is to minimize the effect of a
carrier system failure on switching system load by making failed trunks
appear busy. Otherwise, these trunks continue to be seized, the call
attempt fails due to the faulty trunks, and the customer tries again.

LMX Group Bank
The second frequency translation in the FOM hierarchy places five
groups (sixty voiceband channels) into the basic supergroup band. Each of
the group signals modulates one of five carriers, which are spaced 48 kHz
apart between 420 and 612 kHz. After bandpass filters select the lower
sidebands of the translated group signals, they are combined and occupy
the basic supergroup spectrum. The current vintage of group bank is the
LMX-3.
L T-1 Connector

In the past, long-haul analog carrier Signals, in order to be switched by
the 4ESS switching equipment or interconnected to digital transmission
facilities, were required to be demodulated by A-type channel banks to
voice frequency and then converted to a time-division format. This
conversion from FOM to time-division multiplex (TOM) required considerable equipment floor space, power, and intraoffice VF cabling. The LT1 connector, the Bell System's first transmultiplexer, converts signals from
FOM at the basic group level to TOM at the digital signal level 1 (051).
As shown in Figure 9-19, the LT-1 converter converts two analog group

366

Network and Customer-Services
Systems

Part 3

signals into a single, digital, 1.544-megabit-per-second (Mbps) OSI signal.
The OSI signal, in turn, goes to a digital interface frame on the 4ESS
switch or to a digital system cross-connect OSX-l frame. 14 A standard LT1 frame accommodates 480 VF channels via twenty LT-l connectors.
LMX Supergroup Bank
The third frequency translation in the FOM hierarchy following the formation of the 60-channel basic supergroup is performed by the LMX
supergroup bank. It places ten supergroups (600 channels) into the basic
mastergroup frequency band from 564 to 3084 kHz. The frequency format
of this band, which is designated U600 (universal 600 channel), is the
standard format used with most microwave radio and coaxial cable systems. It consists of six supergroups, a 56-kHz guard band, and four more
supergroups. This wide guard band is used to place a line pilot for carrier system regulation. Between all other supergroups in the basic
mastergroup band there are 8-kHz guard bands.
A standard basic mastergroup format common to several long-haul
carrier systems greatly facilitates emergency broadband service restoration. In addition, the U600 format has the advantage that digital signals,
such as lA-ROS, with frequency components less than 500 kHz may be
carried over TO /TH microwave radio systems.
MMX Mastergroup Multiplex and MGT-B Mastergroup Translator
The fourth frequency translation in the FOM hierarchy following the formation of the 600-channel basic mastergroup places between two and
eight mastergroups (as shown in Figure 9-19) in frequency bands that are
used directly as line-frequency signals for broadband facilities or as input
signals to higher levels in the multiplex structure. The following paragraphs discuss two examples of how this translation is accomplished: the
MMX mastergroup multiplex and the MGT-B mastergroup translator.
Figure 9-21 illustrates the MMX mastergroup multiplex, which was
initially designed to provide the line signal for the L4 coaxial cable system. The MMX terminal places six mastergroups (3600 channels) into the
basic jumbogroup band from 564 to 17,548 kHz. (It should be noted that
MGl is fed straight through the terminal without modulation.)
The mastergroups in the basic jumbogroup spectrum are separated by
guard bands whose widths are proportional to their center frequencies.
Wide guard bands are needed to permit individual mastergroups to be
dropped or reinserted along a route without requiring the entire jumbogroup to be demodulated to basic mastergroup frequencies.

14 Section 8.6.2 describes the DSX-l cross-connect.

~ BASIC JUUBOGROUP BAND ~
VOICE CHANNEL
SIDEBAND
ORIENTATION

BEl

3084

MG1

J~

~~

:

~

FREQUENCIES
(kHz)

I II II II II II I
L_ElGGGGG

MASTERGROUP
NUMBER

""I\'

§~

IQC'O

P,)P,)

2

564

~ ""
,,8

...............

J

II)

564

3

"" ""
~:
q;q;

~~

5

~~

6

3084

----------------~BPFr_------_,-----------------------

TO JUMBOGROUP MULTIPLEX
OR L4 LINE

564

3084

3252

5772

MG2--~I--~'~~------~

I

I

lEtl

564

6336 kHz

3084 1

E

15028

I

17548

MG6 ----~

18112 kHz
BPF

BANDPASS FILTER

Figure 9-21. MMX mastergroup multiplex frequencies.

As shown in Figure 9-19, the MMX terminal is also used to combine
either two or three basic mastergroup signals to provide the line signals
for TD systems and for TH and L3 systems.
The MGT-B mastergroup translator terminal illustrated in Figure 9-22
differs in many ways from the MMX terminal. The MGT-B terminal
translates up to eight mastergroups (4800 channels) into the multimastergroup band from 3252 to 24,588 kHz. This wider bandwidth, combined
with fixed 168-kHz guard bands between all mastergroups, results in
additional VF channel capacity: More efficient use of the frequency spectrum is gained at the expense of mastergroup branching flexibility. The
MGT-B terminal does not provide the mastergroup dropping and adding
capability of the MMX terminal, although branching can be done at the
basic mastergroup frequencies by demodulating the entire mastergroup
signal.

367

r--~~

MUL TlMASTERGROUP MGT-B BAND

~~

~
~

~~

Jig

I

II

II

~

:~
~~
II
1# ~~
Ii ......
&~
;;~
22 :?~

II

II

II

II

II

~
~

FREQUENCIES
(kHz)

IV

I

I ..... II~II~II~II~II ..... II ..... I 1-.1
2

3

4

5

6

7

8

VOICE CHANNEL
SIDEBAND
ORIENTATION

G=J

I

[::;]

r:=Fl

G=l

564

MASTERGROUP
NUMBER

3084

3252

5772

MG1
TO MUL TIMASTERGROUP TRANSLATOR

I

6336 kHz

Fl

564

3084

r::;]
5940

8460

MG2

IFl
564

9024 kHz

3084

r=;:]
22068

24588

MG8

25152 kHz

BPF

BANDPASS FILTER

Figure 9-22. MGT-B mastergroup translator frequencies.

In addition to better bandwidth utilization, the MGT-B requires no
costly protection switching equipment, as the result of improved reliability in design.
JMX Jumbogroup Multiplex and
MMGT -C Multimastergroup Translator
Following the formation of the basic jumbogroup / multimastergroup signal shown in Figure 9-19, the fifth and final frequency translation in the
FDM hierarchy places between two and three of these signals into frequency bands appropriate for the AR6A, LS, and LSE carrier systems. The
JMX jumbogroup multiplex (developed specifically for the LS carrier system) and the MMGT-C multimastergroup translator (developed for the
368

Chap. 9

Transmission Systems

369

expansion of L5 to L5E) are discussed below. The MMGT-R, developed
for use with the AR6A microwave radio system, is similar in many
respects to the MMGT -c.
The JMX terminal places three jumbogroups (10,800 channels) into the
L5 line spectrum between 3124 and 60,556 kHz. A total of four steps of
modulation and demodulation are used in the transmitting and receiving
terminals to ease filter design and minimize interference problems. Protection switching is provided on a one-for-one basis to ensure high reliability in consideration of the large number of circuits carried by each
working jumbogroup.
The MMGT-C multimastergroup translator terminal places three multimastergroups (13,200 channels)15 into the L5E carrier system spectrum.
Instead of two or three steps of modulation, the terminal uses a single
step of modulation to translate MMG2 and MMG3; MMG1 is not modulated at all.
The guard band between multimastergroups in the MMGT-C terminal
is about half that provided between jumbogroups in the JMX terminal.
Even so, the guard band is sufficient to permit dropping and adding
multimastergroups at L5E line frequencies, thus avoiding the need to
demodulate the line signal to multimastergroup frequencies. The combined effect of (1) closer spacing between multimastergroups, (2) a
broader L5E line frequency band, (3) closer spacing between mastergroups, and (4) a broader multimastergroup frequency spectrum permits
twenty-two mastergroups to be loaded onto L5E as compared with eighteen mastergroups on L5.
A multimastergroup pilot is inserted between MG4 and MG5 at
13,920 kHz. It is used in conjunction with automatic multimastergroup
protection switching and also as a continuity signal. Protection is provided on a basis of one spare modulator for every twenty working modulators in the MMGT-C terminal.

9.4 DIGITAL CARRIER TRANSMISSION
The digital transmission network consists of four major types of system
components: terminals, multiplexers, cross-connects, and transmission
facilities (or systems). Each of the system components has been designed
to operate at one or more of the six bit rates, or levels, of the TDM hierarchy used in the Bell System digital network. The TDM hierarchy is
described in Section 9.4.3.
Digital signals are created with the use of digital terminals. These terminals take a continuous-wave analog input and transform it, through
the use of sampling and encoding, into a digital waveform.
15 As shown in Figure 9-19, the MMGT-C combines two 4200-channel mastergroups and
one 4800-channel mastergroup to produce the 13,200-channel L5E output.

370

Network and Customer-Services
Systems

Part 3

Digital multiplexers provide interfaces between the different bit rates in
the digital network. Many of the multiplexers include performance monitoring, failure detection, and alarm and automatic protection-switching
features.
The digital cross-connects (DSXs) are the interconnection points for terminals, multiplexers, and transmission facilities. They are equipment
frames where cabling between the system components is cross-connected
to provide flexibility for restoration, rearrangements, and circuit order
work. Each digital signal rate is handled by its own cross-connect.
Hence, there are six cross-connects, identified as DSX-n, where n is one of
the six TDM levels. The DSX-l is discussed in Section 8.6.2 and illustrated in Figure 8-13. The other digital cross-connects are not discussed
in this section.
Digital signals are transmitted from one location to another by
transmission facilities or systems using a multitude of media (paired cable,
coaxial cable, radio, optical fiber, and satellite) at the various bit rates.
The rest of this section describes various transmission systems in relation
to their application in the Bell System network and includes a discussion
of the TDM hierarchy and several types of multiplex equipment.

9.4.1 LOOP AREA APPLICATIONS
Loop carrier systems bring electronic technology to the traditionally
costly and active loop plant-the pairs of metallic conductors that connect
subscribers to the central office. In suburban areas, the present loop plant
is affected by rapid growth and movement, requiring costly cable installation to serve new customers. In rural areas, these conductors must extend
over many miles to serve relatively few people. By substituting electronics for cable, subscriber loop carrier systems offer an economical way
to serve suburban and rural areas. Because these systems increase the
number of customers served by existing facilities (wire pairs), they are
often called pair-gain systems.
Figure 9-23 is a block diagram of a digital carrier system
representative of those used for loop transmission. Channels 1 through n
on the left represent individual customer lines, which are multiplexed at
a nearby customer terminal. Hybrid circuits separate the customer
transmitted and received signals for processing. Following conditioning
(for example, amplification, band limiting), transmitted signals are
converted to a digital pulse-code-modulated (PCM) format as described in
Section 6.4.3. The time-division multiplexer interleaves the digital
signals from the n sources and transmits the combined pulse stream over
a repeatered line to the central office. At the receiver, a demultiplexer
must be synchronized with the transmitting multiplexer so that the
received pulses may be detected and routed to the appropriate channel.

REPEATER

•
•
•

CHANNEL
1

~~

CHANNEL
1

•

MULTIPLEXER

•

•
CUSTOMERS

•

•

LOCAL
SWITCHING
SYSTEM

•
•
•

~~

•

CHANNEL

n

CHANNEL

n
DEMULTIPLEXER

L

, --

-

-

-

-

-

-

-

CUSTOMER TERMINAL - -

-

-

-

-

-

-

+

-

DIGITALLY
REPEATERED LINE

._+ __

I 1

CENTRAL OFFICE
TERMINAL
(IDENTICAL TO CUSTOMER
OFFICE TERMINAL)

Figure 9-23. Digital loop carrier system.

372

Network and Customer-Services
Systems

Part 3

The synchronization circuits are shared by all the channels in a digital
system.
In contrast to analog loop systems, digital systems usually use two 2wire pairs, each pair being dedicated to a particular direction of transmission. The signal transmitted over each pair occupies half the bandwidth
of the signal that would have been necessary if a single pair had been
employed for both directions of transmission. Generally, the choice of
two pairs rather than one pair for transmission in digital systems results
in considerable economic savings.
The SLC 16_40 system is a digital carrier providing forty full-time
voiceband channels between a central office terminal and a single remote
(customer) terminal. Using Tl-type digital repeater line facilities, the
remote terminal can be located up to about 20 miles from the central
office terminal on 22-gauge buried cable. The remote terminal can connect customers on cable pairs up to 900 ohms or 5-dB loss beyond the
remote terminal. This results in a customer-to-remote-terminal distance
of 26 kft on 22-gauge loaded cable and almost 50 kft on 19-9auge cable.
Radio systems that handle Tl-type transmission can also be used between
the remote and central office terminal. The modulation and signaling are
performed in per-line channel units that can be added as customer
demand develops, thus minimizing common equipment. This results in a
low start-up cost and makes the system economical in long-route, lowgrowth areas as well as in shorter loop or higher growth areas where
cable and structure relief are expensive. The system provides maintenance and alarm status information to the central office terminal and has
a simple, straightforward maintenance plan.
The latest digital subscriber carrier system is the SLC-96 carrier system.
It permits ninety-six customers to be served by as few as three digital
transmission lines. The SLC-96 system is more economical, accommodates
more customers, provides more services, is easier to maintain, and offers
greater design flexibility than any of its predecessors. In addition, it has
been designed to be compatible with the interoffice digital network.
Studies of the application of the SLC-96 system indicate that about
one-third of the existing routes studied and roughly 10 percent of the
growth expected on suburban routes could be economically provided by
the SLC-96 system.
Along with these economic advantages, the SLC-96 carrier system provides a wide range of readily available benefits for rapidly growing
suburban and rural areas. The specific benefits include:
• a full range of customer services, including single-party, multiparty,
and coin service-both coin first and dial tone first

16 Trademark of Western Electric Co.

Chap. 9

Transmission Systems

373

• most special services such as Digital Data System service, foreign
exchange lines, private branch exchange trunks, and private-line
services
• remotely controlled testing of customer channels and wire pairs for
single-party, multiparty, and coin service
• continuous transmission monitoring
• built-in aids for isolating system troubles
• a variety of remote terminals in different sizes and styles to accommodate conditions in the serving area.
The system consists of three basic components: a central office terminal, a remote terminal located in the area served, and TI digital lines
linking the two terminals. Each TI line, composed of two wire pairs, can
handle twenty-four channels. The TI lines contain digital repeaters
spaced about I mile apart.
The SLC-96 system employs time-division multiplexing and optional
digital concentration to achieve pair gain. With optional digital concentration, a concentrator digitally switches active circuits to available channels. A two-to-one concentrator permits forty-eight customers to share
the 24-channel capacity of a single TI line. Only two TI lines are needed
to serve ninety-six subscribers, with a third line provided for protection.
This is the carrier/concentrator configuration. When the SLC-96 system is
used without concentration (in its carrier-only configuration), ninety-six
subscribers are served by five TI lines. Each of four lines carries twentyfour 2-way conversations, and a fifth line is available for protection in
case a working line experiences transmission difficulties. If a working
line fails, the protection line is automatically switched into service.
9.4.2 INTEROFFICE AREA APPLICATIONS
Metro Area Systems
This section describes the evolving technology that provides transmission
services in local exchange, or metro, areas where interoffice distances are
relatively short.
In short transmission systems, the cost of the terminals is considerably
greater than the transmission line costs; simple baseband transmission on
paired cable is preferred for these applications. For long routes, multiplex techniques prove economical since line costs now become
significant, and multiplex techniques reduce the per-channel cost. The
economic break-even point for baseband transmission on paired cable
versus TI digital carrier is about 5 miles, although this can vary considerably in any specific metro area. When baseband transmission on paired

l .... t:lWUCK ,UlU

374

~U:Slurnt:r-;::)I:~rVlce:s

Part 3

Systems

cable is combined with digital switching, the break-even point drops to 0
miles.
In recent years, almost all metro area trunks 17 have used paired cable
as the transmission medium. IS It is expected that, in the future, where
traffic is sufficient, long trunks will be grouped into larger cross sections
to achieve economies of scale by using higher speed digital carrier on
paired cable, digital carrier on coaxial cable, digital microwave radio,
and digital lightwave systems. The following paragraphs discuss these
systems.
Paired Cable Systems (Tl). T1 digital carrier was the initial Bell System
short-haul digital transmission line; the first commercial application
occurred in 1962. It has had wide application in metropolitan networks
because of favorable cost and operational experience.
The signal transmitted in T1 is a 1.544 Mbps pulse stream that can be
generated by a variety of terminals. Signals are applied directly to cable
pairs in a bipolar format in which positive and negative pulses, always
alternating, represent one state; and absence of pulse represents the other
state. Figure 9-24 shows an example of a bipolar signal. The use of bipolar signals provides four significant advantages:
1)

A bipolar signal spectrum has the significant portion of the signal
power density below the frequency corresponding to the pulsestream rate (that is, the frequency that is the reciprocal of the pulse
period). A polar signal has twice the bandwidth.

2)

The bipolar signal has a null in the power density spectrum at
o Hz (dc) and at integer multiples of the- pulse-stream rate. This
avoids problems of dc (baseline) wander in ac-coupled input, output, and equalization circuits.
o

o

o

0

0

BINARY SIGNAL

WAVEFORM

Figure 9-24. Bipolar signal.

17 Traditionally, these have been called exchange area trunks.
18 Section 9.3.2 discusses analog carrier systems used in the metro area.

o

Chap. 9

Transmission Systems

375

3)

The bipolar signal can be examined at any point along the
transmission path, and any single-bit transmission error can be
detected. If an error occurs in a 1, thereby converting it to a 0,
adjacent l's will be of identical polarity, violating the coding rule.
If an error occurs in a 0, thereby converting it to a 1, there will be
two successive l's of identical polarity, again violating the rule.

4)

A bipolar signal has good transition density for timing recovery.

Property (2) above is useful for remote maintenance margin testing,
and property (3) above is useful for maintenance because working lines
can be monitored automatically for excessive bipolar violation errors.
Error performance is a transmission performance quantity that must be
controlled in digital system design. Line errors produce impulse noise in
VF channels and can introduce serious errors in data transmission. A
high occurrence of errors can cause terminals to lose synchronization,
which is catastrophic (until reframing occurs)19 in time-division multiplex
syste~s, If line errors occur at rates less than, for example, 10-6 errors
per bit transmitted, noise and signal distortion will be determined primarily by the terminal coding and decoding circuitry. The T1 objective is
for 95 percent of all systems to have an error rate better than 10-6 errors
per bit, under worst-case conditions.
Sources of error in T1 systems are similar to sources of noise and
interference in analog multiplex carrier systems. Misequalization,
interference from other services occupying the cable, ·and electronic circuit noise, for example, are sources of error. Examination of the components of a T1 repeater and its essential element, the T1 regenerator,
yields a better understanding of how these sources of error are controlled
when a system is engineered.
The T1 line repeater is a solid-state, plug-in unit, suitable for pole
mounting or manhole placement and usually powered over the transmission pairs from the central office. Nominal spacing between repeaters is
1 mile. The repeater performs three functions: equalization of the input
pulse stream to correct for linear distortion, extraction of the appropriate
timing strobe for the regenerator, and determination of the input pulse
value (that is, 0 or 1) with corresponding regeneration of a correct and
properly shaped output pulse stream.
The optimum equalization does not correct for line distortion over the
complete signal bandwidth but instead achieves a balance between noise
and intersymbol interference. The amount of equalization required is
determined by the type and length of cable used and is also influenced
by temperature and manufacturing variations.
19 A pulse-code modulation frame and location of the framing bit are shown in Figure 6-17.

376

Network and Customer-Services
Systems

Part 3

Tl lines are self-timed, that is, timing information is extracted from
the input waveform. Since a bipolar signal spectrum has a null at the
pulse-stream rate, timing information cannot be directly extracted
without conversion to a unipolar signal.
Tl carrier can be applied to a wide variety of cables. Although all
cables exhibit an attenuation characteristic that increases with frequency,
the loss-versus-frequency shape varies with different cable types and will
change for a given cable with variations in temperature. Two approaches
have been used in Tl: selection of an appropriate equalizer from a series
of fixed equalizers to match a specific regenerator requirement and use of
a fixed amplifier gain characteristic together with a network to "build-out"
the measured cable loss at a given regenerator to a predetermined value.
Allowances are made in either case for temperature effects and variations
between cable pairs.
Tl systems must be designed to cope with impulse noise generated by
switching equipment in central offices. This type of impulse noise
couples into carrier systems in the cables leaving the central office, where
pairs used for VF trunks and carrier pairs occupy the same sheath .. Usually, the repeater spacing in the first section out of an office is shortened
so that the carrier signals are not allowed to be attenuated to too low a
level, thereby achieving greater margin relative to impulse noise.
The crosstalk noise allocation determines the total number of Tl systems allowed in a given cable sheath. Near-end crosstalk is greatest
where both directions of transmission are on pairs in the same sheath and
in adjacent binder groups. In this case, as few as about 10 percent of the
total pairs may be usable for Tl systems. If separate sheaths are used for
each direction of transmission, near-end crosstalk is eliminated and only
far-end crosstalk remains as a possible constraint. In practice, with existing regenerator designs, this does not limit full utilization of available
cables.
Tl systems operate and are designed on a span basis. A span is
defined as the aggregate of all span lines between two central office
buildings. A span line is a regenerative repeatered line between the office
repeater bays in two central office buildings. For longer routes, a number
of span lines may be connected in tandem and together with channel
banks on each end, constitute a 24-channel Tl system. By engineering
and maintaining on a span basis, span lines can be connected
("hardwired") together as required to provide longer trunks without the
need to engineer the system on a custom end-to-end basis. An average
system contains four span lines and is 15 miles long. Where additional
circuit capacity is needed on a route, additional span lines can be added
to augment the span cross section by following a basic span design applicable to all span lines on that route.
Maintenance operations also are organized on a span basis. A span
will have spare span lines provided to "make good" failed lines by patching when a working line fails. Individual repeaters are tested from the

Chap. 9

Transmission Systems

377

central offices by transmitting a digital signal that contains a superimposed audio component. The audio tone is returned to the testing office
over a separate voiceband pair, and the tone's level and variation give
significant information about the health of the individual repeater
section.
Based on experience with T1 performance, there have been changes in
operating parameters and improvements in design and engineering
methods. One example has been new digital repeatered line designs that
achieve an increased pulse-stream rate (and, therefore, a greater number
of channels) by establishing a tighter balance between cable characteristics and regenerator design. These new lines, TIC and TID, use techniques similar to the T1 design and provide for transmission at a 3.125Mbps rate in 22-gauge plastic-insulated. cable with the same regenerator
spacings as Tl. When D channel banks· are combined with a new M1C
multiplexer20 for transmission over TIC or TID, the system will provide
forty-eight voiceband channels. Because the pulse-stream rate is twice
that of T1, the equalization of the regenerator must accommodate a
greater line loss at the higher frequencies required to support the 3.125Mbps rate. Automatic line buildout (see footnote 6) is used to simplify
engineering and installation. The engineering rules and design methods
used provide greater control of crosstalk and interference required by the
higher pulse-stream rate. Fault location principles for TIC and TID are
similar to those for Tl.
Coaxial Cable Systems (T4M). The T4M digital repeatered line is a
metropolitan area system but can be used for applications of up to 500
miles. It has manhole-located regenerators powered over the coaxial
tubes and spaced up to 5700 feet apart. The line is divided into maintenance spans, with a maximum length of 111 miles between terminating
offices where equipment for maintenance and restoration is located.
Arrangements to add and drop circuits can be provided within the
maintenance span without providing span-terminating equipment on
through systems. and without requiring additional protection lines over
and above the common protection line.
T4M operates at a transmission speed of 274.176 Mbps and uses a
polar rather than a bipolar signal on the line. Thus, only a single decision threshold is required instead of the two thresholds used for. the 3level (+, -, 0) bipolar signal. This provides increased margin, although
certain advanced techniques are required to extract timing information
and to compensate for cutting off the transmitted spectrum at low
frequencies.
Design principles have involved consideration of many of the same
factors as in the digital systems mentioned previously. However, T4M is

20 Later sections of this chapter discuss 0 channel banks and multiplex equipment.

378

Network and Customer-Services
Systems

Part 3

different in one important way: Interference between systems operating
in a coaxial cable is not significant because of the high degree of shielding provided by the coaxial structure. As in all digital systems, intersymbol interference is an important factor. Allowance must be made in
the design of the decision, timing, and cable equalization circuits so that
even with normal environmental, production, and component-aging
tolerances, the regenerator will be error-free. Automatic equalization
without the need for installation adjustment is provided by automatic
line buildouts. Four regenerator designs have been developed to cover
the full cable loss range of from 0 to 56 dB. 21
No routine testing of regenerators is contemplated, and a regenerator
is considered to be operating satisfactorily with bit error rates better than
10-7 error per bit. An error rate of 10-6 error per bit will initiate alarms
and a switch to a protection system.
The full system (SOO-mile) error objective of T4M is less than 10-6
error per bit for 99.9 percent of the time per line. The jitter22 accumulation objective is no more than six time slots peak-to-peak
(22 nanoseconds) over 500 miles. This value avoids a requirement for
special "dejitter" circuits over and above those provided in M34 multiplexers (see Section 9.4.3).
Lightwave Systems (FT3). FT3 is the Bell System's first standard digital
lightwave system. It transmits at the 44.7-Mbps rate, one of the Bell
System's standard digital transmission rates.
A key element of the FT3 system is the lightwave regenerator.
Mounted on a single plug-in circuit board, the regenerator contains a
laser transmitter, an avalanche photo diode receiver, and electrical circuitry for timing recovery and pulse regeneration. This circuit pack contains both lightguide and electrical connectors, so both optical and electrical contacts are made when it is plugged into the equipment frame.
The FT3 system contains extensive maintenance features, including
in-service performance monitoring of the error rate on digital transmission line, automatic protection switching to a spare line when the error
rate exceeds one error in 106 bits, and test equipment to permit remote
identification and isolation of faulty components.
The principal application of FT3 is for interoffice trunking in metropolitan areas. Digital transmission is already extensively used for metropolitan trunking; FT3 offers considerably greater duct efficiency and
repeater spacing than metallic (wire pair) systems. The repeater spacing
for FT3 is initially 4 miles-about four times that of metallic systems-

21 At 137 MHz, the highest frequency important for the 274.176-Mbps transmission rate.
22 Section 6.6.2 discusses jitter.

Chap. 9

Transmission Systems

379

and as further reductions are made in fiber loss, increases in repeater
spacing are anticipated.
In addition to standard FT3 applications, Western Electric is customengineering systems based on FT3 to meet special telephone company
needs. One has been in service in Trumbull, Connecticut, since October
1979. This 4-mile system is in the feeder portion (the cables that leave
the central office) of the loop plant and interfaces with the newly introduced SLC-96 system.
In the Connecticut system, the Tl signals coming from the SLC-96 system terminals in the central office are multiplexed by the MX3 and carried on fibers from the office to a remote terminal. MX3 has the capacity
to multiplex twenty-eight Tl lines; thus, in the Connecticut
configuration, a single MX3 can accommodate five SLC-96 system terminals (with five Tl lines each) serving 480 customers. For the initial service, two SLC-96 system terminals, serving 192 customers, have been
installed. But additional capacity can readily be achieved by simply
adding plug-ins to the MX3 multiplexer and more SLC-96 system terminals. Compared with metallic Tl, the FT3 lightwave system employs a
smaller cable, requires no intermediate repeaters, and has- greater growth
capability. The FT3C system, introduced in 1983, operates at 90 Mbps
and provides twice the capacity of FT3 by transmitting two 44.7-Mbps
signals over the same fiber.

Outstate Area Systems
To bring Tl quality and cost advantages to rural and suburban routes, Tl
capabilities were expanded to adapt it to a telephone plant environment
that includes long transmission distances, relatively few circuits, and
unmanned switching exchanges (community dial offices). In September
1975, the first installation of this modified Tl-known as Tl Outstate
(Tl jOS)-was placed into service between a toll office in Lamar,
Colorado, and five of its outlying community dial offices. Three of the
exchanges are independent telephone company offices.
The Tl lOS system meets the needs of the rural environment with
little modification to the basic Tl digital transmission equipment and
without large developments. Improvements were added to Tl in three
broad areas:
• New guidelines or engineering rules were developed to allow the
operating company engineers to assemble long systems from the various TI/0S components.
• Features were provided to enhance reliability and maintenance. These
include line automatic protection switching (APS), remote monitoring
and APS control by means of built-in telemetry mated to the APS, and
a new repeater fault-locating system.

380

Network and Customer-Services
Systems

Part 3

• Modular and economical central office equipment arrangements were
developed to provide the variety of combinations needed for small
installations.
The outstate system shares basic components with Tl-the 03 and 04
digital channel banks that encode twenty-four voiceband circuits into the
1.544-Mbps digital signal and the Tl repeatered transmission line for connecting between offices. The new maintenance subsystems are added to
these components and, together with the set of engineering rules, make
up a total system that gives Tl/0S digital carrier a number of advantages
over analog carrier for many rural routes.
Advantages obtained from using the 03 and 04 stem from the digital
channel bank's low cost, low noise, and ability to handle a wide variety
of switching system interfaces. In addition, because of the digital format,
the quality of the line signal is monitored easily and directly. This makes
maintenance and trouble location easier.
While conventional Tl is limited to about 50 miles, for Tl/OS it was
necessary to extend this range to over 150 miles. The new engineering
rules achieve this goal while ensuring that overall performance objectives
are met. The rules are based on connecting in tandem no more than 200
repeaters. The actual length of the system is a function of the quality of
the individual repeater sections and the permissible repeater spacing.
The new fault-locating system is essentially a separate VF transmission
facility and, as such, requires its own set of engineering rules. In Tl / as,
fault-locating layouts can be complex; thus, the rules include a computer
program for analyzing networks with branches, multiple terminations,
and bridge taps. The program is called T-Carrier Fault-Locating Applications
Program (TFLAP) and is a subprogram of the Universal Cable Circuit
Analysis Program.
APS is central to the maintenance concept in Tl / as. Each Tl line is
monitored at the ends of each protection-switching maintenance span; a
failure triggers an alarm that identifies the troubled section.
When the APS is operating fully automatically (it can also be operated
semiautomatically or manually) and a service line fails, through transmission is restored by transferring the digital signal to a waiting (good) protection line. The transfer is so fast that message traffic is not affected.
The APS switches upon loss of signal, high error rates (10-4 or 10-6 error
per bit), or wide pulses. When the problem is cleared, traffic is automatically returned to the original line, clearing the protection line for use by
another failed system.
Typically, a number of maintenance spans are connected in tandem to
make up a total Tl/0S system. The APS is designed so that failures
within one maintenance span do not affect APS operation in the others.
An important feature of the APS allows faulty repeaters to be located
from one end of a maintenance span. On command, the two directions of
Tl transmission at the far office are connected within the APS, thus

Chap. 9

Transmission Systems

381

"looping-back" the fault-locating signal to the testing office. The APS
controls can be executed remotely.

Intercity Systems
Cable Systems. The T2 system provides transmission at 6.312 Mbps
using paired cable. T2 provides intercity digital transmission for distances up to approximately 500 miles. Four OSl signals, equivalent to
four T1 lines, can be combined.
T2 uses a separate 22-gauge low-capacitance (LOCAP) cable for each
direction to minimize crosstalk, permitting repeater spacings of 2.8 miles
on repeatered sections of buried or underground cables not adjacent to
office or powering stations. Aerial cable sections and sections adjacent to
offices or powering stations require shorter spacings. The T2 system
design is based upon a maximum of 250 repeater sections in tandem.
Most systems are constructed in blocks of twenty-four T2 lines, including
one protection line to which traffic from a working line can be switched
automatically when errors on the former become excessive. In fully
developed systems with 0 channel banks, this represents a cross section
of 2208 voice channels. Using the largest size (104 pair) LOCAP cable
being manufactured, a 2-cable route will provide a maximum cross section of 8832 voice channels. Because of the large cross sections available,
T2 lines serve the intermediate-distance intertoll trunk market, connecting population centers between which traffic can be grouped to take
advantage of economies of scale.
The T2 repeater consists of two separate I-way regenerators powered
over the transmission pairs. They may be mounted in manholes or above
ground. The regenerator for each direction of transmission is mounted in
a separate apparatus case. The T2 repeater performs the same functions
as the T1 repeater: equalization, timing extraction, and regeneration.
Equalization is achieved with a combination of fixed equalization and
automatic line buildout (ALBa). Five codes (designs) of equalizers
feature different amounts of fixed equalization, selected on the basis of
the cable route makeup. The ALBa consists of a variable equalizer controlled by a feedback signal developed from the fully equalized pulse
stream at the ALBa output.
The pulse stream is in a modified bipolar format called bipolar with six
zero substitution (B6ZS). In this format, l's alternate in polarity as in T1,
but a special code word is substituted when six O's occur in a row in the
original signal. This avoids loss of energy in the tuned timing extraction
circuit. In addition, the restriction to no more than five O's enables rapid
detection of a loss of signal. It should be noted that B6ZS is defined as
the format for the 6.312-Mbps signal (OS2 level in the time-division
multiplex hierarchy) and is required from every terminal that will be
connected to the T2 line.

382

Network and Customer-Services
Systems

Part 3

T2 lines are divided into maintenance spans that can contain up to
forty-four repeaters in manholes, offices, or intermediate power points.
As an aid to sectionalization of troubles, the signal received at each end
of the maintenance span is monitored on each T2 line. If errors in the
B6ZS format exceed a threshold, appropriate alarms are activated. The
signal is modified appropriately to eliminate format violations at any
error rate, or in case of total signal interruption, a special all-l's signal is
inserted. This is done to avoid erroneous error detection beyond the
span where the fault occurred. It should be noted that violation-removal
circuits do not correct errors; rather, they prevent the false indication of
errors in succeeding spans.
Fault location is accomplished by a method similar to that used in Tl,
that is, with a special test signal containing a strong audio component.
T2 lines are being used for both intertoll trunks and private-line service between cities. The system error requirement for T2 is that 95 percent of all systems have an error rate of 10-7 or better under worst-case
conditions. It is evident that T2 has been designed to more stringent
requirements than Tl, since the maximum length of T2 lines is about 500
miles, an order of magnitude greater than the usual maximum for Tl, and
the error objective is an order of magnitude better. With regard to jitter,
it has been shown that 250 T2 regenerators (the maximum) in tandem
have a root-mean-square (rms) jitter accumulation much better than the
system objective of 10 nanoseconds.
Radio Systems (lA-RDS, 3A-RDS, DR6-30, DR11-40). The lA-Radio
Digital System (IA-RDS) is designed to transmit data signals over existing
microwave facilities, including TO and TH. The primary application of
lA-RDS provides intercity connections for the Digital Data System. The
lA-RDS has 4000-mile, 2-way capability. As many long-haul systems as
are necessary may be linked together to form the lA-RDS channel. lARDS provides radio facilities with the capability to transmit a DSI signal
in the radio baseband from 0 to 500 kHz, below the lowest frequency
(564 kHz) used for transmission in the frequency-division multiplex basic
mastergroup (described in Section 9.3.5). Since the basic mastergroup carries 600 voiceband channels, the lA-RDS channel is referred to as data
under voice (DUV).
The 3A-RDS is a high-capacity digital carrier system that operates at
44.736 Mbps. It provides for a total capacity of 560 DSI signals
(1.544 Mbps), or 13,440 voice channels,23 in the ll-GHz radio band.
The 3A-RDS was intended primarily for applications on high-density
intercity routes up to approximately 250 miles, as feeder for T4M or other
systems, and to provide route diversity within the network. Repeater

23 Voice channel or voice circuit refers to a 64-kbps channel.

Chap. 9

Transmission Systems

383

spacings range from 6 to 25 miles depending on the intensity of local
rainfall.
3A-RDS uses a combination of existing TN-l analog radio equipment
and a 3A-RDT digital terminal. The 3A-RDT provides for the digital processing modulation and demodulation steps necessary for transmitting
the DS3 signal over a TN -1 radio channel. Regeneration is required at
approximately every other repeater site; at these sites, a 3A-RDT regenerator is provided in addition to the TN-l radio equipment.
DR6-30 is a high-capacity digital carrier system that operates at 6 GHz
in channels of 30-MHz bandwidth and can carry a maximum of 9408 digital voice circuits, or the equivalent in data circuits, in seven channels.
It is most often used on hops of about 15 to 30 miles. The first DR6-30
system went into service early in 1981, between Eugene and Roseburg,
Oregon-a 75-mile route over rugged mountain terrain in Pacific
Northwest Bell territory.
DRII-40, operating at 11 GHz in 40-MHz channel bandwidths, has a
capacity of 13,440 voice circuits, or the equivalent in data circuits, in ten
channels. It is used for shorter hops, usually 15 miles or less. Signals at
11 GHz are more susceptible to rain-induced fading than 6-GHz signals,
and so are usually restricted to the shorter distances. However, II-GHz
systems are used for the longer routes when the 6-GHz frequency spectrum is already fully occupied.
The first DRII-40 systems went into service in combination with
DR6-30 on a Michigan Bell route between Flint and Kalamazoo, with
branches to Saginaw and Big Rapids.
DR6-30 and DRII-40 are compatible with analog radio systems and
can share the same towers and antennas. This joint use is possible
because the design of the modulation scheme controls the spillover of
digital signal power in adjacent channels. The DR6-30 and DRII-40 systems use 16-state quadrature-amplitude modulation. In this scheme,
transmitted signal strength drops off sharply at the upper and lower frequency limits of a given channel, and for a given transmitted bit rate, the
bandwidth is smaller than in some other modulation methods. The digital channel, therefore, presents a very small and acceptable amount of
interference to any adjacent channels.
Lightwave Systems (FT3C). The FT3C lightwave digital transmission
system transmits digitally encoded information between offices in the
form of light pulses at the rate of 90.524 Mbps. One FT3C lightwave signal contains information for up to 1344 two-way voice channels over a
pair of glass fibers within a lightguide cable. A cable may contain up to
144 individual fibers, giving a maximum capacity of over 80,000 voice circuits per cable, including protection.
The FT3C facility is intended for applications with medium-to-Iarge
service cross sections. The lower time-division multiplex transmission

384

Network and Customer-Services
Systems

Part 3

rates are multiplexed to the FT3C rate. An FT3C terminal can also be
equipped to serve as an express terminal; that is, a terminal to interconnect back-to-back FT3C maintenance spans without appearing at a crossconnect bay. This permits violation monitoring, protection switching,
and other maintenance functions to be performed in the express
configuration.
Figure 9-25 is a simplified block diagram of a typical FT3C facility.
The lightguide cable used between offices consists of glass fibers packaged in ribbons of twelve fibers each. One to twelve ribbons are
enclosed in a protective sheath with an outside diameter of about onehalf inch independent of the number of ribbons. A separate fiber is
required for each direction of transmission.
Regenerators are required along the transmission path and are located
in line repeater stations (LRSs). The maximum distance between LRSs is
determined primarily by the grade of lightguide cable selected and the
number of splices used to construct the cable path from regenerator to
regenerator. At present, for example, a regenerator section length of 5.3
miles is attainable using the best grade of cable and an average distance
of about 1600 feet between splices. Longer regenerator spacing is permissible when fewer splices are used. The grade of cable and permissible
fiber path loss are system parameters subject to frequent modification due
to rapid technological change.
The lightwave regenerator requires an environment similar to that
found in central offices, but it may be located in a suitable hut, controlled
environment vault (CEV), or leased space. In a hut, CEV, or leased space,
power may be provided from a power plant that converts local ac power
to dc power with a battery reserve.
The FT3C facility uses an MX3C lightwave terminating frame (LTF) to
terminate up to ten 2-way lightwave service lines and up to two 2-way
lightwave protection lines.
Lightguide cable interconnection equipment (LCIE) terminates the
lightguide cable sheath and provides appearances of the individual fibers
on an array of single-fiber connectors. Single-fiber interconnection cables
connect the lightguide cable to terminal equipment or repeater station
regenerators.
The FT3C lightwave line is designed to the following transmission
objectives for system lengths up to 4000 miles:
• an average error rate of better than 10-8
• a service outage caused by equipment failures of not more than 0.02
percent, that is, 1.7 hours per year.
Outage is that percentage of time during which service over a given line
is interrupted because of an unprotected failure.

OFFICE A

TO
DSX-1
DSX-1C
DSX-2
OR
DSX-3
CROSSCONNECT
BAY(S)

I
Ij4- OUTSIDE PLANT ~I

I

I

~ OUTSIDE PLANT ~

OFFICE B

MX3C
LTF

MX3C
LTF

LOCAL DC POWER
(-24V. -48V. OR
+140V)

LCIE
LRS
LTF

I

LlGHTGUIDE
CABLE

LlGHTGUIDE CABLE INTERCONNECTION EQUIPMENT
LINE REPEATER STATION
LIGHTWAVE TERMINATING FRAME

Figure 9-25. Block diagram of a typical FT3C lightwave system facility.

TO
DSX-1
DSX-1C
DSX-2
OR
DSX-3
CROSSCONNECT
BAY(S)

Network and Customer-Services
Systems

386

Part 3

9.4.3 DIGITAL MULTIPLEX EQUIPMENT
The digital carrier systems described in the previous sections operate at
several different transmission or bit rates. As noted earlier, the bit rates
are levels in the time-division multiplex (TDM) hierarchy. A variety of
multiplex equipment has been designed to interface between the
different bit rates. This section describes the TDM hierarchy and several
kinds of multiplex equipment.
The TDM Hierarchy
Figure 9-26 shows the TDM hierarchy, generally described as DS levels 0
through 4. The 0- to 4-kHz nominal voiceband channel is shown below
the DSO level to illustrate the first step in combining voice or other analog signals into 64-kbps pulse-co de-modulated signals. As described in
Section 6.5.2, twenty-four voiceband analog channels are combined or
multiplexed to form a DSI signal (1.544 Mbps), also called a digroup (for
digital group). Digital transmission systems are shown on the right at the
corresponding transmission rates.
In the case of a short-haul Tl carrier system carrying a 24-channel circuit group, the channel terminal equipment bears a unique relationship
to (and is normally thought of as part of) the transmission facility. For
DIGITAL
FACILITIES

DIGITAL
LEVELS
274.176 Mbps (4032 CHANNELS)

DS4

-----------------------------------------r------~T4M,DR18

DS3

_44_.7_3_6_M_b_PS_(_6_72_C_H_A_N_N_EL_S_)__- r__________~----~~~____~FT3,3ARDS
DR6-30

DS2

---------------------+------r---~----~----~--~T2

6.312 Mbps (96 CHANNELS)

3.152 Mbps (48 CHANNELS)
DS1C

1
DS1

+-.......______+-__---1____

1-..._ _ _ _

T 1C, T1 D

_1._5_44_M_b_P_S_(2_4_C_H_A_NN_E_L_S)__--.a.-z-______----L---I______--.L_ _ _ _-+__.... T 1, T1 lOS
1ARDS

0.064 Mbps (1 CHANNEL)
DSO

(1 ANALOG CHANNEL)
0-4 kHz ______________________
......1.._ _ _ _ _ _ _ _--.L_ _ _ _ _ _ _ _---.,;K.__ __

Figure 9-26. PCM-TDM hierarchy.

Chap. 9

Transmission Systems

387

the higher bit rate systems (for example, T2, T4M, DRI8, and FT3), the
multiplex arrangements have evolved in a hierarchical manner to be consistent with the capabilities of the various media for high-speed digital
transmission.

Channel Banks
Digital channel banks (or primary pulse-code modulation [PCM] multiplexers, as they are called in CCITT terminology) have two basic functions: They convert analog voice signals into digital form (and vice
versa), and they combine the resulting digital signals into a single digital
bit stream by time-division multiplexing (and vice versa). The D-type
channel banks have a number of characteristics in common:
• They have equipment to provide the proper VF and signaling interfaces with the central office trunk circuits or other assigned circuits.
• They use filters to limit the transmitter input frequency range and to
reconstruct voice signals at the receive output.
• They use sampling gates to sample transmitter input signals and to
deliver receiver output samples to the correct channels.
• They have either a nonlinear encoder and decoder or a linear codec
(coder-decoder) combined with a compressor and expandor to convert
pulse-amplitude-modulated samples to PCM samples.
• They provide a means of controlling the timing of the sampling gates
and the location of PCM and signaling bits in the bit stream format.
• They use the carrier group alarm to provide carrier failure alarms and
trunk conditioning to minimize the effects of outages.
• They have power supplies to provide the dc-to-dc conversion needed
to convert -48 volt office supply voltages to the voltages required in
the bay.
D-type channel banks of the types DIA, B, C through D4 are 24-, 48-,
or 96-channel carrier terminals that are composed of a commonequipment portion and a set of up to twenty-four, forty-eight, or ninetysix individual channel units. The common-equipment units have those
functions that are common to all the channels and have characteristics
that do not vary with the types of trunk circuits assigned to the channels.
Channel units are individual to a particular channel, and their basic function is to provide the interface between the central office trunk circuits,
or other assigned circuits, and the channel bank. They include hybrids
(if needed) to convert between 2-wire and 4-wire transmission, transformers for 4-wire circuits, and maintenance access for each channel.

388

Network and Customer-Services
Systems

Part 3

Plugging in the correct channel unit furnishes proper transmission and
network signaling for a particular trunk.
The OIA channel bank was the first digital channel bank developed.
It consists of twenty-four plug-in channel units and twenty-nine plug-in
common-equipment units. The OIB channel bank was developed to
allow 4-state signaling; it is identical to the OIA bank except in the Jormat used to code signaling information. The OIC is a specialized bank
designed to provide remote operator capability for Traffic Service Position
Systems (TSPSs).24 It permits transmission and reception of twenty-four
voice channels and a 24-bit parallel data link between toll connecting
trunks at a central office (TSPS base location) and a TSPS No. 1 at a
remote office (remote operator group). The 010 channel bank provides
direct, tandem, and toll connecting trunk facilities on an end-to-end basis
with 02, 03, 04, other terminal equipment with a compatible OSI interface, or another 010 bank.
All the O-type channel banks, except the 01 series, are suitable for use
on intertoll trunks. A performance level superior to that of the 01 bank
is required because toll calls may include several digital trunks interconnected by switching systems. For switching systems that can handle signals only in voice-frequency form, quantizing itoise will accumulate
because of analog-to-digital and digital-to-analog conversions at each
switching system. For digital switching, a signal converted into digital
form by one channel bank should be capable of being reconstructed by
any other channel bank (except OlA, B, C). Thus, a high degree of standardization and uniformity is required of all such channel banks.
The 02 channel bank is a 96-channel carrier terminal delivering four
independent 1.544-Mbps outputs, each one with the same output capacity
as a 01 bank. It is effectively four independent channel banks of
twenty-four channels each, even though some of its circuits are shared by
all ninety-six channels. The 02 banks have 104 plug-in common units
for ninety-six voice channels with circuits serving eight, twenty-four, or
ninety-six channels on each unit. The number of channels, the sampling
rate, and the output bit rate for 02 were chosen to match those of the 01
channel bank and the T1 line, but the 02 channel bank is not end-to-end
compatible with the 01A or 01B because the amplitudes of voice samples
are encoded differently.
The 03 channel bank is a carrier terminal used for processing twentyfour VF channels into a OSl signal. It is suitable for use in direct, tandem, toll connecting, and intertoll trunks. The 03 channel bank is much
smaller than 01; just five plug-in units are required for the commonequipment circuitry versus twenty-nine for 01. All the circuitry of a
given channel is contained wholly within the channel plug-in unit and

24 Section 10.4.1 discusses TSPSs.

Chap. 9

Transmission Systems

389

not within common equipment, so that single-channel failures can be
easily restored by simply replacing the unit.
The 04 channel bank is a 48-channel carrier terminal that is organized
into two 24-channel digroups. It can be operated in anyone of five
different modes to produce OSl, OSlC, or OS2 outputs (2 times
48-channel terminals in the case of OS2) to match the line and far-end
equipment. It has' from eight to eleven plug-in common units, depending on the mode, for forty-eight channels with circuits serving twentyfour or forty-eight channels. 04 is end-to-end compatible with DID, 02,
03, and other 04 channel banks. 04 channel banks provide a wide range
of channel unit types to serve both message and special-services applications. In addition, the 04 bank can be used to provide the 64-kbps rate
of the Digital Data System using dataport channel units.

ESS Switching Equipment Interfaces

Digital interfaces between ESS switching equipment and transmission
facilities provide cost, space, and maintenance advantages over analog
interfaces. That portion of the interface that directly terminates transmission lines is called the transmission interface. With this definition in mind,
the ESS switching equipment interfaces can be thought of as comprising a
number of transmission interface units (digital terminals), along with
associated controllers that serve as control and maintenance interfaces
with the particular ESS system used.
The first integrated digital terminal, the digroup terminal, was
developed in conjunction with the 4ESS switching equipment. The
digroup terminal unit (OTU) provides a bidirectional transmission interface between five OSl-rate lines and one time-slot interchange (TSI) port
on the 4ESS switch. Eight working OTUs, protected by a switchable
spare, and a digroup terminal controller (OTC) that serves as the control
and maintenance interface with the 1A processor (which also controls
operation of the 4ESS switching equipment) make up a digroup terminal
frame. Each frame has the capability of terminating forty digroups or 960
working channels. Thus, a digroup terminal frame provides an interface
between forty OSl-rate lines and the 4ESS switching equipment. In
actual practice, a digroup terminal complex comprises as many digroup
terminal frames as are required to accommodate digital trunks interfacing
with a 4ESS switching office.
The digital interface frame (OIF) is a more recent 4ESS switch digital
interface that performs functions previously performed by the digroup
terminal and its associated signal processor (SP2). One OIF can replace
four digroup terminal frames and one SP2. The design philosophy of the
OIF is oriented towards a per-digroup approach rather than the commonequipment design approach used in the digroup terminal. The ability to

390

Network and Customer-Services
Systems

Part 3

do so economically stems from the incorporation of several custom largescale integrated circuits in the per-digroup circuitry. The per-digroup
arrangement also simplifies maintenance and isolates faults in perdigroup circuitry from common-control functions. The digital interface
unit (DIU) provides a transmission interface between five DSI-rate lines
and one TSI port of the 4ESS switch. Thirty-two working DIUs, protected
by two spares, plus a digital interface controller (DIC), which provides
the signaling and control interface with the lA processor within the 4ESS
switch, make up a DIF. Thus, each DIF has the capability of terminating
160 digroups or 3840 channels.
The digital carrier trunk (DCT) bank provides a transmission interface
between digital trunks provided by the carrier facilities trunk link network (TLN) and the trunk distributing frame of the 1/ lAESS switching
equipment. It is a representative of flow-through technology: equipment
based on the D4 bank designed to operate in the lESS switching environment. DCT banks differ from stand-alone D4 channel banks in two
respects:
• DCT channel units combine both digital channel unit functions and
switching system trunk circuit functions.
• DCT banks have a third port connection that permits the exchange of
signaling, supervisory, and maintenance information with the
l/lAESS switch in digital format.
Ten DCT banks plus a peripheral unit controller-a redundant
microprocessor system that serves as the interface with the processor of
the 1/ lAESS switching equipment-make up a digital carrier trunk frame.
Since each DCT bank can accommodate 48 VF channels, each DCT frame
has the capability of terminating 480 VF channels. On the carrier side,
these channels are organized into digroups of twenty-four channels each
and may be carried over T-carrier facilities at either the DSI rate, DSIC
rate, or the DS2 rate. So, the DCT frame may terminate twenty, ten, or
five T-carrier lines depending upon the line rate.
The digital facility interface (DFI) is an interface unit that provides a
transmission interface between a digital transmission facility and the 5ESS
switching equipment. Up to fifteen DFIs compose one digital line/trunk
unit (DLTU), which directly interfaces with DSI terminations. Interfaces
to DS1C and to the international 32-channel digital carrier are being
planned.
Not covered within this section are the 2/2BESS and 3ESS switching
equipment interfaces. The 2/2BESS switch direct interface with T-carrier
permits direct control of T-carrier D3 and/or D4 channel units by the
2/2BESS switch processors. The 3ESS switch direct interface with Tcarrier permits direct control of T-carrier D4 channel units by the 3ESS

Chap. 9

Transmission Systems

391

switch processor. These procedures entail the use of standard D3 and D4
channel banks with a new channel unit.
Digital Access and Cross-Connect System
The Digital Access and Cross-Connect System (DACS) provides perchannel DSO (64 kbps) electronic cross-connection and test access to individual channels in analog or digital form from the DSI signal. DACS
terminates up to 127 DSI signals (3048 DSO channels), providing a
maximum of 1524 DSO cross-connections. DACS is a new element in the
facility makeup of Tl systems. It is not a direct replacement for, nor an
alternative to, any single existing hardware entity. The cross-connect
capability of DACS permits the" assignment and redistribution of 64-kbps
channels among Tl systems at the digital level. This capability can be
used to collect channels with common destinations and thus increase the
fill of Tl lines and terminal equipment. It also reduces the need for
back-to-back channel banks to convert signals to analog form and then
back to digital form. The cross-connect capability of DACS can be used to
segregate channels by type (for example, message / special services,
4-wire/2-wire), which will result in simplification of central office interconnection and testing arrangements. The test access capability of DACS
may serve as an alternative to jack or Switched Maintenance Access System (SMAS)25 arrangements in the digital environment. Benefits derived
from DACS applications include capital savings resulting from reductions
in the number of intermediate distribution frames and associated wiring,
reductions in back-to-back D-type channel banks and SMAS connections,
and more efficient use of Tl lines and digital terminals. Expense savings
stemming from easier administration and improved facility maintenance
are also expected.
Other Multiplex Equipment
The digital multiplexers provide the interfaces between the different bit
rates in the digital network. A multiplexer contains one or several muldems, each consisting of a multiplexer-demultiplexer combination. The
multiplexer may also contain in-service performance monitoring and
automatic protection-switching circuitry. All multiplexers except the
MIC are available with time-shared, in-service performance monitoring
and automatic protection switching. These two features are usually combined and are optional in some of the multiplexers. In multiplexers
25 Through the use of relays, SMAS provides concentrated metallic access to individual
circuits to permit remote access and testing by the Switched Access Remote Test System
(SARTS).

392

Network and Customer-Services
Systems

Part 3

equipped with monitoring and switching, the monitors sequentially
check each muldem by comparing the inputs and outputs of the muldem
with the inputs and output of a reference demultiplexer contained in the
monitor. All input signals are also checked for integrity. Muldem
failures are alarmed and result in an automatic switch to the standby muldem (if available). The failure detection and switch in the M13 and M34,
but not the M12, are completed rapidly enough to avoid dropping calls in
progress, although data service may be affected. Input signal failures
cause an alarm but do not cause automatic switching. The time-shared
monitor provides diagnostic indicators and issues the appropriate office
alarms. These can also be sent to a remote location. For maintenance
purposes, it should be noted that OSI signals out of a multiplexer are
always free of bipolar violations, even when bipolar violations are
present at the entrance to the multiplex at the preceding end. For example, if a higher capacity transmission facility with the required multiplexers at each end constitutes one link within a cascade of Tl spans,
bipolar violations occurring ahead of that facility cannot be measured
downstream from that facility (as they could if that facility were just
another Tl span). The multiplexer units described below and their position in the TOM hierarchy are illustrated in Figure 9-26.
MIC Multiplexer. The MIC multiplexer bay contains up to forty-eight
operating muldems and one standby muldem. An MIC muldem accepts
two digital signals at the OSI rate and multiplexes them into a single
OSlC signal that is transmitted over a TIC or a TID digital line. The
MIC muldem can also accept a OSIC signal from a TIC or TID digital
line and demultiplex it into two digital signals at the OSI rate. The
standby muldem is a hot spare that must be manually switched to replace
any failed in-service muldem.
M12 Multiplexer. The MI2 digital multiplexer combines four OSI signals to form a OS2 signal. One standby muldem is used for the protection of up to twenty-three working muldems. An MI2 multiplex frame
can contain up to a maximum of twenty-four M12 muldems, depending
upon the particular frame configuration desired.
M13 and MX3 Multiplexers. The MI3 digital multiplexer provides a
transmission interface between the OSI and OS3 digital signal rates.
Access to it is normally provided through the OSX-l and DSX-3 crossconnects. An MI3 multiplexer frame contains two M13 muldems. Multiplexing is accomplished in two steps. In the first step, up to four asynchronous incoming OSI signals are multiplexed into a single internal
OS2 signal. This is accomplished in the low-speed multiplexer portion of
an M13 muldem. In the second step, up to seven of those OS2 signals are

Chap. 9

Transmission Systems

393

multiplexed into a single DS3 signal. This is accomplished in the highspeed multiplexer portion of the M13 muldem. The demultiplexer portion first recovers the seven DS2 signals from a received DS3 signal and
then the four DSI signals from each of the DS2 signals, thus arriving at
the original twenty-eight DSI signals. Protection for the Ml3 muldems is
accomplished by providing one standby for each working high-speed
muldem portion and one standby for up to fourteen working low-speed
muldem portions.
The MX3 is a family of digital multiplexers that will be available in a
number of configurations and options; the MX3 will serve as a standalone multiplexer, a line-terminating multiplexer for FT3 or DR6, or as a
span terminal for FT3. The MX3 will interconnect a DSX-I, DSX-IC, or
DSX-2 cross-connect with a DSX-3 cross-connect, the FT3 fiber optic line,
or the DR6 digital radio. The MX3 multiplexer is expected to replace the
Ml3 in multiplex applications and, as a line terminal for FT3, find use in
up to 70 percent of the metropolitan digital trunk applications. The
stand-alone version of the MX3, the MX3 multiplexer assembly, is a
single MX3 muldem intended for small installations where cost, space,
and power are major considerations.

M34 Multiplexer. The M34 digital multiplexer provides a transmission
interface between the DS3 and DS4 digital signal rates. Access to it is
normally provided through the DSX-3 and DSX-4 cross-connects. One
standby muldem is provided for the protection of up to ten working muldems. A fully equipped M34 multiplexer arrangement contains up to
eleven M34 muldems in a total of seven frames. The multiplexer portion
of each muldem time-division multiplexes up to six asynchronous incoming DS3 signals into a single DS4 signal. The demultiplexer portion
recovers the original six DS3 signals from a received DS4 signal.

9.5 TRANSMISSION APPLICATION OVERVIEW
Previous sections of this chapter describe various transmission systems.
This section provides a breakdown by transmission facility category for
the loop and interoffice application areas. Data are presented to give both
a recent (1980) picture of the breakdown and an indication of the trends
over the past 20 years.

9.5.1 LOOP APPLICATIONS
Nearly all customer loops are now carried at voice frequency on individual wire pairs. Only a few percent of all loops are on carrier facilities
because most loops are short, making it difficult to compensate for the
cost of carrier terminals. Where loops are long, the cross sections tend to

Network and Lustomer-~ervices
Systems

394

Part 3

be small, which in turn, tends to make carrier facilities uneconomical.
Recent digital technology improvements, together with the increasing
cost of wire, have resulted in some penetration of carrier systems into the
loop plant. However, the increasing use of local digital switching systems, which permit lower termination costs for digital facilities, should
permit digital carrier systems to become an important factor in the loop
plant in the future.

9.5.2 INTEROFFICE APPLICATIONS
The application of different facilities to the interoffice area can be viewed
in two important ways: in terms of the number of circuits (or voiceband
channels) and in terms of circuit-miles. The first view is provided by
Table 9-8, which shows the breakdown of circuits by carrier terminal type
as of 1980. As indicated by the footnotes, each terminal type is used for
two or more transmission systems. Because of the large number of
transmission systems, a breakdown by system type would result in a large
table and obscure an important point: that digital facilities terminated on
D-type channel banks account for most (68 percent in 1980) of the
interoffice circuits. Table 9-9 shows the trend in terminal types from 1960
to 1980. The evolution to digital facilities in the interoffice plant is
clearly seen, with an increasing rate of replacement of analog terminals
by digital terminals.

TABLE 9-8
CARRIER CIRCUITS BY TYPE (1980)
Type

Thousands
of Circuits

Percent

A-type channel banks*
N-carrier terminalst
D-type channel banks:j:

1164
1979
6752

11.8
20.0
68.2

Total

9895

NOTE: Circuits shipped prior to N-carrier, with the exception of A-type
channel banks, are not included since they have almost all been retired.
* Used for analog L-carrier and analog radio.

t Total includes a

and ON terminals.

:j: Includes digital switching system interface terminals; used for all
digital interoffice facilities.

395

Transmission Systems

Chap. 9

TABLE 9-9
PERCENTAGE OF TERMINALS SHIPPED, 1960 TO 1980
1960

1965

1970

1975

1980

A -type channel banks * 21.9
78.1
N -carrier terminalst
D-type channel banks:j: 0.0

23.8
64.8
11.4

24.0
43.0
33.0

21.0
29.3
49.7

11.8
20.0
68.2

Type

NOTE: Circuits shipped prior to N-carrier, with the exception of A-type channel banks, are
not included since they have almost all been retired.

* Used for analog L-carrier and analog radio.

t

Total includes

a

and ON terminals.

:j: Includes digital switching system interface terminals; used for all digital interoffice
facilities.

Table 9-10 provides another view of interoffice applications. Here, the
categories are generic facility types, and the data indicate that digital
facilities account for only about 11 percent (in 1980) of the total
interoffice circuit-miles. Analog systems dominate by this measure, with
analog radio alone accounting for about 60 percent of all circuit-miles.
As shown in Table 9-11, analog radio has essentially maintained a constant share of total circuit-miles in recent years, largely because of
improvements discussed in Section 9.3.4 that increased system capacity.
TABLE 9-10
CARRIER CIRCUIT-MILES BY TYPE (1980)
Type

Analog paired cable
Analog coaxial cable
Undersea cable
Analog radio
Satellite
Digital paired cable
Digital coaxial cable
Digital radio
Total

Millions of
Circuit-Miles

38
197
15
623
36
102
1
11
1023

Percent

3.7
19.3
1.5
60.9
3.5
10.0
0.1
1.1

Network and Lustomer-~ervices
Systems

396

Part 3

TABLE 9-11
PERCENTAGE OF CARRIER CIRCUIT-MILES
BY TYPE, 1960 TO 1980
Type

1960

1965

1970

1975

1980

Analog paired cable
Analog coaxial cable
Undersea cable
Analog radio
Satellite
Digital paired cable
Digital coaxial cable
Digital radio

29.8
31.5
0.4
38.3
0.0
0.0
0.0
0.0

19.6
25.7
1.0
52.7
0.0
1.0
0.0
0.0

10.3
23.6
0.9
61.9
0.0
3.3
0.0
0.0

6.6
22.8
1.0
62.1
0.0
7.5
0.004
0.00

3.7
19.3
1.4
60.9
3.6
10.0
0.14
1.0

Thus, while the use of digital carrier has been growing rapidly in the
interoffice plant, the amount of penetration differs significantly according
to the application. Short-haul circuits, such as those in metropolitan
areas, have a high penetration of digital facilities. The long-haul intercity
circuits are mostly on analog facilities, primarily analog radio. It should
be noted, however, that there has been an active program to provide digital capability over long-haul analog facilities, thereby increasing the
amount of end-to-end digital connectivity in the network. The lA-RDS
channel referred to as data under voice (see Section 9.4.2) is one example of
this kind of adaptation.

AUTHORS
E. J. Anderson
R. A. Bruce
W. C. Roesel
H. R. Westerman

10
Network Switching Systems

10.1 INTRODUCTION
Switching systems used in the Bell System can be divided into two broad
functional categories: those designed for local switching and those
designed for tandem switching. Local offices connect customer lines to
each other for local calls and connect lines to trunks for interoffice calls.
Tandem switching has two applications. Offices that connect trunks to
trunks within a metropolitan area are referred to as local tandem offices.
Offices that connect trunks to trunks to form the toll network portion
(class 1 to class 4) of the hierarchical public switched telephone network
(PSTN) are called toll offices (see Chapter 4).
There are significant differences in requirements for these areas of
application. Consequently, systems designed primarily for local switching applications (often called local switching systems) are different in architecture and function from those designed primarily for local tandem or
toll switching. As discussed in Section 4.2 and shown in Figure 4-4,
many switching systems serve more than one role in the PSTN, and in
particular, they may provide both local and tandem switching functions.
Because lines and trunks connecting nearby offices usually generate a
load per termination that is less than the link capacity within a switching
network, local switching systems generally employ concentration of lines
and expansion to trunks (see Section 7.3). In tandem and toll applications,
however, trunks are more heavily used, so expansion of trunks to the network is frequently used.
Because lines and short-distance trunks generally use 2-wire transmission, the telephone connections within local and local tandem offices
have generally been 2-wire. Time-division digital systems, however, usually switch on a 4-wire basis (that is, separate paths for each direction of

397

398

Network and Customer-Services
Systems

Part 3

transmission). Toll facilities are usually also 4-wire, so toll switching systems switch on a 4-wire basis.
Functions needed to provide exchange services l are built into the local
switching system because of the convenience resulting from the direct
interface with customer lines. The geographic centralization of the tandem office, however, offers efficiency in providing centralized billing and
operator and network services.
The Bell System formally began installing automatic switching in
1919. Since that time, the market for switching systems has expanded
continuously in terms of both the number of systems and their applications. The growth of cities, establishment of new population centers in
the suburbs, increased use of the telephone, and demand for new services
all contribute to the expanding market.
The evolution of switching systems has been marked by a flow of new
technology. As new technology is developed, new capabilities become
available at lower cost for the basic switching functions and new customer services. Switching networks have evolved from progressive switching to coordinate switching to the current technology of time-division
switching. Switching control has evolved from direct progressive control
through common control by electromagnetic registers and markers to the
current technology of stored-program control.
Because of the large base of existing systems and the cost and effort
associated with replacement, older systems that still provide satisfactory
service may remain in service for a considerable time after introduction
of a new system. As a result, several systems of various vintages exist in
the PSTN.

10.2 ELECTROMECHANICAL SWITCHING SYSTEMS
10.2.1 EVOLUTION
Local Switching
The earliest automatic switching system was the step-by-step (SXS) system, known around the world as the Strowger system. Although it was
invented in 1889 by A. B. Strowger and first installed in· 1892, the Bell
System did not begin using the step-by-step system extensively until
1919, and even then, the equipment was installed by the Automatic Electric Company. One reason for the delay in applying step-by-step systems
was the high percentage of Bell System customers who were located in
large cities where step-by-step systems were not economically attractive.
However, by 1921, Western Electric did begin installing them in the
smaller cities. Western Electric acquired licensing agreements with
Automatic Electric in 1916 and began many design improvements leading
1 Section 2.4 describes exchange services.

Chap. 10

Network Switching Systems

399

to its own system design, which appeared in 1926. Today, step-by-step
systems are used in rural and suburban areas and even in some metropolitan areas that began small but grew extensively.
In the early 1900s, the Bell System began working on an automatic
system to provide efficient service in large cities; the result was the panel
system, first placed in service in 1921. Originally introduced as a local
switching system, it was later adapted for local tandem operation. This
system used a register, called a sender, to store dialed digits and to control
progressive originating and terminating switching networks. At their
peak, in the 1950s, panel offices served nearly 4 million customer lines in
Bell operating companies in many large cities. Their replacement by
more modern systems began in the 1960s, and the last panel system was
retired on September 11, 1982, in Newark, New Jersey.
In 1938, following the invention of the crossbar switch and advances
in relay technology, the Bell System introduced another metropolitan
switching system, the No.1 Crossbar System. The No.1 Crossbar used
separate originating and terminating coordinate switching networks.
Each network was controlled by a group of markers that interpreted
digits, selected a network path, and caused the proper network switches
to operate. The No.1 Crossbar was designed for large-size offices and
used primarily to meet the substantial growth in demand for telephone
service in cities.
After World War II, design was started on a crossbar system adapted to
the needs of suburbs and small cities. This system, the No.5 Crossbar
System, met these needs through effective use of the common-control
principle-using faster markers controlling a single coordinate switching
network instead of separate originating and terminating networks. Local
automatic message accounting (see Section 10.5.3) was incorporated into
the design. Later, centrex service was added. In fact, this flexible
crossbar system found economical applications beyond its original design
range-in large city central offices and, as a special version, in
AUTOVON. 2 The last new No.5 Crossbar to be installed went into service in November 1977.
Toll and Tandem Switching
To a large degree, toll and tandem switching systems have evolved following the same technological advances as local systems. Automatic
switching was first introduced in the toll environment during the early
1920s in Los Angeles, California, using step-by-step systems. Step-by-step
toll systems were initially used to carry short-haul toll traffic. But, by the
2 AUTOVON (automatic voice network) is a private voiceband network serving the
Department of Defense. It employs automatic switching and handles both voice and data
traffic.

400

Network and Customer-Services
Systems

Part 3

early 1940s, cities were being tied together by step-by-step systems in
long-haul dialing networks. All step-by-step toll systems are 2-wire, and
most are directly controlled systems (no senders, decoders, translators,
etc.). About sixty have been modified for centralized automatic message
accounting (discussed in Section 10.5.4). These use common-control
features for the routing and charging associated with customer-dialed
traffic. Nearly all step-by-step switching systems that do tandem switching are in class 4 offices.
Many early panel systems were used within cities as local tandem
offices. The crossbar tandem switching system was designed as the successor for these local tandem panel systems and went into service in 1941.
In cases where a local crossbar tandem office had access to all the local
offices in an area, it was naturally positioned to handle toll calls originating or terminating in that area. The ability to complete toll traffic was
added in 1947, and originating toll functions were added in 1953.
Improved transmission techniques made the introduction of crossbar
tandem (a 2-wire system) as a through toll (class 3, 2, or 1) switching system possible in 1955. Foreign-area translation (the ability to translate 6digit area and office codes to derive more extensive trunk group choices)
was added to crossbar tandem in 1958.
The first crossbar switching system designed exclusively for toll service was the No.4 Crossbar System introduced in 1943 in Philadelphia,
Pennsylvania. In 1953, an improved version of the No.4 Crossbar, called
the No.4A Crossbar System, added foreign-area translation, automatic
alternate routing (the ability to route the call to other trunks groups if
the first route is busy), and address digit manipulation capabilities (converting the incoming address to a different address for route control in
subsequent offices, deleting digits, and prefixing new digits if needed).
The No. 4A Crossbar was intended for metropolitan areas and was the
largest of the Bell System's toll systems until 4ESS switching equipment
became available. The No.4A Crossbar is a 4-wire switching system.
The 4-wire design was chosen to eliminate the echo problems associated
with converting 4-wire to 2-wire transmission after projections of both
intertoll and toll connecting equipment indicated increased carrier (4wire) operation in the future.
10.2.2 STEP-BY-STEP SYSTEMS
The term step-by-step describes both the manner in which the switching
network path is established and the way in which each of the switches in
the path operates. The basic step-by-step system is classified as a direct
progressive control system because the dial pulses generated by the customer's telephone directly control the stepping switches of the progressive networks.

401

Network Switching Systems

Chap. 10

The switches are functionally described as line finders, selectors, and
connectors. Each of these switches combines vertical stepping and horizontal rotary stepping motions in a 2-stage selection process. A set of
wiper brush contact fingers is moved, first in a vertical direction to select
one of ten level positions, then in a horizontal direction until the
selected position of ten at that level is reached. Figure 10-1 shows the
terminal bank of a switch.

SLEEVE
BANK
SLEEVE
WIPER

_-ez.....~1l"'.;

VERTICAL
COMMUTATOR
(USED IN LINE
FINDERS)

WIPER
CORDS

Figure 10-1. Terminal bank of 100 customer lines.

Figure 10-2 shows, in simple form, a step-by-step system interconnecting 1000 lines. For such a system, customers would be numbered from
111 to 000. 3 When a customer goes off-hook, an idle line finder locates
the line requesting service through a vertical and horizontal hunt. The
line finder is wired to a selector switch that returns dial tone to the
caller. The selector switch moves up to the level determined by the first
digit dialed (in this example, the hundreds digit). Next, the switch
moves horizontally across the contacts in the selected level, hunting for a
circuit to an idle connector switch in the called customer's hundreds
group.
3 Step-by-step switches respond to rotary-dial pulses from 1 to 10 (zero); in actual practice,
the initial 0 is reserved as a single digit for connection to an operator.

r-------------------------·

I

I
I
I
I

I
I
I

I
CALLING
CUSTOMER'S
TELEPHONE

I
I

I--~­

I
CALLED
_.....:.1--1 CUSTOMER'S
I
TELEPHONE

L ________________________

J

Figure 10-2. A 1000-line switching system requiring 3-digit selections.

When the second digit is dialed, the connector switch is stepped vertically to the level corresponding to the digit, thereby selecting a particular
row of ten lines in that hundreds group. When the third digit is dialed,
the connector switch moves horizontally across the row of terminals to
the line dialed. If the line is idle, the called customer's telephone rings.
If the line is busy, a busy signal is returned to the calling customer.
The step-by-step system has proven to be a popular system in the past
because it is economical for basic functions and can be readily expanded
as the need develops. On January 1, 1983, only 13 percent of all Bell System lines were served by this equipment, but more than 45 percent of the
local switching system~ (including community dial offices) were step-bystep. However, the progressive control nature of the system precludes
the addition of new functions, such as TOUCH-TONE calling and alternate routing, without adding costly equipment to the office. Another
disadvantage, or limitation, is that the customer's telephone number is
determined by the physical termination (appearance) of the line or connector on the system. Customer lines cannot be moved to other terminations without changing the telephone number. Finally, the maintenance
cost of electromechanical, large-motion switches is high. For these reasons, operating companies are replacing step-by-step equipment with
flexible electronic switching systems (see Section 10.3).

10.2.3 NO.1 CROSSBAR SYSTEM
The No.1 Crossbar was developed for use in large metropolitan areas. It
first went into service in 1938 in Brooklyn, New York, and over 300 systems were subsequently installed, serving more than 7 million lines. It is
now being replaced by more modern systems. On January 1, 1983, 180
systems still remained, serving nearly 4 million lines.

402

Chap. 10

Network Switching Systems

403

As described in the previous section, step-by-step systems use devices
that cause selector brushes to wipe over contacts in either rotary or linear
motions to establish the network path progressively. The No.1 Crossbar
apparatus, in contrast, is a matrix of crosspoints associated with horizontal
and vertical bars that are selected to operate an associated crosspoint
within 60 milliseconds-much faster than the hundreds of milliseconds
of the step-by-step switch. A common set of logic (a marker) makes network connections in a few tenths of a second. The marker finds a path
through the network by finding a set of idle switch paths, one in each
stage of the network, that can be connected serially. It then establishes
the connection by direct orders to each switch. Other systems using
crossbar switches preceded the No.1 Crossbar, but this system was
unique in that its markers could set connections for a call in less than 1
second and then move on to the next call. This also contrasts with stepby-step systems where the control logic-approximately six relays associated with each of four or more step-by-step switches-is retained for the
duration of a call, however long it may take.
The No.1 Crossbar derives its name from and is built around the
crossbar switch shown in Figure 10-3. The name crossbar is derived from
the use of horizontal and vertical bars to select the contacts. There are
five selecting bars mounted horizontally across the front of each crossbar
switch. Each selecting bar can choose either of two horizontal rows of
contacts. The five horizontal selecting bars can therefore select ten horizontal rows of contacts. There are ten or twenty vertical units mounted
on the switch; each vertical unit forms one vertical path. Each switch has
either 100 or 200 sets of contacts, called crosspoints, depending on the
number of vertical units.
When one or the other of the two horizontal select magnets controlling the selecting bar operates, the bar is rotated up or down. This action
chooses one of the two horizontal paths available to this selecting bar by
moving ten or twenty flexible selecting fingers either up or down to positions adjacent to a crosspoint on each vertical column of crosspoints.
When rotated by a "hold magnet" and armature at one end, a vertical, or
holding, bar along a column of contacts presses against deflected fingers
in that column, closing the selected crosspoint.
After the operation of the hold magnet, the select magnet releases.
This restores the horizontal bar and all selecting fingers (except that held
at the closed crosspoint by an operated hold magnet) to normal. The
other horizontal path can then be selected for connection to another vertical path by rotating the select bar in the other direction. In ten separate
operations, up to 10 different crosspoints can be closed independently of
each other in a 100-crosspoint switch. This allows ten calls to pass
through a switch simultaneously, as opposed to only one call in a stepping switch.

Figure 10-3. A typical 20o-point crossbar switch.

Figure 10-4 is a simplified block diagram of a No. 1 Crossbar. The system uses separa te originating and terminating networks. (The first stage
of the line link frame is common to both networks.) A call is established
as follows : On requesting service, the calling customer is connected to a
district junctor4 and subscriber sender. The sender )'rovides dial tone and
then receives the called number as it is dialed by the customer. It is then
connected to an originating marker, which selects the trunk to be used
for the call and sets up the originating-end connection (by way of additional stages of crossbar switches in the district and office link frames)
4 The term district junctor is derived from a switching stage in panel systems that selected
the district of a city to which a call was directed.

404

FROM OTHER OFFICES

TO OTHER OFFICES

..

~

I
I

I
I
CALLING
LINE

LINE
LINK
FRAME

I--

DISTRICT
JUNCTOR

I--

DISTRICT
LINK
FRAMES

I

OFFICE
LINK
FRAMES

I--

TRUNKS

I
I

4

INCOMING
TRUNK
FRAMES

INCOMING
LINK
FRAMES

I--

CALLED
LINE

LINE
LINK
FRAME

I--

I
I

I

I
I

I

I
I

I
I
I

i

i.

SUBSCRIBER
SENDER

ORIGINA TING
MARKER

~-------

TERMINATING
SENDER

TERMINATING
MARKER

I

I
I
NUMBER
GROUP
FRAME

+- - - - - - - - - - - - - -

ORIGINATING -

-

-

-

-

-

-

-

-

-

-

-

-

-

-

- ..... -

-

-

-

-

-

INDICATES CONNECTIONS BY LINK
FRAMES OR CONNECTOR RELAYS

Figure 10-4. No. 1 Crossbar System block diagram.

-

-

-

-

TERMINATING -

-

-

-

-

-

-

-

-

-

....

406

Network ana Lustomer-:,ervlces
Systems

Part 3

from the district junctor to a trunk, to another office, or to the terminating network of the same office. When seized, the incoming trunk circuit
at the terminating end is connected to a terminating sender; this sender
registers the called number when it is sent by the subscriber sender. The
terminating sender is then connected to a terminating marker, which sets
up the connection of the trunk to the called customer's telephone.
Figure 10-4 also shows the number group. The terminating markers
consult this circuit to determine the called line location. The number
group also gives the marker the busy-idle status of the line being called
before attempting to set up a network connection. This was an early
design change that greatly increased the efficiency of the terminating
markers, allowing them to handle more traffic.

10.2.4 NO.5 CROSSBAR SYSTEM
The No.5 Crossbar was developed to fill a need for a switching system
suitable for suburban residential areas and smaller cities where it was
expected that a high percentage of calls would be completed to customers
in the same office. Another design consideration was the concept of
direct distance dialing that required automatic recording of call details for
billing purposes. The resulting system design proved to be suitable for
applications and features that went well beyond original plans. First
introduced in 1948, the No.5 Crossbar went on to serve more than
28 million lines and more than 40 percent of the Bell System's telephones
in the 1970s.
Figure 10-5 is a simplified block diagram of the basic No.5 Crossbar
equipment units. The equipment may be divided into two broad
categories: the switching network, through which all talking paths are
established, and the common-control equipment, which sets up the talking paths.
Customer lines appear on the line link frames, and trunks and originating registers on the trunk link frames. Each of these frames consists of a
number of crossbar switches. Connections are established from lines to
trunks or from lines to lines through an intraoffice trunk by the crossbar
switches on the line link and trunk link frames. The common-control
equipment used to establish the various connections includes registers,
markers, senders, number groups, and connectors. A dial-tone marker
sets up the connection to an originating register that provides dial tone
and receives the digits dialed by the customer. When dialing is complete,
a completing marker selects an appropriate idle trunk and sets up the
network path to complete the call. For calls leaving the office, the completing marker connects a sender to the selected trunk so that the necessary signaling between offices can take place. For incoming calls, an

r------------------------ --- - --...,

I

I
CUSTOMER
LINES

I

SWITCHING NETWORK

I
I
I

INTRA-

I

OFFICE
TRUNK
CIRCUIT

I
I
I
I

TRUNKS

COMMONCONTROL
EQUIPMENT
CONNECTORS

Figure 10-5. No.5 Crossbar System block diagram.

incoming register receives the directory number of the called party from
the distant office. A completing marker is required to obtain the line link
frame location of the called customer from the number group. Once the
connection is established, the common-control equipment is released;
only the line link, trunk link, and trunk circuits (that form the transmission path) remain in the connection.
During the 30-year history of the No.5 Crossbar, important improvements have been made and functions added:
• centralized automatic message accounting, making the No.5 Crossbar
more attractive as a small toll office.
• line link pulsing to facilitate direct inward dialing to stations served
by a dial private branch exchange (PBX).
• international direct distance dialing, allowing customers to dial overseas calls with up to twelve digits.
• centrex service, including station-controlled dial transfer.
• automatic call distributor (ACD) capability. (In this capacity, a
No.5 Crossbar is used only as a large ACD for directory assistance and
intercept service; it does not perform central office functions.)
407

408

Network ana Lustomer-::,ervlces
Systems

Part 3

• the No.5 Electronic Translator System, which uses software instead
of wired cross-connections to provide line, trunk, and routing translations. The system also stores billing information for transmission via
data link to a centralized billing collection system.
10.2.5 NO. 4A CROSSBAR SYSTEM
No. 4A Crossbar is an electromechanical common-control system
designed for toll service, with crossbar switches making up its switching
network (see Figure 10-6). Trunk circuits provide signaling and transmission paths. Incoming and outgoing trunk link frames, using 4-wire
crossbar switches, provide paths for connecting incoming trunks to outgoing trunks. Senders register destination codes pulsed from preceding
offices and transmit the necessary information to the decoders. Decoders
and translators determine the proper trunk group to the called office.
Markers set up the connection from the incoming trunk to a selected outgoing trunk. A separate group of markers is associated with each specific
train, as shown in Figure 10-6. A marker, sender, decoder, and translator
are only associated with a call during part or all of the call setup time.
A number of improvements have been made to the No. 4A Crossbar.
In 1960, centralized automatic message accounting (CAMA) equipment
was added to record billing information automatically for local and toll
calls at a central point. CAMA provides automatic billing of calls from
customers served by local offices in which no AMA facilities are available.
In the No. 4A Crossbar, the translation function determines routing in
terms of specific trunk groups from address digits. s It also provides other
information, including the digits to outpulse and the type of outpulsing
(for example, multifrequency or dial pulsing). Originally, translation was
done by a device called a card translator; this device used phototransistors
and represented the first use of transistors in Bell System equipment. In
1969, the card translator was replaced by the No. 4A Electronic Translator System. The No. 4A translator, a stored-program control processor,
also allows the billing and route translation functions to be changed by
teletypewriter input. This expedites the task of making routing changes
to many switching offices in response to network changes, such as the
addition of a new toll office or the creation of a new numbering plan
area.
In 1973, the peripheral bus computer (PBC) was added. The PBC uses a
minicomputer associated with the No. 4A translator to provide traffic and
maintenance data. Summary reports are provided either by local
cathode-ray tube (CRT) terminals or via data lines to centralized operations systems (see Chapter 14).
5 The first three digits are used for calls within a numbering plan area (NPA); the first six
digits are used for calls to another (foreign) NP A. Section 4.3 discusses NPAs.

r----------------------,
INTERTOLL TRAIN

FROM
ANOTHER
TOLL OFFICE

I

TO A LOCAL
~i-+--~. OR TANDEM
OFFICE

I

I
I

I

I
I

I
I _____________________ _
L

L ______________________________________

~

Figure 10-6. No. 4A Crossbar System block diagram.

In 1976, common-channel interoffice signaling (see Section 8.5.5) was
added to the No. 4A translators to allow more efficient signaling between
toll offices.

10.3 ELECTRONIC SWITCHING SYSTEMS
10.3.1 CONCEPTS
During the 19505, Bell Laboratories began working on a new type of
switching system to meet the growing demand for telephone services
more economically. These electronic switching systems, made possible by
the invention of the transistor in 1947, applied the basic concepts of an
409

410

Network and Customer-Services
Systems

Part 3

electronic data processor, operating under the direction of a stored program, and high-speed switching networks. The stored-program control
concept allowed system designs to be far more flexible than before. With
stored-program control, it is still expensive to design new features but
relatively easy to install them.

Stored-Program Control
In most electronic switching systems, the logical steps involved in making telephone connections and providing services reside in a stored program (software). In wired logic systems, in contrast, the logic is designed
into electromechanical or electronic circuits (hardware). To change the
operation of an electromechanical system (for example, to provide a new
service), it was often necessary to redesign circuits and rewire them
extensively in the field. In electronic switching systems, new services
can frequently be implemented by making changes to the stored program. These services can be made available to customers soon after the
software has been centrally developed and distributed to the switching
systems.
The stored program controls the sequencing of operations required to
establish a telephone call, including such items as timing the duration of
signals. Thus, stored-program control can control a line or trunk circuit
according to its application, so that system cost can be reduced by minimizing the types of these circuits and the number of manufacturing wiring options needed.

The Processor
The switching network and peripheral units of electronic switching systems operate under orders received from an electronic processor.
Depending on the particular switching system, a central processor and / or
distributed processors are used. Each processor consists of generalpurpose registers and buffer circuits that perform information-processing
operations based on instructions in the stored program. These processors
are time shared by all calls they serve.
The processors used to control switching systems are similar to
general-purpose computers with some important differences. Processors
for switching systems are generally oriented toward logical rather than
arithmetic operations. Therefore, they do not require sophisticated logic
arithmetic units to multiply, divide, and perform high-order number
manipulation. Switching system processors are input/output driven and
are designed to process information in real time. That is, the system must
respond promptly to signals and data transmitted by customers and other
switching systems. Delays in processing could result in misrouted calls

C:hap.l0

Network Switching Systems

411

or other incorrect treatment. Switching system processors must be
extremely reliable. The system may be expected to provide continuous
service 99.999 percent of the time-about 2 hours downtime in 40 years.
In general, processors have access to two types of memory. The first
type, which contains the executable instructions, is usually protected
against inadvertent program overwrites. The second type, a temporary or
transient memory, is used by the system as a "scratch pad" and therefore
must be fully readable and writable. The temporary memory contains
information such as busy / idle status of lines and trunks, the digits being
received on a particular call, billing information for an existing call, and
the results of diagnostic tests.

Program Organization
The organization of the program ,is strongly influenced by the fact that it
must operate in real time. It must also respond to trouble-detection circuits designed into the hardware to ensure dependable operation. For
the program to meet all these requirements, it is necessary to establish a
hierarchy of program tasks. Some tasks must be performed on a strict
schedule; others may be delayed without significant adverse effects.
The central processor generally incorporates an interrupt mechanism
that momentarily seizes control of the system when a demand (nonscheduled) or clock (scheduled) interrupt occurs. The interrupt circuit
causes the system to stop its present program task, store the program
address at which it was interrupted, and then transfer control to the
appropriate program. When the interrupt programs are completed, control is returned to the program that was interrupted.
All systems of a particular type use one set of program instructions
(called a generic program) that is the same for each office. New functions
and features are provided in successive issues of the generic program,
and different offices of one type may have different issues at any given
time. This generic program includes all the functions necessary to cover
the possible office sizes that the type of system serves and also includes
means for handling changing traffic conditions and growth. The detailed
differences for each individual system are listed in parameter information. This approach simplifies record keeping since only the parameter
information that specifies the present office size and operating conditions
is unique to each application of a switching system.
In 1975, feature loading was adopted to allow the selective inclusion
of new program packages that, generally speaking, are large in size and
limited in application. By this time, the designing and debugging tools
and techniques had advanced to allow this capability. This made possible
the addition of ESS-ACD (automatic call distributor) switching systems
and cellular mobile telephone service with an electronic switching system
as the serving switch.

412

Network and Customer-Services
Systems

Part 3

Additional information that describes characteristics of each line and
trunk within the office, as well as the routing and charging characteristics
of the office, are found in the translations area of the program. This
information is used to relate directory numbers to equipment numbers
for lines and trunks, define classes of service and special treatments, and
determine charging rates based on the called number.

A Typical Call
The call program analyzes a call in discrete steps. These steps start when
a calling customer lifts the receiver or an incoming trunk is seized by a
distant switching system. The program then proceeds through a chain of
actions that ultimately establishes a talking connection. The system deals
with each step of the chain in turn. The following paragraphs are a
simplified description of how an electronic switching system processes
telephone calls.
At a fixed periodic rate, the system checks (scans) each customer's line
and each office trunk and records its state: on-hook (idle) or off-hook
(dialing, busy in a talking connection, etc.). The state detected at each
scan is compared with the state recorded at the previous scan. If there is
no change, no action is taken. If a change is detected, the system consults the program for the action to be taken. A change of state in a line
from receiver on-hook to receiver off-hook indicates that a customer
wishes to place a call. The program, therefore, directs that dial tone be
sent to the customer's line. Usually, this involves establishing a switching network path between the subscriber's line and a digit receiver. The
digit receiver connects the dial-tone source to the calling line. Dial tone
is removed upon the detection of the first received digit. A similar state
change on a trunk indicates an incoming call. In either case, scanning of
the digit receiver continues at regular intervals, and each digit is
recorded in turn.
After the system registers the digits, it again consults the program.
For an intraoffice call or an incoming call to a line, the next step is to
determine, by scanning that line, the state of the called telephone. If the
called telephone is busy, the program causes a busy tone to be sent to the
calling party. If the called telephone is idle, the program causes audible
ringing tone to be sent to the caller and the called telephone to begin
ringing. For calls to stations on distant switching systems, the· program
directs the seizing of a trunk and the transmission of the called number.
If the call is answered, this state is registered, and the ringing signal
and audible ringing tone are removed. A talking path is then established
between the calling and called terminals. The lines, or line and trunk,
involved in the call continue to be periodically scanned. No action is

Chap. 10

Network Switching Systems

413

taken until one of the parties hangs up, at which time the program
directs that the connection be taken down.
Networks
An electronic switching system may use either space-division or timedivision switching networks (see Section 7.3 for general descriptions of
these types of networks). The principles of stored-program control can
be applied independently of the network selected. The type of network
chosen for a system depends largely on the characteristics of the line and
trunk transmission environment in which the system will be used.
Space-division networks are economically compatible with analog line
and trunk transmission interfaces.
Time-division networks have
significant advantages where transmission facilities employing digital
multiplexing6 are expected to dominate. Use of a space-division switching network in areas dominated by digital transmission facilities or use of
a time-division network in an analog transmission environment requires
extensive use of digital-to-analog and analog-to-digital converters at the
interface between the transmission facilities and the switching system.

10.3.2 EVOLUTION OF ELECTRONIC SWITCHING
The Bell System's first trial of electronic switching took place in Morris,
Illinois, in 1960. The Morris trial culminated a 6-year development and
proved the viability of the stored-program control concept. The first
application of electronic local switching in the Bell System occurred in
May 1965 with the cutover of the first lESS switch in Succasunna, New
Jersey.
The lESS switching system was designed for use in areas where large
numbers of lines and lines with heavy traffic (primarily business customers) are served. The system has generally been used in areas serving
between 10,000 and 65,000 lines and has been the primary replacement
system for urban step-by-step and panel systems. The ease and flexibility
of adding new services made lESS switching equipment a natural replacement vehicle in city applications where the demand for new, sophisticated business and residence services is high.
The need for economical electronic switching in the 2000- to 10,000line offices was met with the introduction of the 2ESS switching system
in 1970. While there are many similarities between the lESS and 2ESS
switching systems (for example, stored-program control and switch elements), there are also important differences: The 2ESS switching equipment network architecture was designed to interface with customer lines
6 Section 6.5 discusses multiplexing.

414

Network and Customer-Services
Systems

Part 3

carrying lighter traffic, the features were oriented toward residential
rather than business lines, and the processor was smaller and less expensive. The market for 2ESS switching equipment was suburban residential
\
areas previously served by step-by-step systems and new suburban wire
centers.
Electronic switching was extended to rural areas serving fewer than
4500 lines with the introduction of 3ESS switching equipment in 1976.
Again, the 3ESS switching equipment design borrowed concepts from
lESS and 2ESS switching equipment but implemented a less expensive
processor and more economical network architecture.
In 1976, the first electronic toll switching system to operate a digital
time-division switching network under stored-program control, the 4ESS
system, was placed in service. It used a new control, the 1A processor,
for the first time to gain a call-carrying capacity in excess of 550,000
busy-hour7 calls. The toll transmission environment was experiencing an
accelerated growth in digital time-multiplexed facilities. A toll storedprogram control system with a digital time-division network provided
large economic advantages over its space-division electromechanical
predecessors.
The 1A processor was also designed for local switching application. It
doubled the call-carrying capacity of the lESS switching system and was
introduced in 1976 in the first 1AESS switch. The 1A processor was
designed to be retrofitted into a working lESS switch. The network capacity of lESS switching equipment was also doubled to allow the 1AESS
switch to serve 130,000 lines. Similarly, a new processor, added to 2ESS
switching equipment in 1976, doubled its traffic capability.
In 1981, local digital time-division switching systems first began service in the Bell System. Northern Telecom's DMS8-10 switching system
serves in rural and suburban offices and is more cost effective at these
office sizes than comparable space-division switching systems. A second
digital time-division switching system, the 5ESS system, began service in
1982 in local offices in Bell operating companies.
Electronic switching was also used to modernize toll operator functions. The stored-program control concepts and much of the lESS switching equipment hardware were used to implement the first Traffic Service
Position System (TSPS) in 1969. Like other electronic switching systems,
the TSPS has undergone many technological upgrades since 1969, including the application of a new processor. (Section 10.4.1 describes TSPS.)
Modernization of operator intercept service with stored-program control
concepts and automatically generated announcements began with the
Automatic Intercept System (AIS) in 1970. (Section 10.4.2 describes AIS.)

7 Section 5.2.3 discusses the busy-hour concept.
8 Trademark of Northern Telecom, Ltd.

Chap. 10

Network Switching Systems

415

10.3.3 SPACE-DIVISION ELECTRONIC SWITCHING SYSTEMS
lESS and lAESS Switching Equipment
lESS Switching Equipment Processor. Figure 10-7 shows the lESS
switching equipment processor community. It includes a fully duplicated
No.1 Central Processor Unit (central control), program store bus, call
store bus, program stores, and call stores. This duplication allows full
interchangeability; for example, each processor has access to all the
busses, and both call store busses and program store busses have access to
both of their duplicated memory units.
The lESS switching equipment uses permanent magnet twistor program
store modules as basic memory elements. These provide a memory that is
fundamentally read only so that neither software nor most hardware malfunctions can alter the information content. Program stores contain
131,072 words, each forty-four bits long (thirty-seven bits of information
and seven bits for error-correcting coding). The program store has a
cycle time of 5.5 microseconds.
Call store provides "scratch pad" or temporary duplicated memory in
the lESS switch for the storage of information related to the progress of
calls and the status of equipment. Originally, 8192 words, each twentyfour bits long, were provided per call store in the form of ferrite sheets.
Later, call stores containing 32,768 words were developed using core
memory. Both types of memory are readable and writable.
The lESS switching equipment may optionally be equipped with a
duplicated signal processor, a duplicated signal processor call store bus,
and duplicated signal processor call stores. The signal processor performs

CALL STORE BUS

I

...

CALL
STORE

r-

I
I
I
I

I
I

I

PROGRAM STORE BUS

I

I I

CALL
STORE

I
PROGRAM
STORE

CENTRAL
CONTROL

----------------------,
SIGNAL PROCESSOR CALL STORE BUS

I

I
SIGNAL
PROCESSOR

I
SIGNAL
PROCESSOR
CALL STORE

...

I
I
I

I

I

I
I
I

SIGNAL
PROCESSOR
CALL STORE

I

OPTIONAL
EQUIPMENT

~

I
I
I
IL _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ I
Figure 10-7. 1ESS switching equipment processor
community with optional signal processor.

...

I
PROGRAM
STORE

416

Network and Customer-Services
Systems

Part 3

many of the input/output functions, thus relieving the central processor
of this work and, thereby, increasing the call-processing capacity of the
system. The signal processor call stores contain the signal processor program instructions.
lA Processor. The 1A processor community is, in some ways, similar to
the control processor of the lESS switching equipment. As with the lESS
switching equipment, the central processor, program store bus, and call
store bus are fully duplicated. However, the 1A processor uses readable
and writable memory for both program store and call store memory.
Originally, 6s,s36-word core-type stores were used, but 262,144-word
semiconductor stores are now available. Each word contains twenty-four
bits of information and two parity-checking bits. The cycle time of the
1A processor with the faster semiconductor stores is 700 nanoseconds,
compared to 5.5 microseconds in the lESS switching equipment central
control and 1.4 microseconds in the 1A processor operating with core
stores.
Unlike the lESS switch, program stores are not fully duplicated in the
1AESS switch, but two spare stores are provided for reliability. A portion
of the 1A processor call store memory is duplicated. However, only one
copy of certain fault recognition programs, parameter information, and
translation data is provided. Another copy of the un duplicated program
store and call store information is provided in file store. File store uses a
disk bulk storage device that can be used to load either the spare program
stores or call stores. In this way, additional reliability is provided. In
addition, magnetic tape units in the 1A processor provide for system reinitialization and detailed call billing functions.

Networks and Periphery. The 1/ 1AESS switches use the same peripheral
equipment. This allows for the transition from a lESS switch to a 1AESS
switch as the office capacity requirements increase, without the replacement of the entire switching system. The major peripheral equipment
items are: the switching network and junctors; the scanners, signal distributors, and central pulse distributors; and the line, trunk, junctor,
and service circuits. The following paragraphs discuss these items in
more detail. Peripheral equipment typically makes up 90 to 95 percent of
the l/lAESS switching equipment.
Switching Networks and Junctors. The 1/ 1AESS switching equipment
networks perform two major functions: They provide a means for interconnecting lines, trunks and service circuits, and they match the relatively lightly used lines to the comparatively heavily used trunks and service circuits.

Chap. 10

Network Switching Systems

417

Figure 10-8 depicts the 1/ 1AESS switching equipment switching networks. They are com posed of two 4-stage switching arrays (line link networks [LLNs] and trunk link networks [TLNs]), interconnected by wires
called junctors,9 to provide a total of eight stages of switching.
Lines (generally associated with telephones) are connected to one side
of a line link network. Because of the statistical characteristics of telephone traffic and the relatively small amount of traffic generated by a
typical line, the LLN does not have to be a nonblocking network. That
is, it does not have to provide paths to all customers simultaneously.
Therefore, the LLN has a fixed concentration ratio between the number
of lines terminated on one side and the number of junctors terminated on
the other side. Standard 1/ 1AESS switch concentration ratios are 2 to 1, 3
to 1, 4 to 1, and 6 to 1. All are based on the expected customer usage
characteristics of an office. The smaller ratios are generally associated
with urban offices. By comparison, a trunk link network, because of the
heavier usage of trunk circuits and service circuits, uses a 1-to-1 or 3-to-2
concentration ratio between the number of trunks and service circuits
and the number of junctors.
LINES

JUNCTORS

••
•

J

LINE LINK
NETWORK
1

J

TRUNKS

TRUNK LINK
NETWORK
1

••
•

JUNCTOR CIRCUIT

Figure 10-8. 1/1AESS switching equipment
network structure.

9 Except that the connections from LLN to LLN contain junctor circuits to provide control
access and power for line-to-line calls.

Network and Customer-Services
Systems

418

Part 3

The 1 IIAESS switching equipment switching networks employ a coordinate network (see Section 7.3.2) and use a ferreed or remreed switch as
a crosspoint element. These crosspoint switches are compatible with both
the existing outside plant (by providing metallic interconnection) and the
high-speed electronic central control. The remreed crosspoint, the newer
version of the two, is more compact and thus requires less floor space.
Scanners, Signal Distributors, and Central Pulse Distributors.
Figure 10-9 depicts the communication links between the lilA processors and the peripheral circuits. Every telephone switching system embodies some mechanism for detecting service requests and for supervising
calls in progress. Input information of this nature is furnished to the
1 IIAESS switch by the operation of scanners that sample or scan lines and
trunks at discrete intervals of time as directed by the system. The result
of any given scan is generally compared to the previous scan to determine if any change in state (on-hook to off-hook, for example) has
occurred. The sensing element used in all 1 I IAESS switching equipment
scanners is the ferrod sensor, a current-sensing device that operates on
electromagnetic principles. It consists of a ferrite rod whose magnetic
field changes depending on the state of the line or trunk being scanned.
Signal distributors translate orders received from central control into
high-power, long-duration pulses that are distributed to the appropriate

LINES

•
•
•

TRUNKS

NETWORK

•

LINES AND TRUNKS

SCANNER
ANSWER BUS

CENTRAL
CONTROL

SCANNERS

SIGNAL
DISTRIBUTOR

PERIPHERAL
UNIT BUS

CENTRAL PULSE
DISTRIBUTOR BUS

Figure 10-9. 1/1AESS switching equipment peripheral systems.

Chap. 10

Network Switching Systems

419

relays in trunk, service, and power control circuits in 1/ lAESS switching
equipment. These relays are controlled by polarized signals and are magnetically latched (held in an operated state), thus providing a memory
function in the end device. The signal distributor has 1024 outputs. The
decoding of the order from central control to provide access to one of
these outputs is performed by relay contacts. Relay contacts were
selected because, at the time of original design (around the early 1960s),
no electronic device could economically compete with a contact on a
large relay for such a decoding function when the required access cycle
was about 20 milliseconds. Such was the case for most of the relays in
1/IAESS switching equipment trunk circuits as well. Since then, fully
electronic signal distributors have been designed to replace the original
signal distributor design. Western Electric began to manufacture those
new distributors in 1975.
In electronic switching systems, some control functions must be carried out at electronic speeds or at speeds exceeding the capability of the
magnetic latching relays controlled by the signal distributors. These control functions are provided by a central pulse distributor (CPO). In 1/ lAESS
switching equipment, a diode-transformer gate was chosen as the decoding element in the CPO. Two primary functions of the CPO are the enabling of other peripheral units to receive orders from the central processor and the control of outpulsing over trunks. The transformer provides
a balanced output, so a pulse from such an output point can be transmitted over a twisted pair to remote locations without interference. In addition, bipolar pulses can be easily generated and transmitted. These pulses
can control the operation and release of a relay over a single pair of wires
by using a receiving device (called a bipolar flip-flop) that can recognize
the two polarities.
Line, Trunk, Junctor, and Service Circuits. Individual circuits are
required on a per-call and even per-line basis to match the widely variable outside world to the standardized "inside world" of the central processor. These individual circuits are the line, trunk, and service circuits.
A line circuit is connected to each line. It provides initial battery power to
the line as well as an indication of on-hook or off-hook state to the
scanners and allows the removal of the scan elements during connection
to the network to minimize transmission losses. A trunk circuit is connected to each trunk. It provides power to the associated line after the
connection is established, provides trunk supervision signaling (an indication of the on-hook or off-hook state to the scanners), and isolates the
network reed contacts during switching intervals. Junctor circuits serve in
place of trunk circuits for the supervision of intraoffice calls. Service circuits include circuits that receive dial pulses or signals from TOUCHTONE service, circuits that ring the called line, circuits that provide an
audible ringing indication to the calling line, circuits that provide a busy

420

Network and Customer-Services
Systems

Part 3

tone, and circuits that transmit and receive multifrequency pulses. Each
circuit performs only a few functions under program control, and
different circuits are connected to the communication path as needed via
the switching network.
Maintenance. The 1/ 1AESS switching equipment reliability objective
specifies that system downtime should total no more than 2 hours in 40
years. Because some failures of individual components are bound to
occur over decades of system service, duplication is essential to meet this
objective. Every system unit required to maintain service in 1/ 1AESS
switches is provided in duplicate, and troubles are found and corrected
quickly in order to minimize the possibility of system failure caused by
multiple troubles.
Somewhat less than half of all the stored-program instructions in
l/lAESS switches are used for maintenance. Some of these programs, in
conjunction with logic wired into the hardware, detect and report faults
and troubles. Other programs control routine tests, diagnose troubles,
and control emergency actions to ensure that the system operates satisfactorily, either by eliminating faulty subsystems or by reorganizing usable
subsystems into a new operating combination. When trouble occurs, telephone switching actions are interrupted as briefly as possible to reestablish an operational system. Then, on a less urgent basis, the defective
unit is diagnosed by the system itself, and the results are printed on the
maintenance terminal-a teletypewriter for the lESS system and a
CRT /keyboard terminal 10 for the 1AESS switch.
Services. In addition to local telephone service, the 1/ 1AESS switches
offer a variety of special services. The Custom Calling Services described
in Section 2.4.2 are available to all customers who are served by electronic switching systems. Business customers may select offerings such as
centrex, ESS-ACD, Enhanced Private Switched Communications Service,
or electronic tandem switching. (Chapter 2 discusses these services.)
2ESS and 2BESS Switching Equipment

The 2ESS switching equipment was designed to extend electronic switching economically into the suburban market. Its first application was in
1970 in Oswego, Illinois. The 2ESS switch has a call capacity of 19,000
average busy season busy hour calls,ll with a maximum of 24,000 terminals per system.
10 Teletype Corporation's DATASPEED 40 terminal, described in Section 11.1.3, is used.
11 Section 5.2.3 discusses the average busy season busy hour concept.

Chap. 10

Network Switching Systems

421

Network. The same design principles used in the development of lESS
switching equipment were applied in 2ESS switching equipment. As
in lESS switching equipment, the network in the 2ESS switching equipment uses eight stages of switching. The original network element was
the ferreed switch, later replaced by the remreed switch. The 2ESS
switching equipment network differs from the lESS switch in that lines
and trunks terminate on the same side of the network. This is referred to
as a folded network. There is no need for separate line and trunk link networks as in the lESS switch; however, a junctor circuit for power feed
and supervision must be included in line-to-line connections. The 2ESS
switching equipment allows for either a 2-to-1 or 4-to-1 concentration
ratio.
Processor and Program. The 2ESS switching equipment processor is
duplicated and constructed of transistor-resistor logic. Its memory is
organized similarly to that of lESS switching equipment with ferrite
sheet call stores and magnetic twistor card program stores. Unlike the
lESS switching equipment, 2ESS memory units are associated with one of
the dual processors and cannot be switched between them. To make
stored-program control economically attractive for suburban local central
office applications, the hardware and software design for the 2ESS system
emphasized savings in call store and program store instead of real time.
The need for new and more complex features and the availability of
im proved technology led to the modernization of the 2ESS switching
equipment. This was accomplished by the development of the 3A central
control (3ACC), which replaces the 2ESS switching equipment processor.
The 2BESS switch is a 2ESS switch with a 3ACC as its processor.
Older 2ESS switches can be updated by a processor retrofit procedure
similar to that for 1AESS switching equipment.
By combining integrated circuit design with semiconductor memory
stores, the 3ACC doubles the call capacity originally available in the 2ESS
switch. In addition to this advantage, the 3ACC requires one-fifth the
floor space and one-sixth the power and air-conditioning that the 2ESS
switching equipment central processor requires. Because of its state-ofthe-art processor, newer features (such as automated data linking of
automatic message accounting data) are available only on the 2BESS and
not on the older 2ESS switching equipment.
The 3ACC is a self-checking, microprogram-controlled processor capable of high-speed serial communication. Its average instruction execution time is 1.25 microseconds. Its memories are not separated by function as in lESS, 1AESS, or 2ESS switching equipment. Instead, there is a
single, insulated-gate, field-effect transistor main store, with portions allocated for call data, resident programs, and translation data. The resident
programs are hardware write protected. Nonresident programs, such as

422

Network and Customer-Services
Systems

Part 3

maintenance and recent change, and backup for translations and resident
programs are stored on a tape cartridge.
The first 2BESS switching equipment application was in Acworth,
Georgia, in 1976. That was the same year that the first lA and 3ESS
switching equipment went into service. (The 3ESS switch also uses the
3ACC, as is explained below.)

3ESS Switching Equipment

The 3ESS switching equipment is the smallest Western Electric spacedivision, centralized electronic switching system, serving 2000 to
4500 lines. It was designed to meet the needs of a typical community dial
office, with one busy hour call per line and approximately 65-percent
intraoffice calling. Like the 2BESS switch, 3ESS switching equipment uses
the 3ACC as its processor. As explained in the previous section, the
memory configuration has microcoded instructions stored in read-only
memory. The 3ESS switching equipment network uses remreed technology with a fixed 6-to-l concentration ratio. As in 2/2BESS switching
equipment, it is a folded network, with a maximum of 5760 terminals.
The 3ESS switching equipment uses five switching stages on every
connection.
The 3ESS switch was designed for unattended operation. To accomplish this, extensive maintenance programs are built into the software.
Also, a sophisticated remote maintenance capability connects the Switching Control Center System (see Section 14.3.2) to the 3ESS switch.
One of the unique features of the 3ESS switch is the ability to "hot
slide" at installation. Normally, when replacing an older system, the new
system is assembled beside the old and brought into service while the old
system continues to operate. This requires enough available building
space for two complete switching systems. However, the 3ESS switching
equipment is assembled at the factory, brought to the installation site on
a platform, and put into service outside the building. The older system
can then be taken out of service, removed from the building, and the
new system "slid" into the building. A great savings in construction is
realized. This type of installation is also available on a limited basis with
the 2BESS switching equipment.

No. lOA Remote Switching System
A survey taken in the mid-1970s showed that approximately one-half of
all switching systems in the Bell System served 2000 or fewer lines and
that a large portion of these systems were No. 355A step-by-step systems,
yet no electronic switching system had been developed to serve this

Chap. 10

Network Switching Systems

423

market economically. The principal obstacle to electronic switching system penetration into small offices was the cost of a processor and its associated memory. Remote switching avoids this problem by sharing the
processing capabilities of a nearby electronic switching system and using
a microprocessor for certain control functions under direction of the host
central processor.
Design Philosophy. The No. lOA Remote Switching System (RSS) is
designed to act as an extension of a lESS, lAESS, or 2BESS switching
equipment host and is controlled remotely by the host over a pair of
dedicated data links. Figure 10-10 illustrates the interconnection arrangement used between a No. lOA RSS and an electronic switching system
host. Customer lines are terminated at the RSS in the conventional
manner. The voice pairs are concentrated at the RSS and connected to
the host system over T- or N-carrier facilities. 12 Outgoing and incoming
calls are processed by the host system. Intra-RSS calls are initially set up
through the host, with the final voice connection through the RSS only.
However, if a data link between the host and RSS is severed, the RSS is
capable of stand-alone operation for intraoffice calls. This is an important
feature in the rural market, where as many as 65 percent of the calls
placed are intraoffice calls.
Whether the call is outgoing or intraoffice, the RSS and host system
take the same initial steps. Customer lines are scanned by the RSS for
off-hook. When an off-hook occurs, the RSS asks the host system to set
up a call path. The electronic switching system locates an idle channel in
the T- or N-carrier, an idle digit receiver within itself, and a path from
the receiver to the channel through its network. The RSS then sets up a
path from the customer's line to the channel. The customer's call is now
handled by the electronic switching system as if there were no RSS
involved.
The RSS can handle a maximum of 2048 customer lines. It consists of
termination and subscriber loop interface circuits for each line, signaling
circuits, digit receivers, a data link to the host system, a switching network, carrier equipment including multiplexers, and a WE 8000 microprocessor and memories. The subscriber loop interface circuit feeds
power to its customer line and monitors on-hook and off-hook signals.
This circuit is scanned by the RSS's microprocessor, which is directed by
the control programs contained in erasable, programmable, read-only
memory. The control programs are activated by data link signals from
the host electronic switching system. The RSS also has a short-term
random-access memory to store data needed for call processing.

12 Sections 9.3 and 9.4 discuss carrier systems.

REMOTE TERMINAL

ELECTRONIC SWITCHING SYSTEM HOST

r----------------,
.-----I

CENTRAL
CONTROL

NETWORK

l-

I

I-

I

I
I
I

L

(T- OR N-CARRIER)
VOICE CHANNELS

••
•

I

I

CUSTOMER
LINES

I

-

DATA
LINK
CONTROL

I
1----1

I

I

I

I

L _______________

~

I

I

I

DATA CHANNELS

I
I
I
I
I
I
I

MICROPROCESSOR
CONTROL

NETWORK

I

I

I

I

I

I
CARRIER
FACILITIES

,-----------------.

••
•
I
DATA
LINK

~

.L ________________
I

Figure 10-10. Block diagram of RSS host with T-carrier interconnection links.

~

Chap. 10

Network Switching Systems

425

When the duplicated data link between the RSS and its host fails, the
RSS switches to a stand-alone mode. Only basic service on intra-RSS calls
is allowed during stand-alone operation; custom calling, billing, traffic
measurements, and similar services normally provided by the host are not
available.
When operating normally, the host passes a message to the RSS every
10 seconds over the data link. Upon data-link failure, the RSS waits 30
seconds, then initiates its own call-processing procedures. The RSS has a
copy of its line translations that is updated periodically. Established
intra-RSS calls are maintained when the failure occurs, but calls in the
process of being set up at the time of failure are lost.
The network in an RSS is composed of PNPN solid-state switching
devices that draw little power and are small enough to fit on a standard
circuit pack. However, because of the switch's low voltage tolerance,
high-voltage signals such as ringing are switched through a separate
high-voltage network.
10.3.4 TIME-DIVISION SWITCHING SYSTEMS
4ESS Switching Equipment

Design Objectives. The motivation for designing an electronic tandem
switching system is basically the same as that for a local electronic
switching system: to lower first-cost and operating expenses relative to
an electromechanical design and to provide a flexible system that can be
adapted to changing needs. An obvious approach to this challenge
would have been to use the basic lESS switch control system with the
substitution of 4-wire switching frames and appropriate trunk circuits.
This -approach was studied and abandoned because:
• Rapid growth of toll traffic and the penalties associated with multiple
toll systems in a metropolitan area pointed to the long-range need for
a system of very large capacity. The lESS switch would have required
substantial increases in both processing and network capabilities.
• Since digital (pulse-code modulation 13 ) transmission was predicted to
become dominant in the toll network, it appeared economically attractive to use a digital time-division switching network in the toll
machine.
Other important advantages of a time-division network in a large toll
switching system arise from considerations of the installation costs of
space-division networks and the cost of trunk rearrangements. One factor in both of these costs in space-division systems is the extensive
13 Section 6.4.3 discusses pulse-code modulation.

426

Network and Customer-Services
Systems

Part 3

cabling needed to connect a large number of network frames together
and to many other functional units. With the 4ESS switching equipment,
these costs are reduced through the integrated modular design of toll terminal equipment and switching equipment and by switching multiplexed
signals to reduce the number of interframe conductors required. This
also makes the rapid interconnection of frames using precut connectorized cables more economical. The choice of time-division switching for
the network results in an economically practicable, nearly nonblocking
network that obviates the need for equipment rearrangements to avoid
network congestion.

Features. The 4ESS switching equipment is a large-capacity 4-wire tandem system for trunk-to-trunk interconnection. Initially, it handled
550,000 peak busy hour calls (for a typical call mix using the 1A processor
equipped with core stores), 107,000 terminations, and 1.8 million
hundred-call seconds (CCS) per hour. The 4ESS switching equipment
forms the heart of the stored-program control network that uses
common-channel interoffice signaling, while still supporting multifrequency and dial-pulse signaling. Stored-program control network
features provided by 4ESS switching equipment include the Mass
Announcement System that supports DIAL-IT network communications
service (see Sections 2.5.1 and 11.3.2) and expanded inward Wide Area
Telecommunications Services screening and routing. Further storedprogram control network capabilities will be provided with future generics. The 4ESS switching equipment also provides international gateway
functions.
Capacity. Rapid growth in toll traffic had created situations in metropolitan areas in which two or more of the available toll switching systems
were needed to provide adequate capacity. When multiple toll offices
serve the same area, trunking penalties result. In view of these situations
and the projected continuing growth of toll traffic, an important objective
in development of 4ESS switching equipment was to achieve a substantial
increase in capacity.
The design of 4ESS switching equipment grew out of the lESS switching equipment concept in that it uses a high-speed processor as central
control to handle the most complex aspects of call completion. Signal
processors are provided for preprocessing the more elementary tasks, thus
decreasing the per-call usage of the main processor. Through its use of
core memories and higher speed logic, the 4ESS switch main processor,
the 1A processor, is about five times as fast as the lESS switching equipment processor.
A large-scale integrated circuit semiconductor memory was developed
to replace the 1A processor core memories and deployed in 1977. It provided a 3-to-1 size reduction and a 6-to-1 reduction in power consumption and was easier to maintain. Subsequent improvements in technology

Chap. 10

Network Switching Systems

427

have resulted in further reductions in size and power consumption.
Increased speed of operation has resulted in up to 3D-percent increases in
the call-carrying capacities of the 1AESS and 4ESS switches. These
improvements began to appear in the field in 1979.

Software Structure. The 4ESS switching equipment software structure is
based on a centralized development process using three languages: a
low-level assembly language, an intermediate language called EPL,14 and
a high-level language called EPLX. The high call-processing capacity
of 4ESS switching equipment is sustained through the judicious application of these languages. Real-time functions such as call processing are
generally programmed in the assembler language, while measurements
and administrative functions frequently are programmed in EPL. Some
of the maintenance programs and audits that are not frequently run are
programmed in EPLX.
The 4ESS switching equipment has an executive control loop operating
system. The loop consists of short segments of low-priority callprocessing tasks. High-priority call-processing tasks may be interjected at
the end of any low-priority segment. This mode of operation contrasts
with interrupt-driven schedulers found in the earlier systems. The 4ESS
system executive control avoids the bookkeeping overhead and much of
the memory-writing conflicts between high- and low-priority tasks.
In order to meet the stringent electronic switching system reliability
objectives, highly defensive coding and an extensive audits structure have
been used. Structured programming techniques have been and will continue to be used for development of the 4ESS switching equipment
software generics.
Administration and Maintenance. Based on experience with the lESS
switching equipment, greater emphasis was given to easing the operation
and administration of the 4ESS switching equipment. For example, CRT
input/output terminals permit an interactive mode of communication
between human and machine. Low-cost, large-capacity magnetic disk
memories (see Figure 10-11) store within the system data that formerly
were kept as paper records. This permits activities such as assigning a
trunk to a terminal to be done by the system rather than manually. The
chosen network design maintains low blocking, even with undistributed
loads; consequently, physical wiring changes traditionally associated
with the periodic redistribution of trunks over terminals (required to
intermix lightly and heavily loaded trunks) have been eliminated. Magnetic tapes controlled by the Auxiliary Data System provide for the output of traffic data, performance data, and billing records, as well as for
the input of new program and translation information.

14 Electronic Switching System Programming Language.

I

DIGITAL CARRIER

ANALOG CARRIER

DIGITAL
:
INTERFACE I--~-..,
FRAME
I

I

I

____ .J
DIGITAL CARRIER

ANALOG CARRIER

WIRE TRUNK

CCIS CHANNEL

MAGNETIC
TAPES

DISK
MEMORIES

Figure 10-11. 4ESS switching equipment block diagram.

Up to six 4ESS switches can be remotely administered and maintained
from centralized work centers. Remote trunk testing, data-base administration, and switching maintenance mean that very few functions must be
performed on site.

Time-Division Switching. Because of the rapid growth of digital
transmission systems in the network, the switching network in 4ESS
switching equipment has been designed specifically to pass pulse-co demodulated (peM) signals without conversion. This pays off on interfaces
with the toll connecting facilities, where short-haul (up to 50-mile) peM
systems are already in extensive use. Short-haul intertoll trunks (up to
500 miles) using peM are also in service and are expected to grow
rapidly. Long-haul intertoll trunks are still predominantly analog, but
long-haul digital transmission systems are now being enthusiastically
introduced.
For analog transmission systems, the 4ESS switching equipment design
provides for conversion of analog signals to digital form and vice versa.
428

Chap. 10

Network Switching Systems

429

Figure 10-11 shows these arrangements. The equipment below the dotted
line indicates the original 4ESS switching equipment configuration. The
triangular shape of the converter, known as a voiceband interface unit
(VIU), is intended to signify the multiplexing of 120 voice channels onto
one digital path. Connections to trunks on PCM facilities do not need
conversion and, therefore, use a much simpler interface, the digroup terminal. (Digroup is a contraction of digital group; in present systems, a digroup
consists of 24 voice channels.) The digroup terminal (DT) multiplexes
digital groups to put 120 voice channels onto one conductor and separates
out supervisory signaling information. Trunks interface via the metallic
facility terminal frame before conversion to PCM. Beyond these interface
units, all signals are in PCM format and can be switched as required in
the time-slot interchange (TSI)/time-multiplexed switch (TMS) complex
under control of circulating solid-state memories.
Starting in 1981, the voice interface units have been replaced by the
digital interface frame (DIF) shown above the dotted line in Figure 10-11.
Digroups can terminate directly on the DIF, eliminating the need for
digroup terminals, while analog carriers go through an LIT connector
that converts from an analog (L3, L4, L5) to a digital carrier (T1). Metallic
facilities go through a channel bank that converts to digital carrier. The
DIF converts five digroups into one digital stream of 120 voice channels
for input to the TSI.
Figure 10-12 shows the TSI-TMS complex in simplified form. The TSI
associated with an incoming trunk stores the incoming coded PCM
sample until the time slot selected for cross-office transmission of this call
comes up. When it does, the TMS is configured by the circulating
memory to provide a path to the TSI associated with the appropriate outgoing VIU, DT, or DIF. The coded PCM sample is thus sent through TSIs
from the incoming VIU, DT, or DIF to a storage register in the TSI associated with the outgoing VIU, DT, or DIF, where it is held until the time
slot corresponding to the appropriate outgoing trunk comes up. A coded
sample is transferred from the outgoing trunk to the incoming trunk in
the same way in the same time slot. The cross-office time slot is selected
independently of the time slots corresponding to the incoming and outgoing trunks. The space-division portion of the switch (the TMS) is
reconfigured for each time slot (about 106 times per second), in contrast to
a conventional space-division switch that is reconfigured only in response
to the arrival and departure of calls. (Section 7.3.3 discusses time-division
switching further.)
Each TSI has 8 input and 8 output coaxial leads connecting to the
TMS. Figure 10-12 shows only input leads from TSI 1 and output leads to
TSI 128. The 1024 input leads (8 from each of 128 TSIs) and 1024 output
leads are interconnected by the TMS, which is a 1024-input-to-1024output matrix. Since each telephone connection is made up of two paths
through the TMS, one for each direction of transmission, the maximum

120 TIME SLOTS

*\

/

\
DT, VIU,
OR DIF

105 TIME SLOTSt

/

4

1

1

•••

TSI
1

••
•
8

7

"

INPUT

---+
TMS
1024 x 1024
MATRIX

1

1

••

•
7

DIF
DT
TMS
TSI
VIU

TSI
128

•
.OUTPUT..-•
8

DIGITAL INTERFACE FRAME
DIG ROUP TERMINAL
TIME-MULTIPLEXED SWITCH
TIME-SLOT INTERCHANGE
VOICEBAND INTERFACE UNIT

• Actually 128 time slots. Eight reserved for maintenance, leaving 120 for
telephone connections.
tActually 128 time slots. On the average, only 105 can be occupied because
of the 7 -to-8 expansion in the TSI.

Figure 10-12. Diagram of TSI-TMS complex.

number of simultaneous telephone connections that can be accommodated may be found by multiplying the number of time slots on each TSI
lead by the number of possible connections through the TMS and dividing the answer by 2:
105 x

1024
= 53,760 connections.
2

Floor-Space Savings. An important part of an operating telephone
company's initial cost for switching is the building to house the equipment. The 4ESS switching equipment is very compact compared to a
No. 4A Crossbar; the initial 4ESS system occupies about one-fifth the floor
space for equivalent systems serving 10,000 to 20,000 trunks. This comparison includes all equipment and work space and assumes small
crossbar switches in the No.4 Crossbar. Subsequent designs of 4ESS systems accomplished significant further reductions in size; recent designs
occupy about one-fourth the floor space of the initial 4ESS system.
DMS-IO Switching System
The DMS-10 system is a digital time-division switching system for local
offices. It was designed for use in community dial office applications.
The system has a 13,000-per-hour peak call capacity; line size ranges from
430

Chap. 10

Network Switching Systems

431

200 to 6000 subscriber lines. Its first standard application in the Bell System was in March 1981. Flexibility is emphasized through modularity in
software and hardware architecture. Hardware equipment uses printed
circuit packs that plug into printed circuit backplanes for easy and rapid
replacement and growth. The DMS-10 system is designed for service in
an unattended rural office. Figure 10-13 is a functional block diagram of
the DMS-10 system.

Processor Complex. The system architecture is organized into system
control, the switching network, and peripheral equipment. System control is provided by a fully duplicated central processing unit (CPU). The
two processors operate in active / standby mode with automatic switchover
control. Two cartridge tape units also provide system backup. The primary tape unit stores complete system software and office data; the secondary tape contains overlay software consisting of infrequently used programs. System memory for program store, call store, and data store is
subdivided into 32K-word blocks, each block packaged as one printed circuit pack. Memory for the Custom Calling Services resides on a separate
32K-word module. For reliability, n+l redundancy is used, whereby one
spare memory block is provided for every n (n~8) blocks of memory. A
duplicated high-speed control bus interconnects the CPU with memories,
input/output devices, and network modules.
Network Complex. The switching network is composed of one to four
network groups, each serving approximately 2500 lines and associated
trunks and service circuits. Intergroup traffic is carried by digital junctors. The network also provides service circuits for tone generation, dialpulse and multifrequency transmission, conference calls, terminal
input/output, various t~st functions, and scanning of the peripheral
equipment (PE). The PE modules provide the hardware interface to
subscriber lines, trunks, and service and maintenance circuits. The
analog-to-digital and digital-to-analog conversions are provided in the PE
by one channel codec (coder-decoder) for each individual subscriber-line
interface. Custom large-scale integration logic performs the time multiplexing of speech signals from any 1 of 112 analog terminals of two PE
modules to anyone of thirty channels of a digital multiplex (MUX) loop.
MUX loops provide thirty PCM voice channels and two signaling channels to carry the digitized speech and signaling information from the PE
to the network. Sparing of MUX loops occurs at the PE end; one MUX
loop serves two PE shelves, with the capability to handle the load on four
shelves should its "mate" fail.
Digital trunks terminate on a digital carrier module (OCM) in the
DMS-10. OCMs convert the internal 2.048-megabits-per-second (Mbps)
transmission format of the MUX loops to the 1.544-Mbps format of standard digital signal level 1 (OSl) interfaces, along with the time-slot
switching from thirty to twenty-four channels. There is one dedicated

MUX LOOPS

ANALOG LINES
AND TRUNKS

T1 CARRIER

REM LINES

30+2

30+2

CPU
DCM
MUX
OCM
PE
RCM
REM
R/WM

CENTRAL PROCESSOR UNIT
DIGITAL CARRIER MODULE
MULTIPLEX
OFFICE CARRIER MODULE
PERIPHERAL EaUIPMENT
REMOTE CARRIER MODULE
REMOTE EaUIPMENT MODULE
READ/WRITE MEMORY
BACK-UP LINKS

Figure 10-13. DM8-10 functional block diagram.

Chap. 10

Network Switching Systems

433

MUX loop per OCM, each OCM accommodating up to twenty-four digital
trunks.
Use of a remote equipment module (REM) provides the capability to
extend the internal MUX loop functions to PE modules located up to 70
miles away from the central office. Two MUX loops, terminated on an
office carrier module (OCM) at the central office, are interconnected by a
dedicated OSl-type carrier to the two MUX loops from the remote carrier
module (RCM) at the remote location. Thus, the REM handles traffic
from four PE modules corresponding to two paired MUX loops, or 224
subscriber lines and analog trunks. Since the same peripheral equipment
is being used, the same central office physical environment is needed at
the REM location.
The DMS-l digital remote concentrator serves as a pair-gain multiplexer when operating with the DMS-IO switching system. Traffic for up
to 252 customer lines can be concentrated in two Tl lines over a distance
of up to 70 miles to an integrated interface in the DMS-IO switching network. That interface eliminates the need to reconvert and re-expand the
signals and terminal appearances at the central office.

Software Structure. The DMS-IO software and firmware can be viewed
as hierarchical layers surrounding the CPU, with inner layer functions
"hidden" from outer layers. The inner layer closest to the CPU includes
firmware for handling CPU interrupt messages, the bootstrap loader for
starting up the system from scratch, run-time-sensitive routines, and
automatic fault detection and recovery software. The next layer, stored in
random-access memory, contains the scheduler for arranging the
machine's working schedule. The outer layer has the software modules
for terminal input/output, overlay program handling, and call processing.
The software for DMS-IO systems is written in SL-1,15 a high-level
language, similar to ALGOL,16 developed at Bell Northern Research, Ltd.
Scanning for state changes is done in two levels by hardware logic circuits. Subscriber-line state changes are noted by peripheral control packs
that are scanned by network signaling packs before sending interrupt
messages to the CPU. Input messages to the CPU are "time-stamped" and
queued in call store for processing. The CPU processes the state change
interrupts and is not involved with a call during the stable, or talking,
state. Call processing in the DMS-IO system is said to be an event-driven
system, similar to that used in the 2BESS.
The DMS-IO is flexible in allowing the operating telephone company
to set up translations and routing as desired. A full range of residential

15 Trademark of Northern Telecom, Ltd.
16 Developed by a committee of the Association of Computing Machinery in conjunction
with their counterparts in Great Britain, France, Germany, and the Netherlands.

434

Network and Customer-Services
Systems

Part 3

and small business features can be provided, and billing capability can be
provided either locally or remotely. Interfaces for operations systems are
being developed for traffic, maintenance, and memory administration.
5ESS Switching Equipment

The 5ESS system is a digital time-division electronic switching system
designed for modular growth to accommodate local offices ranging from
1000 to 100,000 lines. It was designed to replace remaining electromechanical switching systems in rural, suburban, and urban areas
economically. A major design goal was to use equipment modularity to
achieve an economically competitive system over this wide range of line
sizes. The 5ESS switching equipment uses distributed processing and
modular software and hardware to provide a flexible architecture and
simplify the addition of new features.
The 5ESS switching equipment architecture is shown in Figure 10-14
and consists of a number of interface modules (lMs) connected via a
time-multiplexed switch. A duplex administrative module processor provides centralized routing control and administrative maintenance
features. It is connected to the TMS through a message switch for communication with the interface modules. The initial application, put into
service in March 1982, in Seneca, Illinois, offered features similar to 3ESS
switching equipment and was limited to a single interface module.
Succeeding issues of the generic program, cut over in 1983, offer the
multi module configuration and local/toll features for combined class 4
and class 5 operation. A remote switching module (the No. 5A Remote
Switching Module) is scheduled for 1984. A fully integrated interface,
also scheduled for 1984, will enable a SLC-96 carrier system to terminate
economically and directly on electronic switching equipment.
The 5ESS system's time-space-time network consists of time-slot interchanges in each interface module that are connected via two optical fiber
network control and timing links to the solid-state TMS. These links
operate at the rate of 32 Mbps and are used for messages between the
administrative module and interface modules and among interface
modules as well as for voice transmission.
Administrative Module. The duplex administrative module processor of
the 5ESS switching equipment consists of two 3B 1720 computers, each
equipped with disk storage and input/output processors (lOPs). The
duplex processors operate in active / standby modes and rely on hardware
and software sanity checks to switch modes. The lOPs provide interfaces
(such as teletypewriters) to technicians, data links to operations systems,

17 Trademark of Western Electric Co.

DATA LINKS FOR
OPERATIONS SYSTEMS
INTERFACE MODULE

--,

SYNCHRONIZATION
UNIT 0
UNIT I

MODULE
CONTROL
AND TSI
(DUPLEX)

/
LINES
AND
TRUNKS

r------""I

LINES

--+

5A
REMOTE
SWITCHING
MODULE

~
T-CARRIER

I

r---,-,-

INTERFACE
MODULE

---1- -

INPUT/
OUTPUT
PROCESSOR
(DUPLEX)

I
I

I
I

--1- - - ,

I

NETWORK
CONTROL AND
TIMING LINKS

I

~
TTY/CRT

L _____ ,

I

I
I

MESSAGE
SWITCH **
(DUPLEX)

•

••

ADMINISTRATIVE
MODULE
PROCESSOR
(DUPLEX)

INTERFACE
MODULE

L _______________

~

L ____________

COMMUNICATIONS MODULE

*OMITTED IN SINGLE-MODULE CONFIGURATION
SIMPLIFIED IN SINGLE-MODULE CONFIGURATION

* * SUBSTANTIALLY

Figure 10-14. 5ESS switching equipment system architecture.

ADMtNISTRA TIVE MODULE

I
I

I
I
...JI

436

Network and Customer-Services
Systems

Part 3

and administrative module processor scan-and-distribute circuits via a
number of peripheral controllers. Control messages and data pass
between the administrative processor and the interface modules via the
communications module. The administrative module processor also communicates with operations systems for traffic, billing, and maintenance
data 18 through the lOPs.
Communications Module. The communications module consists of a
message switch and a time-multiplexed switch. In addition to interfacing
the administrative processors to the TMS, the message switch terminates
special data links such as the one for the synchronization carrier (normally from a 4ESS switch) used to maintain digital synchronism. The
TMS is used to connect voice channels in one interface module to voice
channels in other interface modules as well as for data messages between
the administrative modules and interface modules and for data messages
between interface modules.
Interface Modules. Each interface module has a duplex microprocessorbased module controller; a module processor, used for call processing;
and a duplex 512-slot TSI, used for time-division switching. In addition,
there are several interface units. One, called a digital services unit, is used
to generate call-progress tones and interfaces with the TSI. Other interface units terminate lines, trunks, and Tl carrier facilities. Once digitized
at the interface unit, each 8-bit PCM voice sample, either received from
Tl or converted by the interface module from an analog input, has eight
bits of control information added to it. The resulting 16-bit word can be
switched via the TSI (for intramodule calls) or sent via the network control and timing links to the TMS (for intermodule calls). Data interfaces
at each TSI and control interfaces at each module controller terminate
serial data and control busses from the interface units. A duplex digital
signal processor in the interface module receives dial pulses and
processes busy / idle bits to relieve the module controller of that realtime-intensive task.
Each interface module can host analog line / trunk units, digital
line / trunk units, digital carrier line units, digital service circuit units, or
metallic service units in addition to miscellaneous test and access units.
These units are tied via busses to the module controller and (with the
exception of the metallic service unit) to the TSI. The exact mixture of
units in a module depends on the traffic and customers being served and
can be different in each interface module. The interface module can also

18 Respectively, the Engineering and Administrative Data Acquisition System (EADAS)
discussed in Section 14.3.1, the Automatic Message Accounting Recording Center
(AMARC) discussed in Section 10.5.5, and the Switching Control Center System (SCCS)
discussed in section 14.3.2.

Chap. 10

Network Switching Systems

437

be used to terminate the T-carrier connections between a remote switching module and its host office. The digital line/trunk unit terminates the
digital Tl carrier. The digital carrier line unit terminates SLC-96 carrier
systems directly.
Both space division and time division are used for economical termination of subscriber lines. The analog line unit includes a concentrator
available in 4-to-1, 6-to-l, and 8-to-1 ratios. Its solid-state gated diode
crosspoints can pass high-level signals such as ringing and battery so that
separate crosspoints are not necessary. SLC-96 carrier systems using the
integrated T-carrier feature and terminating on a digital carrier line unit
can be concentrated digitally in the digital carrier line unit. Trunks are
unconcentrated. The metallic service unit contains a metallic test access
network, some high-level service circuit functions, and scan-anddistribute points.
No. SA Remote Switching Module. Remote switching is made available
via the No. SA Remote Switching Module (RSM). The RSM can be
located up to 100 miles from the SESS switching equipment and can terminate a maximum of 4000 lines with a single interface module. Several
No. SA RSMs can be interconnected to serve remote offices as large as
16,000 lines.
The No. SA RSM is a standard SESS system interface module with
software augmented to provide stand-alone switching capability in .the
event of a failure in the host-remote link. A No. SA RSM is linked to a
host SESS system via T1 carrier or other facilties such as lightguide. The
T1 carrier facilities terminate on a digital line/trunk unit on the SESS system host. Using the same basic hardware and software modules in the
SA RSM and SESS switching equipment ensures compatibility and
reduces development effort. It also provides for a structural modularity
that allows a smooth transition from a remote to a full SESS system or
vice versa.
An important feature of the RSM is direct trunking. Most remote
switches (for example, the No. lOA Remote Switching System) require
that all interoffice calls pass through the host switch. The SA RSM can be
equipped with trunk units that directly interface facilities to other offices.
This avoids costly trunk rearrangements when remote switches are
deployed.
Software Structure. The SESS switching equipment software is divided
into two segments. The portion in the administrative module processor is
responsible for officewide functions such as the human interfaces, routing, charging, feature translations, switch maintenance, and data storage
and backup. The portion in the interface module is responsible for the
standard call-processing functions associated with the lines and trunks

NetwOrK ana LUStOmer-~ervlces
Systems

438

Part 3

terminating on that interface module. These functions represent about
80 percent of the processing on calls and include interpreting status and
address digits; controlling the interface unit, analog concentrator, and TSI
path hunts; and signaling. Most software is written in C, a programming
language developed by Bell Laboratories, and has a modular structure to
afford easy expansion and maintenance. The administrative module processor and the interface module processors share a common software
environment composed of a common data-base management system and
an operating system for distributed switching. This facilitates software
portability.
The 5ESS switching equipment was designed for flexibility. Processing power is added with the addition of each interface module. Modular
software and modern data-base management concepts allow the software
to be easily extended and modified to accommodate new services. Likewise, the modular hardware allows units to be upgraded to take advantage of new technology.

10.4 OPERATOR SYSTEMS
The same technology used in switching systems has been applied to
operator-services equipment. The goal of this modernization is to reduce
the number of operators required by automating the more routine tasks
and by using operators more efficiently. Two examples of such modernized systems are described below -one for toll services and one for
number services.

10.4.1

TOLL SERVICE

Before the introduction of mechanization, toll operators worked at 3CL
switchboard positions, pictured in Figure 10-15. Many tasks involved in
operator-handled calls were done manually, including switching (connecting circuits with cords), timing the call duration, and billing.
Mechanization of these tasks leaves the operator free to perform the personal functions of speaking with customers, gathering needed information, and entering it into the system by pressing the appropriate keys.
The first major mechanization of the toll operator's functions was the
development of the traffic service position (TSP) in 1965 as part of the
crossbar tandem switching system. The operator's position was a cordless
console with a numerical display and pushbuttons. In order to make the
mechanization of operator-services capabilities independent of the
designs of present and future toll and tandem switching systems, a new
electronic Traffic Service Position System (TSPS) was developed and first
introduced in 1969. The TSPS is an autonomous system-it stands apart

Figure 10-15. Manual toll switchboards.

from both the local and toll offices. Because the signaling and transmission interfaces for TSPS are standard, it functions with all the various
designs of local and tandem switching systems.
The system is able to handle the types of calls shown in Table 10-1. It
can also handle guest-originated calls from hotel rooms and provide th e
hotel with an automatic, immediate teletypewriter printout or a timely
operator voice report of the charges for these calls.
For customer-dialed station-to-sta tion calls, TSPS can serve as a centralized automatic message accounting (CAMA) point to record billing
details without operator intervention. TSPS operators also provide assistance on customer-dialed international calls, and when local offices are
not modified for international dialing, customers can place calls through
TSPS on a dial-zero basis (that is, the customer dials only zero; the TSPS
operator keys in the international number).
The TSPS uses both the basic hardware components and the system
structure of the lESS swi tch. At first, the real-time capacity of the storedprogram control (SPC) processor used in TSPS was approximately 16,000
initial position seizures per busy hour . The maximum number of major
system elements are:
• three thousand trunks
• 310 operator positions, accessible as a single team
• eight chief operator groups (local, remote, or both)
• 62 operator positions per chief operator group.
439

Network and Customer-Services
Systems

440

Part 3

TABLE 10-1
TSPS OPERATOR FUNCTIONS
Type of Call
TSPS Operator
Functions

From Coin
Stations

From N oncoin
Stations

1+,0+

0+

Identifying called customer
on person-to-person calls

0+

0+

Obtaining acceptance of
charges on collect calls

0+

0+

Identifying calling number*

1+,0+

1+,0+

Monitoring coin deposits

1+,0+

Obtaining billing information
for calling card or thirdn urn ber calls

Handling operator assistance
calls
Type of call (as it appears on
TSPS console)

0-

0-

1+ = Customer-dialed station-tostation calls
0+ = Customer-dialed special calls

o = Operator assistance

calls

+ Indicates that more digits are to be dialed, that is, the called telephone number.
- Indicates that no more digits are to be dialed.
* Needed only when calling number is not automatically identified and forwarded from the
local office.
As a result of automating parts of the operator's job and centralizing
several small teams into one larger, more efficient team, TSPS operation
requires many fewer operators than the cord switchboard.
Figure 10-16 shows a typical TSPS operating room. Each console con~
tains two positions in a desk-like arrangement. A position becomes available to receive calls when an operator plugs the headset into its jack.
Calls are automatically distributed to all attended positions, so all operators receive an equal share of the load. When a position is given a call,
the operator hears a distinct tone and is given a lamp display to indicate

Figure 10-16. A typical TSPS operating room.

whether the originating station is coin or noncoin and whether the customer has dialed zero alone or zero followed by seven or ten digits or if
the call requires only operator number identification for billing purposes.
With these indications, the operator is able to respond appropriately. In
particular, on calls received from coin stations, the initial deposit and
duration of the initial period are indicated in the numerical display .
Whenever a call is connected to a TSPS position, all call details are
available from the system memory. The details are directly equivalent to
those that would be written on a ticket if the call were handled at a cord
switchboard. Under key control, the calling number, the number that is
being called, a calling card number if keyed into the system, the number
of a third telephone if one is being billed, or the charging rate on coin
calls can be displayed. Other operator controls allow the operator to
release connections forward or backward, to ring the stations forward or
backward, to collect or return coin deposits, and to connect to specialservices operators over outgoing trunks .
As shown in Figure 10-17, all TSPS trunks have two 2-wire appearances on the TSPS link network. The network connects to various service
circuits: digit receivers, outpulsers, coin control circuits, tone circuits, and
operator positions. The basic logic instructions for handling calls are in
the memory and are executed by the SPC processor. Changes in the state
of trunk, service, and other peripheral circuits, including the positions
themselves, are detected by scanners together with programs and memory

441

TSPS
TRUNK
CIRCUIT

2-WIRE TRUNKS

VOICE
DATA

}

VDO~~!-

{

LINK

4---.t

LOCAL OR
REMOTE
POSITION
SUBSYSTEM

TSPS
CONSOLE

SIGNAL
DISTRIBUTOR

CENTRAL
PULSE
DISTRIBUTOR

SCAN

SPC 1A
PROCESSOR

MAINTENANCE
ACCESS
TELETYPEWRITER,
ETC_

MEMORY

MAGNETIC TAPE
UNITS
CALL RECORDING
PROGRAM LOADING

SPC 1A

Figure 10-17. Traffic Service Position System.

indicating previous states. Output instructions via signal distributors and
central pulse distributors control these circuits and the position lamps.
The processor-memory complex-including such support units as the
control and display panel, a signal distributor, a central pulse distributor,
a master scanner, the maintenance teletypewriter, and the program tape
unit for loading and unloading memory-constitutes a subsystem called
the SPC No. lA. The switching network (link network) used in TSPS to
connect the trunks to the service circuits and positions is a 4-stage, 2wire, space-division network using ferreeds.
The program structure for TSPS closely follows the lESS switch program structure. An executive control program, interrupt levels, priority

442

Chap. 10

Network Switching Systems

443

work lists, fault-recognition programs, and diagnostic programs are used
to provide the real-time characteristics of the system.
Since its initial cutover in 1969, many enhancements have been made
to TSPS. For example:
• The remote trunk arrangement extended TSPS service to remote areas
that could otherwise not be served economically by TSPS.
• The more recent Automated Coin Toll Service (ACTS) automates
operator functions on coin toll calls. The Station Signaling and
Announcement Subsystem required for ACTS synthesizes requests for
deposits, counts the deposits, and sets up and times the call, thereby
eliminating operator intervention from these typ~s of calls in most
cases. By taking over these routine tasks, ACTS frees operators to concentrate on more complex calls, such as collect or third-party billing,
and to help any customers requiring assistance.
• Automated Calling Card Service eliminates operator intervention on
calling card, collect, and bill-to-third-number calls by allowing customers to key in billing information on these calls.
• The SPC No. 1A processor is being replaced with the SPC No. 1B processor, which uses the 3B20 computer control unit (the 3B20 computer
is also used in SESS switching equipment); this has increased the realtime capacity of TSPS by 60 percent. By the mid-1980s, close to 100
percent of the telephones served by Bell operating companies will be
served by TSPS.

10.4.2 NUMBER SERVICES
The Automatic Intercept System (AIS), which became available in 1970, was
developed to automate and centralize the processing of calls to nonworking numbers. The various types of calls that may be routed to AIS
include calls to vacant or unassigned numbers, calls to changed numbers,
calls to disconnected numbers, and trouble intercepts.
Incoming intercept trunks from local offices are connected to equipment that synthesizes recorded announcements specifically tailored to
each intercept call. For those relatively few nonroutine cases requiring
operator assistance, AIS provides an improved method of operator handling. As such, AIS totally eliminates operator intervention on most intercept calls. This, together with the centralization of many small intercept
bureaus into one AIS, has resulted in operator savings of about 75
percent.
As shown in Figure 10-18, an AIS contains one or more Automatic
Intercept Centers (AICs). There is one centralized intercept bureau associated with a home AIC in an AIS. An AIC contains a time-division

AIC

FROM
LOCAL
CENTRAL
OFFICES

CENTRALIZED
INTERCEPT
BUREAU
AIC

[>
1

AIC
ONI

OPERATOR ASSISTANCE,
TRAFFIC, OR OPERATOR
NUMBER IDENTIFICATION
AT LOCAL AIC

G>
2

OPERATOR
INQUIRY
TRUNKS

b
3

OPTIONAL OPERATOR
NUMBER IDENTIFICATION
AT LOCAL AIC

AUTOMATIC INTERCEPT CENTER
OPERATOR NUMBER IDENTIFICATION

Figure 10-18. Automatic Intercept System.

switching network, a stored-program processor, magnetic disk memories
for filing unassigned numbers, service circuits, announcement machines,
and the centralized intercept bureau. The network connects to a maximum of 512 intercept trunks, announcement circuits, and service circuits
and can accommodate as many as sixty-four connections simultaneously.
When a call comes into the AlC from the local office, it is connected to
a multifrequency receiver that receives the called digits and passes them
to the processor. The processor decides whether to connect the call to a
centralized intercept bureau operator or to look up the called number in
the files.
For situations in which local offices are not equipped to identify the
called number on intercept calls, operator number identification is provided at the AlC.
The files of unassigned numbers are stored in duplicated magnetic
disk memories for reliability. The processor uses this information to compose an announcement. Recorded words and phrases, including
numbers, are supplied by a duplicated 96-track announcement machine.
The output of each track can be connected to customer lines through the
switching network.
444

Chap. 10

Network Switching Systems

445

The locations of recorded numbers, words, and phrases are stored in
the administrative processor, which selects the items required for an
announcement in the proper sequence for the particular call. The administrative processor is the same as that developed for 2ESS switching equipment; the program structure and maintenance strategy closely follow
those of the 2ESS switch.
The intercept file represents a large data base that requires frequent
updating. A typical installation serving a metropolitan area (New York
has several) or an entire state, such as North Carolina, may have as many
as 500,000 records and may require as many as 18,000 changes per day. A
minicomputer-based Data Base Administration System provides for
efficient updating via teletypewriter or data-link channels, as well as an
off-premises backup of the entire file.
Recent enhancements to AIS provide for automation of parts of some
of the calls that previously required operator handling. The network
capability has been expanded to a total of 1024 terminations and 256
simultaneous connections. By the mid-1980s, approximately 85 percent of
the operating company telephones will be served by AIS.

10.5 BILLING EQUIPMENT AND SYSTEMS
10.5.1 INTRODUCTION
In 1980, the Bell operating companies handled, on the average, over
500 million calls a day. Under current tariffs, billing data is not required
for every call placed since approximately 70 percent are covered under
the provision of a monthly flat-rate charge. Even on these calls, however,
the switching system must make a decision not to record call details for
billing.
The billing information required for the remaining 30 percent
depends on the type of call. About 10 percent of the total calls originated
are classified as toll. These require the most billing information, including time of answer, call duration, and calling and called numbers. The
remaining 20 percent are measured local calls. These require a range of
billing information from full details to an abbreviated form with no
called number or call duration, depending on local tariffs.
Current trends indicate that the percentage of calls requiring some
form of per-call measurement will continue to increase, reaching close to
90 percent by 1990. At the same time, the volume of total calls placed is
rising at about 3 percent per year. Under these conditions, automatic
message accounting (AMA) equipment will be called on to record substantially increasing amounts of data in the next decade. (See Figure
10-19.) In addition, since most of the additional calls to be measured are
local calls, the average revenue per call recorded will drop. The need to

500

Ui
z

400

0

TOTAL
ORIGINATING
CALLS

:i
...J

§.
(/)
...J
...J

300

······················1·····
r-- -- ---------~----

CONTROL
UNIT

DEMODULATOR

RECEIVER

• • 

......

->

I

UNIT 0

I DS120 LINK

I:

~
UNIT 1

OTHER
TlME·SlOT .......
INTERCHANGE <......
PORTS

TIME-SLOT
INTERCHANGE FRAME

~

lc: -

:::>

SERVING
SWITCHING AND
PERMUTING CIRCUIT

DEDICATED
SWITCHING AND
PERMUTING CIRCUIT

SERVICE UNITS

Figure 11-31. Configuration of the Mass Announcement System.

-

-...

... >

I

.....
....

...

.....
>
L

___

_...I

Chap. 11

Customer-Services Equipment
and Systems

515

The two units contain identical recordings, thereby increasing reliability
in case of failure of one of the units.
In the 4ESS electronic switch, terminations for announcements are
provided by units known as dedicated time-slot interchanges (TSIs), which
are discussed in Section 7.3.3. A dedicated TSI can connect any calling
customer who reaches it with the announcement corresponding to the
number that the caller has dialed. A dedicated TSI for MAS can terminate up to 896 calls simultaneously. Announcements can be associated
with terminations in a completely flexible manner, permitting dynamic
allocation of capacity within the 896 limit. Two dedicated TSI frames,
one associated with each MAS unit, are required for each 4ESS switch
equipped for MAS. This minimum equipment configuration provides
announcement capability to 1792 callers. Additional dedicated units can
be added to increase capacity.
Figure 11-32 shows a simplified event flow for a call using the MAS
capability. A more detailed description is as follows:
• The caller dials either a 7- or 9-digit announcement number such as a
DIAL-IT service number and is routed to a 4ESS switch equipped for
MAS. The 4ESS switch recognizes the number as a request for a particular MAS announcement. It also recognizes whether call counting
is associated with the number. Next, it determines whether cutthrough service is in effect and, if it is, whether the call should be cut
through or connected to the announcement.
• The 4ESS switch determines which dedicated TSI the call should be
connected to in order to' minimize the audible ring interval. If all terminations are busy on the first choice, the call will be connected to
the second-choice dedicated TSI or, if no announcement terminations
are available, to a busy tone. If the MAS equipment is out of service
so no MAS announcement can be given, the call is connected to a nocircuit announcement from the office announcement machine.
• Finally, for successful calls, audible ring ends when the 4ESS switch
connects the call to the announcement over the same dedicated TSI
that provided the audible ring. Answer supervision then begins
(unless specified otherwise). If the incoming trunk has centralized
automatic message accounting (CAMA), CAMA billing is initiated (see
Section 10.5.4).
• The customer can abandon the call at any time during the announcement. In the absence of early abandonment, the call is automatically
removed from its MAS termination when the allocated playing time
elapses.
The Mass Announcement System is an optional feature of the 4ESS
switch. Offices with 4ESS switches equipped for MAS are strategically

MAS
CUSTOMER
CALL

4 ESS TRANSLATES DIALED DIGITS TO
A REQUEST TO HEAR A PARTICULAR
ANNOUNCEMENT
IF CALL IS TO BE
CUT THROUGH

4 ESS OUTPULSES
CUT-THROUGH
DOD NUMBER

CALLER HEARS AUDIBLE RING
(OPTIONAL)

4 ESS SENDS ANSWER

CALLER HEARS MAS ANNOUNCEMENT

AT END OF PLAYING TIME,
CALLER IS DISCONNECTED FROM
MAS ANNOUNCEMENT

Figure 11-32. Simplified event flow for a MAS call.

located throughout the country (see Figure 11-33) so that sponsors can
provide local, regional, or national service. Each of these offices is designated an MAS node and its associated calling region an MAS island.

11.4 MOBILE TELEPHONE SYSTEMS
The demand for telephone service to and from moving or temporarily
stationary users continues to grow rapidly; much more than a convenience, this service can increase the efficiency of today's mobile society.
Of the more than 100 million vehicles in the United States, about 10 million are equipped with 2-way radios (not counting citizens-band radios);
however, in 1980, only about 100,000 could connect to the telephone network. Historically, the bulk of radio communication has been in the

516

ISLAND

MAS NODE

•

ATLANTA

S

CHICAGO (INCLUDES ALASKA)

F~:;!lcl

DALLAS

L::yy:t

DENVER

III

LOS ANGELES (INCLUDES HAWAII)

[))::;tl

NEWARK

ITIIIII1

PHILADELPHIA

Figure 11-33. National deployment of nodes and associated islands
at initiation of Mass Announcement System in 1980.

private, fleet-oriented area. Future growth is expected to shift toward the
mobile telephone sector, which will be split about equally between radio
common carriers and wire-line carriers (for example, the operating telephone companies). In addition to radio-equipped vehicles, service to
hand-held portable units is becoming increasingly popular as the state of
the art permits smaller, lighter designs.
The following sections review existing systems and examine newer
systems now being deployed. The major emphasis will be on systems for
land mobile (the largest market), although systems for airplanes, trains,
and watercraft serve a small but important need.
The radio spectrum is a limited (but renewable) resource. Thus, the
availability of radio channels is a fundamental concern for any mobile
system. The FCC is responsible for making channels available to the
public based on need. It sets standards (bandwidth, channel spacing,
modulation methods, power, etc.) according to the state of the art.

517

l''IIetwOrK

518

ana LUStOmer-::>ervlces
Systems

Part 3

11.4.1 LAND MOBILE TELEPHONE SYSTEMS
Channel Availability
Mobile telephone service began in the late 1940s. By the seventies, it
included a total of thirty-three 2-way channels below 500 megahertz
(MHz), as shown in Table 11-2. The 35-MHz band, which is not well
suited to mobile service (because of propagation anomalies), is not
heavily used. The other bands are fully utilized in the larger cities. In
spite of this, the combination of few available channels per city and large
demand has led to excessive blocking. The FCC's recent allocation of 666
channels at 850 MHz for use by cellular systems (described below) should
change this situation. This allocation is split equally between wire-line
and radio common carriers (each is allocated 333 channels). In many
areas, the wire-line carrier will be the local operating company.
Use of conventional systems on the new channels would increase the
traffic-handling capacity by a factor of about 10. The cellular approach,
however, will increase the capacity by a factor of 100 or more. How this
increase is achieved is discussed later in this section. The potential for
very efficient use of so valuable and limited a resource as the frequency
spectrum was a persuasive factor in the FCC's decision.

TABLE 11-2
MOBILE TELEPHONE ALLOCATIONS

Band*
(MHz)

Number of
Wire-Line
Common-Carrier
Channels

Channel
acing
SRkHz)

35
150
450
850

10
11
12
333

40
30
25
30

;(- This is the nominal frequency that identifies the band. The frequencies
allocated in the 850-MHz band, for example, are between 825 and 845
MHz and 870 and 890 MHz.

Transmission Considerations
Radio propagation over smooth earth can be described by an inverse
power law; that is, the received signal varies as an inverse power of the
distance. Unlike fixed radio systems (for example, broadcast television or
the microwave systems described in Chapter 9), however, transmission to

Chap. 11

Customer-Services Equipment
and Systems

519

or from a moving user is subject to large, unpredictable, sometimes rapid
fluctuations of both amplitude and phase caused by
This impairment is caused by hills, buildings, dense
forests, etc. It is reciprocal, affecting land-to-mobile and mobile-toland transmission alike, and changes only slowly over tens of feet.

• Shadowing -

Because the transmitted signal may travel over
multiple paths of differing loss and length, the received signal in
mobile communications varies rapidly in both amplitude and phase as
the multiple signals reinforce or cancel one another. 25

• Multipath interference -

Other vehicles, electric power transmission, industrial processing, etc., create broadband noise that impairs the channel, especially at 150 MHz and below.

• Noise -

Because of these effects, radio channels can be used reliably to communicate at distances of only about 20 miles, and the same channel (frequency) cannot be reused for another talking path less than 75 miles
away except by careful planning and design.
In a typical land-based radio system at 150 or 450 MHz, one channel
comprises a single frequency-modulation (FM) transmitter with 50- to
250-watt output power, plus one or more receivers with 0.3- to 0.526
microvolt sensitivity. This equipment is coupled by receiver selection
and voice-processing circuitry into a control terminal that connects one or
more of these channels to the telephone network (see Figure 11-34). The
control terminal is housed in a local switching office. The radio equipment is housed near the mast and antenna, which are often on very tall
buildings or a nearby hilltop.

Conventional System Operation
Originally, all mobile telephone systems operated manually, much as
most private radio systems do today. A few of these early systems are
still in use, but because they are obsolete, they will not be discussed here.
More recent systems (the MJ system at 150 MHz and the MK system at
450 MHz) provide automatic dial operation. Control equipment at the
central office continually chooses an idle channel (if there is one) among
the locally equipped complement of channels and marks it with an "idle"
tone. All idle mobiles scan these channels and lock onto the one marked
with the idle tone. All incoming and outgoing calls are then routed over
this channel. Signaling in both directions uses slow-speed audio tone
25 Section 6.3.6 discusses multi path fading, another term for interference.
26 One of several receivers is selected for best reception from the mobile unit as it moves
through the area.

LOCAL
OFFICE

TO OTHER
} LOCAL OFFICES

TO
OTHER
TOLL
OFFICES

,./'

"....--------

~

/

STANDARD
SWITCHING

...............
OMNIDIRECTIONAL
COVERAGE
AREA
(20-25 MILE RADIUS)
'\

EQUIPMENT

\

~4'VER\
+ ~MOBILE

.A

I
I

TRANSMITTER

FM TRANSMITTER
(ONE EACH FOR 1
TO 12 CHANNELS)

(ONE OR MORE
PER TRANSMITTER)
.............

---

/

/

RECEIVER

""'-

\

/

/

/

/'"

--./'"

Figure 11-34. A manual, or MJ/MK, mobile telephone system.

pulses for user identification and for dialing. Compatibility with manual
mobile units is maintained in many areas served by the automatic systems
by providing mobile-service operators. Conversely, MJ and MK mobile
units can operate in manual areas using manual procedures.
One desirable feature of a mobile telephone system is the ability to
"roam"; that is, subscribers must be able to call and be called in cities
other than their home areas. The numbering plan must be compatible
with the North American numbering plan (see Section 4.3). Further, for
land-originated calls, a routing plan must allow calls to be forwarded to
the current location. In the MJ system, operators do this. Because of the
availability of the MJ system to subscribers requiring the roam feature,
the MK system need not be arranged for roaming.

520

Chap. 11

Customer-Services Equipment
and Systems

521

Advanced Mobile Phone Service
Cellular Concept. Although the MJ and MK automatic systems offer
some major improvements in call handling, the basic problems-few
channels and the inefficient use of available channels-still limit the
traffic capacity of these conventionally designed systems. Advanced
Mobile Phone Service27 overcomes these problems by using a novel cellular approach. It operates on frequencies in the 825- to 845-MHz and 870to 890-MHz bands recently made available by the FCC. The large
number of channels available in the new bands has made the cellular
approach practical.
A cellular plan differs from a conventional one in that the planned
reuse of channels makes interference, in addition to signal coverage, a
primary concern of the designer. Quality calculations must take the statistical properties of interference into account, and the control plan must
be robust enough to perform reliably in the face of interference. By placing base stations in a more or less regular grid (spacing them uniformly),
the area to be served is partitioned into many roughly hexagonal cells,
which are packed together to cover the region completely. Cell size is
based on the traffic density expected in the area and can range from 1 to
10 miles in radius.
Up to fifty channels are assigned to each cell to achieve their regular
reuse and to control interference between adjacent cells. This is illustrated in Figure 11-35, where cell A' can use the same channels as cell A.
Because of the inverse power law of propagation, the spatial separation
between cells A and A' can be made large enough to ensure statistically
that a signal-to-interference ratio greater than or equal to 17 dB is maintained over 90 percent of the area. Maintenance of this ratio ensures that
a majority of users will rate the service quality good or better.
Cellular systems also differ from conventional systems in two
significant ways:
• High transmitted power and very tall antennas are not required.
• Wide FM deviation is permissible without causing significant levels of
interference from adjacent channels.
The latter is responsible for the high voice quality and high signaling
reliability of the Advanced Mobile Phone Service.
In any given area, both the size of the cells and the distance between
cells using the same group of channels determine the efficiency with

27 Advanced Mobile Phone Service (sometimes called AMPS) is a generic name referring to the

cellular system concepts and the control algorithms used for mobile service in the United
States in the 850-MHz band.

LAND·BASED
SUBSCRIBER

TO

O~~~~:S

(TOLL AND LOCAL)

t

---

I
I

/
I
/
"
"

DIRECTIONAL
PATTERN
OF ONE
ANTENNA

MOBILE
TELECOMMUNICATIONS
SWITCHING OFFICE
_ . - TRUNKS TO RADIOS
UP TO 50 PER CELL
SITE (NOT SHOWN TO
ALL CELL SITES)
-

-

-

CONTROL PATHS

Figure 11-35. Advanced Mobile Phone Service system plan.

which frequencies can be reused. When a system is newly installed in an
area (when large cells are serving only a few customers), frequency reuse
is unnecessary. Later, as the service grows (a dense system will have
many small cells and many customers), a given channel in a large city
could be serving customers in twenty or more nonadjacent cells simultaneously. The cellular plan permits staged growth. To progress from
the early to the more mature configuration over a period of years, new
cell sites can be added halfway between existing cell sites in stages. Such
a combination of newer, smaller cells and original, larger cells is shown
in Figure 11-36.
One cellular system is the Western Electric AUTOPLEX-100. In this
system, a mobile or portable unit in a given cell transmits to and receives
from a cell site, or base station, on a channel assigned to that cell. In a
mature system, these cell sites are located at alternate corners of each of
the hexagonal cells as shown in Figure 11-36. Directional antennas at
each cell site point toward the centers of the cells, and each site is connected by standard land transmission facilities to a 1AESS switching system and system controller equipped for Advanced Mobile Phone Service
522

Figure 11-36. Staged growth of a Western Electric AUTOPLEX cellular
radio system. (AUTOPLEX is a trademark of Western Electric.) Cell
splitting, by adding new cell sites in areas of high demand, creates more
traffic capacity.

operation (called a mobile telecommunications switching office, or
MTSO). Start-up and small-city systems use a somewhat more conventional configuration with a single cell site at the center of each cell.
The efficient use of frequencies that results from the cellular approach
permits Advanced Mobile Phone Service customers to enjoy a level of
service almost unknown with present mobile telephone service. Grades
of service of P(O.02)28 are anticipated, compared to today's all-too-common
P(O.5) or worse. At the same time, the number of customers in a large
city can be increased from a maximum of about one thousand for a conventional system to several hundred thousand. Also, because of the
stored-program control capability of MTSOs equipped with the lAESS
system, Custom Calling Services and many other features can be offered,
some unique to mobile service. Other, smaller switches provided by
Western Electric or other vendors are also available to serve smaller cities
and towns.
System Operation. Unlike the MJ and MK systems, Advanced Mobile
Phone Service dedicates a special subset of the 333 allocated channels
solely to signaling and control. Each mobile or portable unit is equipped
with a frequency synthesizer (to generate any of the 333 channels) and a

28 As described in Chapter 5, this means a 2-percent probability of blocking.

523

524

l"erworK ana Lustomer-~ervlces
Systems

Part 3

high-speed modem (10 kbps). When idle, a mobile unit chooses the
"best" control channel to listen to (by measuring signal strength) and
reads the high-speed messages coming over this channel. The messages
include the identities of called mobiles, local general control information,
channel assignments for active mobiles, and "filler" words to maintain
synchronism. These data are made highly redundant to combat multipath interference. A user is alerted to an incoming call when the mobile
unit recognizes its identity code in the data message. From the user's
standpoint, calls are initiated and received as they would be from any
business or residence telephone.
As a mobile unit engaged in a call moves away from a cell site and its
signal weakens, the MTSO will automatically instruct it to tune to a
different frequency-one assigned to the newly entered cell. This is
called handoff. The MTSO determines when handoff should occur by
analyzing measurements of radio signal strength made by the present
controlling cell site and by its neighbors. The returning instructions for
handoff sent during a call must use the voice channel. The data regarding the new channel are sent rapidly (in about 50 milliseconds), and the
entire retuning process takes only about 300 milliseconds.
In addition to channel assignment, other MTSO functions include
maintaining a list of busy (that is, off-hook) mobile units and paging
mobile units for which incoming calls are intended.
Regulatory Picture. The FCC intends cellular service to be regulated by
competition, with two competing system providers in each large city: a
wire-line carrier and a radio common carrier. To prevent any possible
cross-subsidization or favoritism, the Bell operating companies must offer
their cellular service through separate subsidiaries. These subsidiaries
will be chiefly providers of service and, in fact, are currently barred from
leasing or selling mobile or portable equipment. Such equipment will be
sold by nonaffiliated enterprises or by American Bell Inc.
11.4.2 PAGING

Another radio service offering of the Bell System is a I-way communication system that provides users with a personal number that identifies a
signaling or alerting receiver, the BELLBOY radio paging set, that can
conveniently be carried about. The Bell System originally developed this
service during the 1960s. During the 1970s, radio common carriers also
introduced this type of service under other trade names. Today the Bell
System has a relatively small portion of this market, although it is a major
market presence in many cities.
This service requires the mobile user to carry a small (3- to 4-ounce)
pocket receiver. The fixed portion of the paging system comprises several

Customer-Services Equipment
and Systems

Chap. 11

525

transmitters and a control terminal. The person (usually a secretary)
doing the paging uses TOUCH-TONE signaling to dial the number of the
control terminal and then dials more digits to identify the specific user.
The transmitters signal the user, who calls the secretary via some convenient land telephone to retrieve the message. In an urban area, several
high-power transmitters are required to provide a signal strong enough
to penetrate buildings and to activate the receiver's small (and hence,
inefficient) antenna. Figure 11-37 is a schematic picture of the system.
All the new paging systems use high-speed digital signaling and can
accommodate as many as 100,000 users on a single radio channel. The
FCC recently allocated 120 channels at 930 MHz for paging-related services. This action is expected to spur rapid innovation and expansion of
this type of service. Some Bell operating companies are already offering a
service that uses a paging receiver with a visual display. A calling party
equipped with TOUCH-TONE service can, via end-to-end signaling, furnish as many as ten digits to the paging controller. In this way, the telephone number of the calling party or some other message with a prearranged meaning can be displayed to the user. Other services either
already available or under study include voice paging (the caller speaks
to the subscriber) and nationwide paging (subscribers may be called in
many cities across the country).

LOCAL/TOLL
NETWORK

URBAN " AREA
"BOUNDARY
"'---

__

~

Figure 11-37. BELLBOY radio paging system plan.

\...u:)(umt!r-;:>ervlces
Systems

'''''''YVV.lA cUlU

526

Part 3

11.5 VISUAL SYSTEMS
At the present time, the Bell System's principal involvement with television is through the cross-country distribution of television signals (see
Section 6.1.3) for the broadcast industry, a private-line service (see Section 2.5.2). However, the Bell System has long been involved with video
telephony, or interactive television, in which two or more parties see as
well as talk to each other during the connection. The requirements for
broadcast and interactive television differ and so do their systems. This
section discusses interactive rather than broadcast television.

11.5.1 EARLY TELEVISION
Long-distance video telephony was first publicly demonstrated on April
7, 1927, when images were sent between Washington, D. C., and Bell
Laboratories in New York City by wire and between an experimental
station in Whippany, New Jersey, and New York City by radio. The
images were crude by present standards: 50-line resolution, eighteen
frames per second, at a bandwidth of 20 kHz. Two different receivers
were designed for two different purposes. One produced a small image,
approximately 2 by 2-% inches, suitable for viewing by a single person,
for use in private conversations. The other receiver produced a larger
image, apprOXimately 2 by 2-V2 feet, intended for viewing by an audience
of some size.
The success of this demonstration led, by 1930, to a 2-way experimental system in Manhattan, connecting AT&T headquarters at 195 Broadway
with Bell Laboratories at 463 West Street. Further developments
improved both equipment and image quality and required more
bandwidth.

11.5.2 PICTUREPHONE VISUAL TELEPHONE SERVICE
Throughout its market trial from 1964 to 1975 (see Section 2.5.6), the
PICTUREPHONE visual telephone service operated to the standards in
Table 11-3. Also shown are the National Television Standards Committee
(NTSC) standards for commercial television. The NTSC standards yield a
resolution of about 140,000 picture elements, which is comparable to 16millimeter movie film. Responding to the high cost of transmission
bandwidth and to a reduced need for resolution in face-to-face video connections, the video telephone provided about one-fourth the resolution of
the NTSC standard. The repetition rate (pictures displayed per second)
was the same as the NTSC's. The aspect ratio (width-to-height ratio) was,
however, quite different and better suited for use by individuals and for
materials such as letters.

bqUlpment
and Systems

Customer-~ervices

Chap. 11

527

TABLE 11-3
NTSC BROADCAST TELEVISION STANDARDS
VERSUS BELL SYSTEM PICTUREPHONE STANDARDS

Characteristic

NTSC
Standards

PICTUREPHONE
Standards
(Early 1970s)

Bandwidth (MHz)
Frames per second
Raster interlace
Lines per frame
Aspect ratio

4
30
2:1
525
4:3

1
30
2:1
267
11:10

Design of the equipment for the PICTUREPHONE visual telephone
service was quite sophisticated. The compact unit contained both the
camera tube and a 9-inch black-and-white picture tube. Among its
features were an electronic zoom capability and automatic adjustment for
light levels. Since the unit carried both picture and sound information,
the camera required shock mounting to eliminate acoustic coupling from
the microphone.
Ordinary telephone wires and newly developed broadband amplifiers
were used for local transmission. Transmit and receive were on separate
wire pairs. For longer distances, broadcast television channels could be
used; a bandwidth of only 1 MHz was needed.
11.5.3 VIDEO TELECONFERENCING
In 1975, PICTUREPHONE meeting service was introduced on a trial basis.
Rooms were set up to accommodate conferences of various sizes and were
equipped with standard large monitors and several cameras. The NTSC
standards were adopted.
Between 1975 and 1981, as part of the market trial, various room
configurations were tested: both "public" rooms, eventually located in
twelve cities, and private rooms on the premises of three customers.
Information was gathered on such matters as customer acclimation to
video technology, network size, business segmentation of potential users,
adaptation to user needs, and exchange versus intercity offerings. A standard service is now offered under tariffs effective July 2, 1982.
Long-distance digital transmission at rates of 1.5 and 3 megabits per
second (Mbps) is provided by the High -Speed Switched Digital Service
(HSSDS). Connection from meeting rooms to HSSDS is provided by
High-Capacity Terrestrial Digital Service (HCTDS, described in Section 2.5.4). The video coder/decoder (codec), a principal component of

528

Systems

Part 3

the meeting room equipment, reduces the transmission rate required for
"conference grade" video to the range of 1.5 to 3 Mbps from the approximately 100 Mbps needed for broadcast quality video. It does this by digitally processing the meeting room video to remove redundancy, which is
more prevalent in conference video than in some types of broadcast
video.
At present, plans for PICTUREPHONE meeting service include public
rooms in many cities, complemented by many more private rooms on customers' premises. Generally, a conference room contains a table for six
people (see Figure 2-11). Each of three cameras is aimed at two of the
participants. The system employs voice camera switching: The camera
pointing at the speaker is the one in use at that moment. A fourth camera, the "overview," shows the whole conference room and is used, for
example, when the local participants are listening to a speaker in the
other conference room.
Three additional cameras are provided for graphics: one for images of
sheets of paper or transparencies, one for slides, and one for a speaker
standing at an easel or chalkboard. The last is a multipurpose camera,
equipped with controllable pan, tilt, and zoom. All cameras and monitors have color capability. A preview monitor is provided to set up the
graphics or slides or to aim the multipurpose camera properly before the
images are transmitted to the other room. A video cassette recorder is
also provided to record the meeting.
Room design for video teleconferencing represents a challenge in
balancing light distribution and sound quality. Enough light must fall on
the participants to provide a good camera image, but if too much falls on
the monitors, the image contrast is degraded. The PICTUREPHONE meeting service audio system provides control of the line from one end or the
other to avoid echoes. Voice transmission is effectively permitted in only
one direction at any time by "squelching" the signal from the other direction by inserting loss in the signal path. In video teleconferencing, however, the amount of loss added has been reduced from that used in the
4A speakerphone to give a better feeling of interaction.
Many human factors experiments have been conducted at Bell Laboratories to evaluate various designs. Based on these evaluations and the
information gained from the market trial, video teleconferencing should
be both effective and pleasant.

11.5.4 RESEARCH ACTIVITIES
Research has produced two concepts of great interest. One is motion compensation (see Netravali and Robbins 1979), in which the picture processor
algorithm attempts to predict the movement of a picture element from
one frame to another. Computer simulations show that the concept is

Chap. 11

Customer-Services Equipment
and Systems

529

technically feasible and would reduce the bit rate required by a factor of
2 or more.
The other concept is continuous presence (see Larsen and Brown 1980).
In one form, it replaces voice switching of the cameras with multiple
cameras and multiple monitors, with two of the conference participants
on each of the three camera-monitor pairs. Research has found that the
processed signals for the three images can be statistically multiplexed,29
so that the three images can be transmitted at less than three times the
individual bit rates.
More work is needed on these concepts to determine if their benefits
are worth their development and equipment costs and, if so, to reduce
them to practice.

11.6 DATA COMMUNICATIONS SYSTEMS
The rapid growth of data communications in terms of both volume of
data and diversity of application has resulted in the development of a
variety of systems to meet customers' needs. Customers' data service
requirements can be broadly classified as either analog or digital. Analog
data transmission is important, but it represents a smaller volume than
digital services. Included in the analog category are much of the current
facsimile market and signals used in alarm and sensor-based systems.
Digital data can be further categorized by the transfer rates required as
either low, medium, or high speed.
• Digital data transmitted at low speeds, generally at 300 bps or less,
include telemetry, terminal-to-terminal message service, and terminalto-computer services such as computer time sharing.
• Medium-speed transfer, generally from 300 bps through 9.6 kbps, can
be accommodated on voiceband telephone facilities. Typical applications include terminal-to-computer transfers where the terminal may
be a CRT or remote job-entry terminal. Credit checking or banking
applications may operate in batch mode or real-time mode.
Computer-to-computer transfer is also common in distributed dataprocessing systems. Other common applications are multiplexing of
lower-speed data signals and digital facsimile.
• High-speed data are generally those transmitted at speeds above
9.6 kbps, for example, at 19.2 kbps, 56 kbps, and higher. Common
applications at these speeds include computer-to-computer operation
and multiplexing of lower-speed data.
29 Section 6.5 discusses multiplexing.

530

Network and customer-Services
Systems

Part 3

Telephone-type facilities are still by far the most popular means of
data transmission. These facilities include both the PSTN and leased
private lines and private switched networks. Data sets (see Section 11.1.2)
are required to condition the customer's data signal for transmission over
these facilities. DATAPHONE II data sets also give private-line customers
powerful diagnostic capabilities and control of their data communications
systems.
A number of other transport services, some of which are described in
Chapter 2, are available now or soon will be. The remainder of this
chapter discusses the equipment and systems that support these services.
The Digital Data System, which is described in Section 11.6.1, provides end-to-end digital connectivity over dedicated facilities and supports
DATAPHONE digital service. The preponderance of data communications, such as those between two digital computers or between computers
and terminals, involves data that are inherently digital. With end-to-end
digital connectivity, the data maintain their digital form throughout the
transfer, and the efficiencies and cost savings discussed in Section 9.1.3
are obtained.
Other capabilities offering end-to-end digital connectivity are emerging. One of these, circuit-switched digital capability (CSOC), is described
in Section 2.5.1. It will provide a 56-kbps end-to-end digital channel as
an evolutionary capability of the PSTN. With the proliferation of digital
facilities in the nationwide network, the integrated services digital network currently being planned will represent a further stage of evolution
towards a network capable of meeting a wide range of telecommunications needs.
Section 11.6.2 describes packet-switching systems, and in particular
the No.1 PSS packet switch. Packet transport networks can serve those
applications where bursts of data are to be transmitted with very short
delays.
Finally, the DATAPHONE Select-a-station service (see Section 2.5.4) is
an example of a capability designed to meet the needs of a fairly specialized segment of the data communications market. Section 11.6.3 provides
a description of the system that provides this service.
11.6.1 DIGITAL DATA SYSTEM
The dedicated digital transport network used to provide and systematically support DATAPHONE digital service (described in Section 2.5.4) is
called the Digital Data System (00S).30
A typical point-to-point DDS channel is illustrated in Figure 11-38.
Customers are connected to the DDS through a local office. Calls to
30 Caution should be exercised in applying the abbreviation DDS. It must not be applied to
the service because DATAPHONE digital service is a registered service mark.

r -------,

r - - ----,
HUB
OFFICE

LONG-HAUL
FACILITIES

HUB
OFFICE

1----::.--

ACCESS LINES
FOR
MAINTENANCE
AND
SURVEILLANCE

SHORT-HAUL
FACILITIES

LOCAL
OFFICE

LOCAL
OFFICE

4-WIRE LOOPS

CUSTOMER
PREMISES

CUSTOMER
PREMISES

SERVING AREA 1
L DIGITAL
_____
---1

SERVING AREA 2
L DIGITAL
__
_ _ _ ..J

Figure 11-38. Typical point-to-point Digital Data System channel.

another metropolitan area, or digital serving area, are connected through
hub offices. The location of ODS equipment and the stages of signal
multiplexing are shown in Figure 11-39.
The channel service unit (CSU) is the first equipment unit on the network side of the customer interface. An additional signal-processing
function can be provided by the customer or by common-carrier equipment to process control and information signals synchronously into a format compatible with the CSU interface. Using digital format, the CSU
connects a customer's data communications equipment over two cable
pairs to a local office. There, the line is terminated at an office channel unit
(OCU) that regenerates the signal and prepares it for transmission
through the multiplexing hierarchy 31 as outlined below. The CSU also
provides the ability to test a DDS channel quickly and decisively up to
the point of interface with the customer.
In the first stage of multiplexing, a number of customer data rates can
be combined by a subrate multiplexer to form a basic 64-kbps (digital signal level 0 [OSO]) channel (see Table 11-4). A second stage of multiplexing combines up to twenty-four OSO signals to form a 1.544-Mbps stream
that corresponds to the OS1 signal of the time-division multiplex hierarchy. The OS1 signal is usually carried to the hub office over short-haul
transmission facilities. Where the data traffic is very heavy, further
31 Section 9.4.3 describes the digital time-division multiplex hierarchy.

531

O~~~E

TO OTHER
HUB OFFICE ..
1 TO 56 DS1 SIGNALS

rl

- -M:L;;P~X-;R; F~R- - -'1
L _ _LONG-HAUL
_ _ _ _FACILITIES
_____ J

LOCAL OFFICE

r- -1

-D-S1-SI-G-NA-L - (1.544

-- -

..L-___

-

----,

MbPS)~_ _

I

I
I

I
I

\
I

64-kbps
SIGNAL

TO OTHER
SUBRATE DATA MULTIPLEXERS
OR 56-kbps USERS

I
I

I
I

I
\~

L_

____

~

_ _ _ _- J I

I
I

~~R~RS_ _ _ _ _ ~

--- -- -

INTERFACE

CUSTOMER DATA
TERMINAL
EQUIPMENT

I

CUSTOMER
PREMISES

I

Figure 11-39. Block diagram of the Digital Data System.

multiplexing for short-haul facilities, which provide multiple OS1 signals
(see Table 11-5), may be required.
Generally, the office chosen to be the hub will be one that serves a
large number of data customers. This hub office also provides test access
to individual data channels, cross-connecting facilities for efficient packing of customer data signals into various outgoing transmission facilities,
532

Customer-Services Equipment
and Systems

Chap. 11

533

TABLE 11-4
SUBRATE MULTIPLEXER CAPABILITY
Bit Rate from CSU
(kbps)

Maximum Number of
Customer Signals

2.4

20

4.8
9.6
56.0

10
5
1

NOTE: Output is a 64-kbps DSO signal, including byte-stuffing, framing,
and control bits, where applicable.

and a highly stable timing source (derived from a master system clock)
for the multiplexers at both this office and local offices. In turn, these
local offices provide system clock information to individual station units
on the customer's premises.
Most of the long-haul interhub transmission capacity at a DSI rate has
been derived from TO and TH microwave radio with data under voice
(DUV) on the lA Radio Digital System. 32 The rapid expansion of DDS in
the 1980s, however, has exhausted most of the available facility capacity
for DUV in many of the intercity routes. Table 11-5 summarizes a
number of facility technologies available to provide long-haul and shorthaul digital connectivity for DDS.
Arrangements are available to handle special cases. For example, a
DSl channel dedicated to DDS may not be economical in areas with a
small number of customers. The percent fill on the channel may be too
low. To meet this need, alternative arrangements of multiplexing equipment are available to mix digital data and voice customers. The most
recent arrangement uses dataport units that plug into a 03 or D4 channel
bank, in place of a voice channel unit, to derive a DSO transmission channel. In a second instance, digital connectivity to the customer's geographic area may not exist. To defer the construction of new routes,
arrangements are available that use analog facilities with the appropriate
'\ data sets to link these customer locations to the DDS network.
The interhub facilities are monitored full time on an in-service basis.
For each DDS interhub link, one of the twenty-four DSO channels is used
to provide monitoring, to transmit results to a centralized data base in
Chicago, and to remote alarms to appropriate centers for maintenance
action. Therefore, in each intercity DSI channel, only twenty-three DSO
channels are available to transmit customer data.
32 Sections 9.3 and 9.4 describe these systems.

Network and Customer-Services
Systems

534

Part 3

TABLE 11-5
DIGITAL DATA SYSTEM NETWORK
REPRESENTATIVE DIGITAL FACILITIES
Yield in
DSls

Facility
Long-Haul

DUV on TO, TH radio*
DOV on L4 coaxial cable
DOV on L5 coaxial cable
DM12 on TO radio
DOM on L5
FT3C (etc.)t
Digital mastergroup on radio, coaxial
cable, etc. (future)
DIM on AR6A
P140 on L4 coaxial cablet
DR6
TD45A

1
2

4
12
2

56
4
2

84
56
28

Short Haul

T1
T1C,Dt
T10S
T2t
T4Mt
6- or 11-GHz radio at 45 and 90 Mbps
2-GHz radio

1
2

1
4
168
28,56
4

*With message load.
tMust be equipped with appropriate multiplexers.
DUV
Data under voice
DOV
Data over voice
DOM
Data on mastergroup
DIM
Data in the middle

Centralized testing, monitoring, and maintenance are the keys to
achieving high-quality service for DATAPHONE digital service. Centralized test centers are the designated points of contact for customers. Test
center personnel verify and sectionalize troubles while the customer is on
the line. As seen in Section 13.3.3, which deals with network operations,
this procedure is not generally followed in network maintenance,

Chap. 11

Customer-Services Equipment
and Systems

535

although it may be found in maintenance of customer loops. The procedure is an important factor in achieving guaranteed performance. A
variety of operations centers and operations systems are involved in
ensuring high-quality performance and rapid restoration of service.
A number of emerging services and capabilities such as circuitswitched digital capability and Basic Packet-Switching Service will use
DDS network elements for connectivity for initial growth, expansion into
new geographic areas, and/or data transport on an ongoing basis.
11.6.2 PACKET-SWITCHING SYSTEMS
In many data communications applications, data occur in bursts separated
by idle periods, and the average data rate may be much lower than the
peak rate. This type of "bursty" data can often be transmitted more
economically by assembling the data into packets and interspersing packets from several channels on one physical communication path. A header
is added to each packet to identify it. The contents of the header depend
on the system used, but in general, the header must at least indicate the
call of which if is a part and where it fits into the sequence of packets in
a call.
A network can be formed to interconnect a number of users of packet
switching. A packet switch sorts packets coming in on one circuit and
switches them out to another circuit according to the header information
in each packet. Potential savings from using a packet-switching network
rather than direct connection of users are similar to those described in
Section 3.1. The switch network trades the cost of the packet switch(es)
against the reduced cost of transmission facilities and equipment for
interfacing with the facilities. The packet-switching network becomes
more economical as the number of user nodes and the distances between
them increase.
Any type of traffic that has a sufficient peak-to-average information
transfer rate is a candidate for packet switching. In the future, even voice
messages may be assembled into packets. The immediate application of
packet switching, however, is in data communications-primarily in the
interconnections among computers. The first Bell System service
designed for packet switching is the Basic Packet-Switching Service
(BPSS) described in Section 2.5.4.
The No.1 PSS Packet Switch
The packet-switching network supporting BPSS will use the No.1 PSS
packet switch. The architecture of the packet switch reflects the service
objectives: high reliability and responsiveness at high capacity. One
important characteristic of the architecture is that it is based or centered

Network and Customer-Services
Systems

536

Part 3

upon a Western Electric 3B200 computer. The 3B200 computer is
designed to be out of service no more than 2 hours in 40 years. The reliability of this system is achieved by redundancy in hardware (duplex processor and disk, for example) and by diligence in software (the bulk of
the operating system is devoted to maintaining the integrity of the system). All centralized functions associated with providing a packetswitching system are performed on this computer. For example, one
function, routing, associates a physical path through the switch with lines
to customers or trunks to other packet switches. Other functions include
billing, traffic measurement and reporting, and system maintenance.
The capacity of the system to switch packets responsively is a function
of the processing capacity available for packet switching. The architecture of the No.1 PSS packet switch provides duplex processors as shown
in Figure 11-40. Access lines to customers and trunks to other switches
are connected to facility interface processors (FIPs), microprocessors especially designed for the system. The FIPs are connected to the duplex central processor (OCP). Technicians interact with the switch through CRT
terminals connected to the OCP.

DUPLEX
CENTRAL
PROCESSOR
DISK STORAGE

MAINTENANCE
INTERFACE

FACILITY
INTERFACE
PROCESSOR

•••

...

FACILITY
INTERFACE
PROCESSOR

•••
LINES AND TRUNKS

Figure 11-40. Architecture of No.1 PSS packet switch.

An important characteristic of a packet-switching system is the protocol it implements. The No.1 PSS packet switch implements the X.2Sthe packet-switching protocol 33 agreed to by the international standards
organization of the Comite Consultatif International Telegraphique et
Telephonique (CCITT).
The X.2S protocol defines standards for three levels of communication
between the data terminating equipment (the customer's terminal equipment) and the data communications equipment (the packet switch).
33 Section 8.8 discusses data communications protocols and X.2S.

Chap. 11

Customer-Services Equipment
and Systems

537

• Level 1 (physical level) defines the electrical interface and is implemented in hardware.
• Level 2 (link level) processing for both trunks and access lines is done
by firmware (code permanently placed in read-only memory) and
special-purpose integrated circuits that are part of the FIP. The
microprocessors that control the level 2 processing for each line or
trunk operate asynchronously with respect to the level 3 processing
on the FIP.
• Level 3 (packet level) processing for multiple lines and/ or trunks is
done by FIPs. The actual number supported depends on the traffic.
The microprocessor chosen for the FIP does all the processing on data
packets; packets associated with setting up calls are also processed by
the oep, where there is enough memory to store the routing tables.
The call setup rate of the No.1 PSS packet switch is determined by
the processing power of the 3B200 computer and is roughly 200 call setups per second. The capacity to handle data packets is primarily determined by the number of FIPs used and is about 1200 packets per second.
To achieve responsiveness for the large number of logical channels
(simultaneous calls) that are supported on each line, the FIPs must have
large amounts of memory to buffer packets. The 3B200 computer must
also have large amounts of memory to store the routing information
necessary to describe potential networks of users.
A characteristic that the system shares with other highly reliable systems is the amount of software devoted to maintaining the system. Of
slightly more than one megabyte of object code for No.1 PSS, only
200 kilobytes are transport (that is, X.25-related) functions. Administration and maintenance programs account for the other 80 percent.
An essential contributor to the high availability of the packet switch is
the recovery strategy implemented by the software. When a failure is
detected, the smallest portion of the system necessary for restoring service is reinitialized. The goal of the strategy is to minimize the effects of
system recovery on customers. For the No.1 PSS, levels of reinitialization result in:
1)

clearing the call associated with the logical channel that was active
w hen the error occurred

2)

clearing all calls on the line that was active when the error occurred

3)

clearing all calls on the switch

4)

at the highest No. 1 PSS level, going through analogous levels of
initialization provided by the operating system of the 3B20D
computer.

!'ICLYVV.lA

CUlU

'-U:HUlllt:r-;:)t~rVlces

Systems

538

Part 3

Service Capabilities
To foster open interconnection of customer equipment, the No.1 PSS
packet switch is designed to adhere to international standards. It supports, for example, all essential services and facilities of the 1980 version
of the CCITT X.25 protocol. It supports access line speeds of 9.6 and
56 kbps. Subscribers can interconnect their terminals, hosts, and office
equipment effectively without large start-up costs or network management expenses. Key design goals include high availability, rapid data
transfer (low end-to-end delay), high throughput (high average packet
rate per channel), and high capacity (large number of packets per unit
time over the switch as a whole). Operations centers and associated
operations systems (see Chapter 15) provide both centralized and distributed support to respond to customer trouble reports most effectively.
The No.1 PSS packet switch can support a common user network to
provide a highly reliable, full-time computer communications network
for users in need of basic data transport. A common user network allows
intercorporate communications and facilitates resource sharing of valueadded vendors who wish to market information services.
11.6.3 DATAPHONE SELECT-A-STATION SERVICE

IMPLEMENTATION
DATAPHONE Select-a-station service is a voiceband private-line data service that is designed for applications-such as alarm service bureaus-in
which a master station exchanges data with a number of remote stations,
one at a time, usually in rapid sequence. The service allows 2-way
transmission between the master and remote stations, but no direct
. transmission is possible between remote stations. Nor is broadcast communication possible between the master station and all remote stations.
The security of this service makes it particularly well suited for alarm service bureaus. Connection control can only come from the master station,
and all remote stations, other than the one connected at a particular time,
are isolated from the connected path and from each other. This ensures
that trouble in any leg cannot affect proper operation of the remainder of
the circuit. This isolation of each point-to-point connection also ensures
the privacy of communication between the master station and each
remote station.
To implement DATAPHONE Select-a-station service, high-speed
switches called data station selectors (DSSs) are located in the telephone
company's central office building to connect the customer's master station
with various remote stations (see Figure 11-41). Connection is established
by the DSS stepping automatically in a fixed sequence or by the customer
at the master station. The master station terminal is a minicomputer or a
specially designed controller owned and operated by the alarm service

ALARM BUREAU

t

PROTECTED PREMISES

CENTRAL OFFICE

CENTRAL OFFICE

SENSORS:

MASTER
STATION
TERMINAL
(CUSTOMER
EQUIPMENT)

,

*

SELECTOR
CONTROL
\

UNIT
DC CONTROL
INTERFACE

PRIMARY
DATA
STATION
SELECTOR

PORT 1

*
*

SECONDARY
DATA
STATION
SELECTOR

FIRE

SMOKE

CHANNEL
SERVICE
UNIT

REMOTE STATION
PORT 128

I

V

REMOTE STATION

REMOTE STATION

MASTER STATION

PORT 1

REMOTE STATION

REMOTE STATION
SECONDARY
DATA
STATION
SELECTOR
REMOTE STATION
PORT 128

* 4-WIRE VOICEBAND FACILITIES
t

2- OR 4-WIRE VOICEBAND FACILITIES

t

CUSTOMER PREMISES

Figure 11-41. System configuration for DATAPHONE Select-a-station service.

t

WATER INTRUSION

540

Network and Customer-Services
Systems

Part 3

bureau. Control signaling consists of the transmission over the
voiceband channel of control tones interspersed in time among the customer's data signals between the master station and the DSS. This inband
signaling avoids the need for a separate control channel.
A selector control unit (SCU), located at the customer's master station,
serves as the interface between the telephone company facilities and the
customer-provided equipment. A multilead interface between the SCU
and the terminal in the master station allows the customer various
degrees of control over the duration and sequence of connection to the
individual remote stations. The SCU will generate the necessary control
signaling in response to the interface commands.
Once the connection to a particular remote station is made, the customer has full responsibility for the end-to-end transmission. The customer provides the terminal equipment that accomplishes the data
exchange at both the master station and the remote sites. The master station can simply pick up an alarm signal, or it can transmit an activating
signal to a remote terminal, causing the terminal to return a report. The
remote terminal, supplied by the alarm bureau and located on the protected premises, has fire, smoke, or other sensors connected to it. Alarm
signals from the terminal go through a channel-service-unit interface
back over either 2- or 4-wire voiceband facilities to the master station.
In the customer-control version of the service, up to two DSSs,
denoted as primary and a secondary, may be placed in tandem. Such tandeming may be desirable to allow geographically dispersed concentrations of stations to be served more economically. In this arrangement,
the customer controls the sequence and interval for each primary port
through the SCU and the primary DSS. The secondary DSS automatically
steps through a fixed sequence to select associated remote stations. The
number of secondaries that may be placed on a single primary is limited
only by the number of output ports available on that primary; the maximum number is 128.

AUTHORS
M. D. Balkovic
H. J. Bouma
R. Carlsen
D. C. Franke
W. G. Heffron

F. G. Dram
K. J. Pfeffer
P. T. Porter
S. P. Shramko
R. W. Stubblefield

12
Common Systems

12.1 INTRODUCTION
In 1982, the Bell System network used approximately 20,000 telephone
equipment buildings, including wire centers (customer-oriented), toll
centers (long-distance oriented), transmission buildings, and radio relay
stations. In each building, common systems provide power, interconnection, and environmental support for network elements associated with
transmission and switching (see Figure 12-1). Since these systems are
shared by switching equipment and transmission facilities, careful
engineering is required. They must have the capacity to support growth
as well as transitions from old to new systems with minimum service disruption. The major classes of common systems are power systems, distributing frames, and equipment building systems. This chapter discusses
these major classes in detail and briefly discusses some other common
systems-cable entrance facilities, cable pathways, and alarm systems.

12.2 POWER SYSTEMS
Power requirements for telephone company equipment buildings vary
greatly depending mainly on the amount and type of equipment housed.
The smallest buildings (for example, community dial offices or small
repeater stations) may require less than 10 kilowatts of power, while a
large central office building may require several thousand kilowatts.
A major function of power systems is to convert alternating current,
supplied by an electric utility, to direct current, which is required for the
proper operation of most devices and electrical circuits (relays, switches,
electron tubes, transistors, integrated circuits, and even station sets) in
telecomm unications systems.

541

TELEPHONE EQUIPMENT BUILDING

TRANSMISSION
EQUIPMENT

SWITCHING
EQUIPMENT

COMMON SYSTEMS
-INTERCONNECTION
-POWER
-ENVIRONMENTAL
SUPPORT

SUBSCRIBER
TRANSMISSION
FACILITIES

TO CUSTOMER PREMISES

TO OTHER TELEPHONE BUILDINGS

Figure 12-1. The role of common systems.

Many power systems also provide reserve energy storage to ensure
service continuity if the normal energy supply is temporarily interrupted.
Reserve energy applies generally to central office switching systems,
transmission systems, and some computer-based operations systems. The
reserve source must be available for instantaneous use since many calls in
progress would be disconnected if power were interrupted even for only
a few milliseconds. The widespread use of computer-controlled systems
makes continuity of power even more important since information stored
in volatile memory is lost during power interruptions and must be
reloaded from backup memory before normal system operations can be
resumed. Normally, telecommunications equipment installed on customers' premises (for example, private branch exchanges [PBXs], key telephone systems, and data sets) is not provided with standby power facilities, although such facilities are available for critical applications at extra
cost. Section 12.2.7 describes the power systems available for customerpremises equipment.
Because the power facilities in a switching office or repeater station
support equipment used in common by many customers, there is extensive redundancy so that the failure of a single power system component
or unit does not disrupt system operation.
542

Chap. 12

Common Systems

543

12.2.1 ENERGY SOURCES
The conventional energy source for most telecommunications equipment
in the United States is alternating current purchased from an electric utility. It is modified and controlled as required for the specific telecommunications system.
Several unconventional power supply systems have been investigated,
and some have been installed in Bell System field experiments or in
remote systems. They include: continuously operated diesel-electric alternators l (installed at a remote microwave station in the Sierra Nevada
Mountains), solar cells (used in a rural carrier field trial and as an auxiliary power source in a large toll center), propane-fueled thermoelectric
generators (for remote repeaters and a digital radio system), and winddriven generators to supplement the electric utility supply. Extensive
research and development efforts by many organizations to develop new
and improved alternate energy sources may eventually bring about their
use in certain applications, such as in remote microwave repeater stations.

12.2.2 ENERGY STORAGE
Two types of energy storage systems are used extensively in the Bell System: electrochemical cells2 and standby alternator sets powered by
in ternal-com bustion engines.
Electrochemical cells provide an instantaneous reserve source of dc
power. One type, unsealed lead-acid cells, costs less than other electrochemical systems. In principle, they are similar to ordinary automobile
batteries. However, the cells used for telecommunications service are
designed for long life (typically 15 years or more) and engineered for
long discharge times (hours) at moderate temperatures. Lead-acid cells
are purchased in accordance with special design specifications and are
available in a wide variety of sizes from 100 to 1680 ampere-hour nominal ratings.
In the early 1970s, Bell Laboratories developed a cylindrical, unsealed
lead-acid cell that lasts longer and requires less maintenance. Detailed
specifications for both materials and fabrication were required to gain
these improvements. These cells are now being manufactured for the
Bell System according to these specifications. Sealed nickel-cadmium and
sealed lead-acid cells have been used in subscriber loop systems deployed
in the outside plant environment to avoid the routine maintenance
required with conventional lead-acid cells.
Lead-acid cells provide only short-term reserve power, so standby
internal-combustion engine-alternator sets (like the ones shown in
1 Machines that produce alternating current.
2 Devices that supply electricity through chemical action.

544

Network and Customer-Services
Systems

Part 3

Figure 12-2) are commonly installed in many telephone company buildings to provide a long-term reserve as described in Section 12.2.3. The
alternator, driven by the engine, produces alternating current to replace
electric utility power. For these systems, energy is stored in the form of a
liquid hydrocarbon fuel (for the engine). A typical communications
center may provide fuel storage for 2 days to 3 weeks of continuous
operation of essential equipment. This fuel supply can be replaced to
achieve unlimited reserve energy storage.
Engine-alternator sets use either diesel engines or turbines. Diesel
engines are generally used in sizes up to about 200 kilowatts, while gas
turbines are used in larger sizes-up to about 2000 kilowatts. Gas turbines are internal-combustion engines and are similar to jet aircraft turbines; the blades of the turbine are propelled by hot gases produced by
the combustion process. Gas turbines may burn different types of fuel; in
Bell System app lications, they use diesel fuel. Gas turbine units are generally much lighter and smaller than comparable diesel engine sets and
can be installed in upper floors or on a building roof, whereas diesel
engine-alternator sets are generally installed in the basement or first
floor. In most buildings, a single turbine unit is installed. Multiple
installations may be used in larger buildings.

Figure 12-2. Standby engine-alternators. Left. 100-kilowatt diesel; right.
750-kilowatt turbines .

12.2.3 POWER SYSTEM OPERATION
The elements of a typical power system are shown in Figure 12-3. While
the dc power plant is common to all telecommunications equipment locations, the engine-alternator may be omitted where long-term reserve
power is not required. The dc-to-ac inverter may be omitted if there is no
essential equipment requiring ac power, and the dc-to-dc converters are

DC POWER PLANT

r-------------------, r-------------,
TYPICAL EQUIPMENT FRAME

5 VOLTS
DC
ELECTRIC
UTILITY - - - -...
CAC POWER)

CONTROL
AND
MONITOR

RECTIFIER

12 VOLTS
DC

•

•
•
ENGINE
ALTERNATOR

RECTIFIER

T

BATTERY

FUEL
STORAGE

TRANSFER
SWITCH

DC-AC
INVERTER
DISTRIBUTION

FACILITIES _
L ____________________

Figure 12-3. Typical power system.

AC
LOADS

546

Network and Customer-Services
Systems

Part 3

provided where voltages not readily available from the dc power plant
are required. Sections 12.2.5 and 12.2.6 discuss converters and inverters,
respectively.
The dc power plant consists of four main elements: rectifiers; a group
of lead-acid cells connected in series, commonly called a battery; control
and monitoring equipment; and distribution facilities.
A typical dc power plant is shown in Figure 12-4. The power plant
accepts alternating current when it is available and rectifies it to produce
dc power that is then supplied to the telecommunications equipment.
The battery provides a source of reserve dc power that is automatically
supplied if the electric utility fails and continues until the lead-acid cells
are discharged. In normal practice, the battery is selected to provide 3 to
8 hours of reserve time.
The engine-alternator set in a power system is normally idle and is
started either manually or automatically after a disruption of the electric
utility supply. Its output then replaces the utility supply, and the

DISTRIBUTION
FACILITIES

Figure 12-4. Typical dc power plant.

Chap. 12

Common Systems

547

rectifiers again supply power to the telecommunications equipment and
recharge the battery. The hours of battery reserve time and the choice of
manual or automatic-start engine-alternator are determined by the operating company for each location, based on the history of commercial power
outages and accessibility of the location to service personnel. An unattended microwave station on a remote mountaintop, for example, may
have as much as 24 hours of battery reserve, as well as automatically
operated engine-alternator sets.
In small community dial offices and readily accessible carrier repeater
stations, a standby engine-alternator may not be furnished. In this case,
the battery may be engineered to provide 24 hours or more of reserve. In
the event of a long-term failure, a portable engine-alternator is brought
to the site.
A number of dc power plants with different nominal output voltages
for different applications have been developed. Table 12-1 identifies
some of the common applications of these plants.
The description of a power system given above is adequate to illustrate its operation as a power conversion and energy storage system.
However, the plants in Bell System installations include additional
features to control the output voltage and maintain the battery in a
proper state of charge.

TABLE 12-1
DC POWER PLANT OUTPUT VOLTAGES
Nominal
Voltage
(Volts)

Application

-24

Microwave radio, transmission multiplex

-48

Electromechanical and electronic switching systems,
transmission systems colocated with switching systems

±130

Electron tubes, remote repeaters, telegraph systems

+140

4ESS switching equipment, L5 power feed stations, large
toll centers

12.2.4 RECTIFIERS
Rectifiers convert alternating current to direct current. In normal operation, they supply dc power to the telecommunications equipment (the
load) and maintain the battery in a fully charged state. In the event of an

548

Network and Customer-Services
Systems

Part 3

electric utility outage, some discharge of the battery will occur. When
utility power is restored or if an engine-alternator set is activated, the
rectifier output will recharge the battery and resume supplying power.
The equipment units are installed and connected in parallel to provide
added capacity as required. Rectifiers with sizes up to about 80 kilowatts
(1600 amperes at 48 volts) are in service in telecommunications power
plants.
12.2.5 DC-TO-DC CONVERTERS
The introduction of new systems and devices has generated a need for
voltages not readily available from a dc power plant. In particular, the
large-scale use of semiconductor and integrated circuits, which require
well-regulated supply voltages between 2 and 12 volts, has provided the
impetus for the development of dc-to-dc converters to convert the output
of a dc power plant to the required dc voltage.
Since the dc-to-dc converter interposes a major system element
between the dc power plant and the load, the effect of a converter failure
must be considered. A common approach is to supply a converter to
power a single functional unit (for example, a channel bank, radio
transmitter, or memory unit) and depend on system reliability arrangements, such as protection switching or redundancy, to maintain service if
a converter fails. This approach may result in high costs, especially when
many small functional units are required in a given system. In these
cases, parallel operation of converters3 may be more desirable. Many converter circuits include monitor, alarm, shutdown, and other peripheral
functions to meet system needs.
Since the dc-to-dc converter must generally provide close control
(regulation) of its output voltage, it should be physically close to the load
to minimize voltage drop4 especially in low-voltage applications. Small
converters are often mounted on the same circuit pack as the load requiring the dc voltage. Medium-size units (20 to 500 watts) are designed as
plug-in modules and are mounted in frames with the circuits they supply, as shown in Figure 12-3. Larger converters are generally mounted in
separate bays in communication equipment rooms.
12.2.6 INVERTERS
Although most loads in telecommunications systems that require reserve
power are operated by direct current, certain loads-such as motors in
tape drives-require alternating current. Where such loads must operate

3 Where the outputs of several converters (including a spare) are combined to supply all
functional units. This is similar to the configuration of rectifiers in Figure 12-3.
4 A reduction in voltage caused by current flow in the conductors between the power
supply and the load.

Chap. 12

Common Systems

549

during electric utility supply interruptions, solid-state dc-to-ac inverters
provide reliable ac power. These units use semiconductor switching devices to convert the dc supply from a battery to alternating current.
The use of commercial computer systems for essential operations has
increased the demand for continuous ac power since these systems generally are not designed for dc operation. Large uninterruptible power
systems employing dedicated dc power plants and multiple inverters for
redundancy are being used to fill these needs.
12.2.7 POWER FOR CUSTOMER-PREMISES EQUIPMENT

Most PBXs, key telephone systems, data stations, and other equipment on
customers' premises do not have standby power facilities. These systems
are energized from the electric utility supply through rectifiers to provide
the dc voltages required. The rectifiers are often mounted in the cabinet
that contains the communication equipment. If power fails, PBX systems
automatically transfer selected stations to central office trunks to provide
limited communication capability. The actual switching takes place when
relays operated by PBX power release because of PBX power failure. In
releasing, the relay contacts transfer the telephones to central office circuits. Until power is restored, service to the transferred telephones is
much like the service provided to a single business or residence line.
In critical locations such as hospitals and emergency services, batteryinverter plants are furnished to provide reserve capability. In some locations, a standby engine-alternator set furnished by the customer may be
used to power communication facilities and other important services if
the utility supply fails.

12.3 DISTRIBUTING FRAMES
12.3.1 GENERAL DESCRIPTION

In a typical telephone equipment building (see Figure 12-5), the distributing frame is a common termination point for all equipment and facilities at that location. The facilities include subscriber cables to customer
premises and trunk cables to other telephone equipment buildings; the
equipment includes switching equipment (both the line and trunk sides
of the switching network),5 and a wide variety of transmission equipment
including amplifiers, signal converters, and test access systems. By interconnecting specific equipment and facilities with distributing frame
jumpers (see Section 12.3.2), basic and special telecommunications services
can be provided. Distributing frames may also connect cables that are

5 Section 7.3 discusses switching networks.

TELEPHONE EQUIPMENT BUILDING

JUMPER
DISTRIBUTING
FRAME

SUBSCRIBER
FACILITIES
TRUNK
FACILITIES

Figure 12-5. Distributing frame role in the
telecommunications network.

part of carrier systems to the appropriate transmission equipment for
functions such as multiplexing6 and demultiplexing.
Each wire center and toll center in the Bell System has its own distributing frame system. The size and complexity of these systems vary
widely from small, single lineups to complex networks involving
numerous large distributing frames interconnected by thousands of tie
cable links. About 20,000 technicians are required to operate distributing
frames in the Bell System.

12.3.2 DISTRIBUTING FRAME FUNCTIONS
Distributing frames provide three basic functions: cross-connection,
electrical protection, and test access.
Cross-connection of two or more outside plant facility and office
equipment terminations is required to provide service to a customer.
Individual cross-connect wires called jumpers, which can remain in place
for a few days to several years, join the termination points of cables
representing particular facilities and equipment. To ensure efficient use

6 Section 6.5 discusses multiplexing.

550

Chap. 12

Common Systems

551

of facilities and equipment, Bell System policy has required total interconnection flexibility on distributing frames (that is, any outside plant
facility or piece of office equipment must be able to be connected to any
other).
Distributing frames also provide electrical protection for equipment
and personnel within a wire center. Fuse-like protector devices are
mounted on distributing frames to guard against spurious voltages and
currents. These electrical incursions could result from lightning strikes
or power-line crosses anywhere in the outside plant.
In addition, distributing frames provide an access point where circuits
can physically be opened and tested. Temporary disconnection of customers is accomplished by setting a protector unit (see Figure 12-6A) to an
inactive position and does not require removal of the cross-connection.

12.3.3 DISTRIBUTING FRAME HARDWARE AND APPLICATION
Figure 12-6 shows conventional distributing frame hardware, which was
the only kind used in the Bell System until 1964. As Figure 12-7 shows
schematically, the conventional frame is a double-sided structure composed of vertical and horizontal parts. Typically, outside plant cables run
to terminal apparatus mounted on the vertical side where electrical protection is provided. Cables from equipment located in the building terminate on the horizontal side. Jumpers between terminals on the vertical
and horizontal sides run on shelves on the horizontal side and along
vertical troughs to the vertical side termination.
The most widely used conventional frame in the Bell System is 11 to
14 feet high, and installations up to 400 feet long are in service. Two
technicians are required to run a jumper, and ladders are needed to reach
the upper shelves. Hence, these frames are somewhat unwieldy to
operate. The low-profile conventional distributing frame was introduced
in the early 1970s. It is about 8 feet high and is supported only at the
floor. It conforms to the New Equipment-Building Systems (NEBS) standards (discussed in Section 12.4.5) and does not require ladders, although
two technicians may still be needed to run jumpers.
Distributing frames may be interconnected for different applications to
form distributing frame networks. The frames that make up these networks are given functional designations according to the type of plant
facilities terminated on them. The functional frames in such networks
are linked by tie cables.
A combined main distributing frame (CMDF) has both subscriber and
trunk outside plant cable terminations (see Figure 12-8A). A subscriber
main distributing frame (SMDF) terminates only customer-oriented
cables, and a trunk main distributing frame (TMDF) terminates only
interoffice cables and transmission facilities (see Figure 12-8B). Finally,

A

PROTECTOR
UNITS

OUTSIDE PLANT
CABLE TERMINATIONS

B

EQUIPMENT
TERMINATIONS
(BOTTOM SIDE
OR
BACK SIDE)

JUMPERS
(FRONT OR
TOP SIDE)

Figure 12·6. Conventional
distributing
frame.
A,
vertical
side :
termination of outside plant facilities on electrically protected connectors ; B,
horizontal side : termination of office equipment with associated jumpers.

CABLE FROM
CENTRAL OFFICE
EQUIPMENT

Figure

12-7.

Conventional

frame

hardware

schematic

diagram.

Terminations for equipment and outside plant cable run all along the frame,
as shown in Figure 12-6.

an intermediate distributing frame (IDF) has no outside plant facilities
terminated on it.
Either conventional or modular hardware may be used in distributing
frame networks. Figure 12-9 shows one modular distributing frame, the
Common Systems Main Interconnection Frame System (COSMIC). Modular frames were first introduced during the 1960s for use with lESS
switching equipment and underwent a major redesign in the early 19708.
Office equipment and outside plant facilities are cabled to alternate
modules by means of the backplane of a modular frame. Each module
contains numerous office equipment or outside plant facility terminations.
Cross-connections are run on the front face of the frame. Vertical and
horizontal troughs are provided to accommodate the cross-connections in
an orderly manner. Modular frames are low, and one technician can run
jumpers; thus, they offer potential operations efficiencies over conventional frames.
Modular frames in common use include: ESS switching equipment
frames, the COSMIC frame (shown in Figure 12-9), and COSMIC II. The
most widely used modular frame is the COSMIC frame, which comprises
553

LINE
EQUIPMENT

SWITCHING
SYSTEM

TRUNK EQUIPMENT

COMBINED
FRAME

MISCELLANEOUS
EQUIPMENT

A

SWITCHING
SYSTEM

LINE
EQUIPMENT

TRUNK
EQUIPMENT

SUBSCRIBER
FRAME

TRUNK
FRAME

B

Figure 12-8. Typical distributing frame networks. A, combined main
distributing frame; 8, subscriber and trunk main distributing frames.

IO,OOO-pair alternating modules of outside plant and office equipment.
Because this COSMIC frame does not accommodate electrical protector
units, a separate frame called the protector frame must also be used.
Cables connect the protector frame and the COSMIC frame.
The COSMIC II distributing frame also uses alternating large modules
of facilities and equipment. However, it differs from the earlier COSMIC
in two important ways: (1) the separate protector frame is replaced by
protected facility terminations on the rear of the outside plant modules,

554

Figure 12-9. Modular (COSMIC) distributing frame showing
modules for terminating outside plant facilities and office equipmen t.

and (2) it has a wider range of applications-it can b e used as a subscriber, a trunk, or a combined main distributing frame, while COSMIC is
always used as a subscriber frame .

12.3.4 DISTRIBUTING FRAME ADMINISTRA nON AND
ENGINEERING
Distributing frame administration includes recordkeeping and assignment
of the facilities and equipment that provide telecommunications services.
Distributing frame engineering provides adequate frame capacity and
ensures systematic layout of facilities and equipment as the office grows.
Providing telecommunications services to customers involves a
sequence of operations related to distributing frames . First, a customer

555

556

Network and Customer-Services
Systems

Part 3

request is transmitted to an assignment center where specific facilities
and equipment are selected from lists of spares at the distributing frame.
This assignment of circuit elements attempts to satisfy several requirements simultaneously, including circuit needs, equipment load balance
(see Section 5.3.4), and short cross-connections (jumpers). After facilities
and equipment are assigned, a service order is transmitted to the frame
force. The frame force makes the necessary cross-connections, tests the
circuits, and sends an order completion report to the assignment center.
It has become increasingly important during recent decades to attain
short jumper assignments because distributing frames have been growing
larger. Larger frames have more jumpers on their shelves and troughs,
and the jumpers tend to congest midway along the frame's length. Figure 12-10 shows an overloaded main distributing frame in a central office.
Preferential assignment (see Figure 12-11) is a method of attaining more
short jumpers than would be obtained by randomly assigning from lists
of spares. With preferential assignment, both the facilities and equipment sides (modules) of the distributing frame are administered in
"zones" with assignment preferences from A to A, then A to B, then A to
C, etc. This type of zoned administration can reduce the number of
jumpers that cross midway along the frame, provided that the various
types of facilities and equipment are distributed in all zones. This latter
condition is accomplished by distributing frame layout control.

Figure 12-10. An overloaded main distributing frame in a central office.

FACILITIES
SIDE

EQUIPMENT
SIDE

FACILITIES
SIDE

EQUIPMENT
SIDE

ZONE A

Figure 12-11.

ZONE B

ZONE C

Conventional distributing frame assignment methods.

Top, random assignment; bottom, preferential assignment.

Layout control of terminations is an engineering technique that was
introduced for conventional frames in the early 1970s with zoned layout
schemes. The objective is to spread old and new terminations of all
categories of equipment and facilities evenly over a distributing frame at
any point in time. This can be accomplished with the aid of a termination
layout mask (see Figure 12-12), which reserves particular regions of a distributing frame for specific termination categories.

SUBSCRIBER
CARRIER

SUBSCRIBER
CABLES

TIE CABLES

SUBSCRIBER
CARRIER

SWITCHING
SYSTEM
LINE EQUIPMENT

SWITCHING
SYSTEM
LINE EQUIPMENT
SUBSCRIBER
CABLES

TIE CABLES
FACILITIES

TIE CABLES

EQUIPMENT

TIE CABLES
FACILITIES

EQUIPMENT

Figure 12-12. Subscriber MDF termination layout mask.

557

558

Network and Customer-Services
Systems

Part 3

Modular frames can be viewed as hardware manifestations of the
zoned layout concept of conventional frames where the layout of facilities and equipment is reflected in the alternating physical hardware
modules. Figure 12-13 shows a technician running a preferentially
assigned short jumper on a layout-controlled modular frame.
Mechanized layout control was first introduced for COSMIC frames
with a system called the Program for Arrangement of Cables and Equipment
(PACE) and is now being extended to other applications with a new system called the Mechanized Engineering and Layout for Distributing Frames
(MELD). Mechanized algorithms spread each type of facility and equipment uniformly when the frame is initially installed and as terminations
are added over its service lifetime.
The manual administration process of maintaining facility ilnd equipment inventories and assignments is complex. Insufficient coordination
and followup can cause errors that result in wasted operations at the distributing frame. Poor records may mean that unused equipment and
facilities are terminated at the distributing frame but not recognized as
spares. In the manual mode, records are difficult to maintain and preferential assignment is almost impossible to implement; hence, efforts to
mechanize record and assignment processes were begun in the early
1970s. The result has been implementation of substantial distributing
frame administrative capabilities (including preferential assignment) in

Figure 12-13. Technician running a preferentially aSSigned
short jumper on a COSMIC distributing frame.

Common Systems

Chap. 12

559

computer-based systems such as the Computer System for Mainframe
Operations (COSMOS) and the Trunks Integrated Records Keeping System (TIRKS).7 With these operations systems, available equipment and
facilities are also used more efficiently.

12.4 EQUIPMENT BUILDING SYSTEMS
12.4.1 TELEPHONE EQUIPMENT AREAS

Approximately 60 to 85 percent of the interior space in a telephone
equipment building consists of equipment areas. These are large, usually
windowless, partition less rooms designed to contain the equipment, the
appropriate cable support systems, and the air ducts for the environmental control the equipment needs to function (see Figure 12-14).
Equipment space and associated cabling are required for switching
systems; transmission systems; and common systems such as power, main
distributing frames, the cable entrance facility (see Section 12.5.1), and
environmental support.
The switching and transmission equipment installed in telephone
equipment buildings is usually mounted on steel framework lineups 1 to
2 feet deep and up to 6-Ih feet wide (see Figure 12-15). The frames are
installed side by side in lineups usually 30 to 50 feet long. Because the
frames are large and heavy, they are bolted to the floor and/or to the

CABLE
ENTRANCE
FACILITY

Figure 12-14. Equipment space in telephone equipment building.

7 Section 14.2.1 describes TIRKS.

Figure 12-15. Equipment frames in telephone equipment buildings.
A, transmission equipment; B, crossbar switching equipment mounted on
11 · J.1! foot high framework lineups; C, ESS switching equipment mounted on
7·foot high framework lineups.

Chap. 12

Common Systems

561

ceiling with concrete-embedded anchors to eliminate vibration and to
prevent toppling.
Typically, aisles 2 to 4 feet wide between lineups allow access to
equipment for wiring and maintenance. Other steel structures above the
equipment framework support the tons of cabling that interconnect the
equipment and connect the equipment to the distributing frames (see
Figure 12-16A). (Section 12.5.2 describes cable distribution systems.)
Main distributing frames (see Section 12.3) are also in the equipment
area. In multistory offices, distributing frames may be located on several
floors, requiring vertical access for cables (see Figure 12-16B). Many
heavy bundles of interfloor cabling are used. In the equipment area, steel
framework is positioned between floors to support the vertical cabling,
and special reinforcement is used in the floors to maintain structural
integrity at points of penetration.
The power systems (see Section 12.2) that provide the power conversion and uninterrupted energy source for the telephone equipment are
also located in equipment areas but are usually located in power rooms
separate from the transmission and switching equipment.
12.4.2 BUILDING ELECTRICAL SYSTEMS
In addition to the dc power systems that directly serve the telephone
equipment, a telephone equipment building contains a number of other
special electrical systems. These systems are designed for extreme reliability and high capacity; they use special control and protective circuits.
• The ac power system, fed by the electric utility, consists of service
entrance, protection, distribution, and control apparatus. The system
serves the building's electrical and mechanical systems, as well as the
telephone power plants discussed in Section 12.2.
• Extensive electrical grounding systems are placed throughout each
building to eliminate noise on lines, reduce high-speed data errors,
and protect the telephone equipment from electrical short circuits and
lightning strikes.
• Shielding systems are provided in offices located near sources of
high-intensity electromagnetic or electrical fields such as television
broadcasting stations, electrical power stations, or certain high-tension
lines. Shielding prevents or reduces the penetration of electromagnetic fields into electronic equipment. Such penetration can cause
malfunctions.
12.4.3 BUILDING MECHANICAL SYSTEMS
Equipment buildings have two general types of mechanical systems. The
first type provides such environmental support as control of the temperature, regulation of the humidity of the surrounding air, and maintenance

Figure 12-16. Cabling in telephone equipment buildings. A. routed above
equipment frames ; B. rout ed between floors .

Chap. 12

Common Systems

563

of the appropriate amounts of outside air of high purity. The second
type provides for vertical access in multi floor structures and for distribution or movement of air, water, and fuel. Mechanical equipment areas
occupy up to 25 percent of the gross space, depending on the process
cooling,S humidification, and air filtration needed by the telephone
equipment and reserve power engines. Mechanical equipment systems
include:
• Cooling systems designed to remove the heat released from telephone
equipment. Recent developments that use solid-state devices have
miniaturized telephone equipment components. The result is closepacked equipment that dissipates large amounts of heat and, therefore,
requires exceptional amounts of cooling. For example, a 10,000square-foot toll office will require up to 100 tons 9 of cooling capacity
when it houses electronic transmission and switching equipment.
Cooling an equivalent floor area for human comfort would rarely
require more than 15 percent of this air-conditioning tonnage.
• Ventilating fan systems with enough capacity to maintain short-term
ambient operating environments for the telephone equipment in the
event of a cooling-system failure. While repairs are being made, these
fan systems use outside air.
• High-capacity air filtration systems necessary to prevent dust and
products of combustion from infiltrating the building and causing
electrical contact failures.
• Vertical access space within the structure provided for water and
drain lines, fuel lines and exhaust stacks for reserve power plants, and
movement of large, bulky equipment assemblies between floors using
freight elevators or hoisting shaftways.
The location of vertical runs is coordinated with the equipment plan
for the building to ensure compatibility with the placement of future
generations of equipment.

12.4.4 SPECIAL CONSTRUCTION
Another important characteristic of an equipment building or transmission station is the provision for expansion. If a horizontal addition to the
building is anticipated, a rear or side wall must be designed so that it can
be removed without interfering with the structural integrity of the roof

8 A procedure for maintaining a cool environment for equipment rather than for people.
9 In this usage, a ton is equal to 12,000 British thermal units (Btus) per hour.

564

Network and Customer-Services
Systems

Part 3

and floors or with equipment assemblies that are operating to provide
service. Extension of the air distribution ducts and refrigeration
machinery must be accommodated. If vertical additions are anticipated,
the footings, columns, and load-bearing walls must be adequate for the
future configuration.
Special construction is also required where offices have roof-mounted
microwave radio towers. These towers must be able to support several
antennas and component assemblies weighing hundreds of tons. When a
tower and antennas are on top of a building, the load is carried through
the building to the foundation and requires a massive internal support
system.
12.4.5 EQUIPMENT BUILDING SYSTEM STANDARDS

During its 100-year history, the Bell System has used three different sets
of building design standards. The earliest equipment buildings were
designed primarily for operator switchboards. In the mid-1920s, with the
introduction of automatic switching, building standards were changed to
accommodate 11-Ih foot equipment framework lineups. Then, in the
early 1970s, the standards were changed to provide 10 feet of clear height
for equipment and cabling. These New Equipment-Building Systems
(NEBS) standards were motivated by the trends in electronics that made it
necessary to fit more compact designs with their higher heat and cabling
requirements into existing space. A second objective was reducing costs
in constructing new buildings. The NEBS documents include a set of
coordinated specifications for equipment and buildings:
• Equipment Design Standards provide the spatial and environmental
performance requirements (for example, air conditioning and lighting
requirements) for all new equipment systems.
• Building Engineering Standards specify the planning and design of
new buildings and additions and provide guidelines for the reuse of
existing space to accommodate modern electronic equipment.
Equipment building standards are important not only to the performance of the facility but also to the cost of buildings, since the physical
characteristics of telephone equipment have an effect on the design of the
building intended to house that equipment. Frame height and cabling
space, for example, control the "clear" ceiling height from the floor to the
lowest overhead obstruction. Equipment weight determines the
building's live load-the weight that foundations, columns, and floors
must support in addition to their own weight. The amount and location
of heat emanating from the equipment determines the size of the cooling
plant and the location of the air ducts and diffusers. 10
10 Fixtures that attach to a duct and distribute air in aisles between equipment frames.

Chap. 12

Common Systems

565

Problems can arise when equipment units differ greatly in physical
characteristics. To avoid these problems, NEBS standards include the full
range of spatial and environmental conditions. The requirements cover
equipment frame areas, distributing frame areas, power equipment areas,
operations systems areas, cable distribution systems, and cable entrance
facilities. The environmental requirements are grouped according to
functional effects and include fire resistance, grounding, radio-frequency
interference, thermal effects, shock, vibration, earthquake, airborne contaminants, acoustical noise, and illumination. NEBS documents provide
standards, design requirements, and planning guidelines for the building
structure, for each equipment area in the building, and for all the building support systems. Use of the standards simplifies building design and
equipment engineering, streamlines equipment and cable installation,
and allows for flexibility in growth patterns. Equipment buildings vary
widely in size and appearance, depending on the application, as the
examples in Figure 12-17 show.

12.5 OTHER COMMON SYSTEMS
12.5.1 CABLE ENTRANCE FACILITY
A typical telephone equipment building must accommodate thousands of
pairs of wires, coaxial cables, and optical fibers from outside plant
transmission facilities. The cable entrance facility (CEF) provides an
entrance area for all types of outside plant cables carrying subscriber
lines and transmission facilities between equipment buildings. As illustrated in Figure 12-18, a typical cable entrance facility is a vault-like
below-grade area. It is typically 12 to 15 feet high and 12 feet wide and
runs the length of the building directly under the main distributing
frame(s). It can be over 400 feet long. One or both of the end walls contain a conduit termination with a built-in gas-venting chamber that is
used to prevent water and hazardous gas from entering the central office
building.

12.5.2 CABLE DISTRIBUTION SYSTEMS
Switching and transmission systems in telephone buildings use multipair
cables to connect submodules within a system or to interconnect that system with other (common) syste1p.s such as distributing frames and power.
The overhead cable distribution systems and associated common
hardware for the NEBS equipment take into account the requirements for
cabling, cooling, assembling, lighting, and maintaining the equipment.
Cable distribution systems are provided in modular arrangements to simplify engineering and installation. Although designed primarily for use
in NEBS buildings, these arrangements can be modified to suit job conditions when NEBS equipment is installed in non-NEBS space.

Figure 12-17. NEBS equipment buildings.
metropolitan toll center; C, urban wire center.

A , suburban wire center; B,

ENTRANCE DUCTS

CABLE PRESSURIZATION EQUIPMENT

Figure 12-18. Cable entrance facility.

The overall coordination of the superstructure and common hardware
required for NEBS equipment is achieved with the Cable Pathways Plan.
This plan standardizes the maximum size and possible locations of cable
racks and integrates the frame and aisle lighting system with the cable
distribution system. The Cable Pathways Plan also incorporates the
building columns, cable holes, cooling air diffusers, fire detectors, and
access requirements in an overall allocation of space that minimizes possible conflicts throughout the life of the building.

12.5.3 ALARM SYSTEMS
Telephone buildings use a variety of alarm systems to indicate equipment
failure and/ or service interruption. Some typical examples are switching
system alarms, transmission terminal alarms, utility power alarms, and de
power alarms. In addition, extensive detector, alarm, and control systems
are employed throughout the building for protection against fires.

AUTHORS

E. J. Kovac
R. J. Skrabal

J. J. Stockert
567

PART FO(JR
OPERATIONS

Part Four expands on the brief discussion of operating telephone company functions in Chapter 1. It examines how these companies manage
their resources and interact with customers to provide services over the
network.
Chapter 13 provides an overview of the major customer- and
network-related operations. It emphasizes relationships, interfaces, and
chronological sequences among a telephone company's major operations.
Chapter 14 examines the role of operations systems in operating telephone companies. Rather than cataloging the many operations systems
in use, the chapter illustrates the use of generic classes of systems and
their impact on operations by describing the functional characteristics
and benefits of a few selected systems. Chapter 15 describes the planning
process that supports telephone companies in organizing their operations
and the impact of this planning on the operating companies and on the
development and evolution of operations systems. Part Three concludes
with Chapter 16, which describes the activities and considerations
involved in defining and maintaining desired levels of service and performance. This is an appropriate closure since meeting these levels is the
central objective of telephone company operations.

569

13
Overview of
Telephone Company Operations

13.1 INTRODUCTION
This chapter outlines some of the many activities or functions required
for operation of the telecommu.nications system. As indicated in the
Foreword of this book, the material presented reflects the Bell System as
it was at the end of 1982 and early in 1983. Then, and for many years
prior, operation of the Bell System was a shared responsibility, with Bell
operating companies responsible for operations in their territories and
AT&T Long Lines responsible for interstate and international operations.
Divestiture of the Bell operating companies from AT&T will result in
changes in responsibility for certain operations functions and, in some
cases, will change the way the functions are accomplished. However, the
basic functions required will, by and large, remain the same, so that the
material in this chapter will remain relevant from the standpoint of
understanding what is involved in telephone company operations.
Although the discussion focuses on Bell System operations, the activities
described are also applicable, in a broad sense, to the many independent
telephone companies.
Telephone company operations can be divided into three kinds of
functions.

• Provisioning is the process of making the various telecommunications
resources (such as switching systems, transmission facilities, and
operators) available for telecommunications services. Provisioning
includes forecasting the demand for service, determining the additions
(or changes) to the network that will be needed, determining where
and when they will be needed, and installing them.

571

572

Operations

Part 4

• Administration covers a broad group of functions that sustain services
once they have been provided. Administration generally consists of
network administration and service administration. Network administration ensures that the network is used efficiently and that grade-ofservice l objectives are met. Service administration includes such diverse
functions as billing; collecting and counting coins from coin telephones; and, for customer switching systems, giving engineering and
service evaluation assistance and keeping detailed engineering
records.
· Maintenance operations ensure that network components work properly once they are installed. Maintenance includes the testing and
repair activities that correct existing malfunctions (corrective maintenance) and those that prevent service-affecting malfunctions (preventive maintenance).
Bell System operations are complex and involve close to a million
employees. To handle this enormous enterprise rationally and efficiently,
operating company employees and the functions they perform have been
grouped into operations centers (see Section 15.2.1). The people in an
operations center report to a common manager and perform closely
related functions for geographic areas that range from national to local in
scope. In addition, operations centers may be customer specific or service
specific. For example, a customer-specific operations center may serve
residence customers; a service-specific operations center may be responsible for a service such as PICTUREPHONE meeting service.
The growing application of computer technology to operations is a
significant factor in the productivity gains made by the Bell System.
Computer-based operations systems (see Section 15.2.2) mechanize much of
the routine, time-consuming, often tiresome tasks (such as recordkeeping)
and enable people to do their jobs more accurately and efficiently. These
systems may also accomplish very complex tasks that could not be done
manually. (Chapter 14 discusses some of the major operations systems in
detail.)
The application of people and machines to operations requires considerable ongoing effort to meet the rapid evolution of market opportunities
and technology or, where possible, to anticipate this evolution. AT&T
and Bell Laboratories are engaged in planning the evolution of operations to ensure that operations related to new products, services, and
technology are efficient, that they benefit from continuing mechanization
opportunities, and that they are responsive to market demands. Because
the Bell System operation involves hundreds of operations centers and

1 Section 5.2.2 discusses the grade-oJ-service concept.

Chap. 13

Overview of Telephone Company
Operations

573

systems, another goal is to see that the whole collection of people and
machines work together. The operations planning process identifies the
need for new or changed roles for existing operations centers and systems, as well as the need for entirely new centers or systems. Thus, it
acts as a starting pOint for the development of methods and procedures
that define how the people in the operations centers should perform their
tasks (largely an AT&T function) and the design and development of
operations systems (largely a Bell Laboratories function). Ultimately, the
Bell operating companies carry out the plans by deploying the operations
centers and systems in their service areas. (Chapter 15 discusses operations planning in detail.)
The rest of this chapter describes customer-related operations (Section
13.2) and network-related operations, the activities required to provide,
administer, and maintain the network and its elements (Section 13.3). It
is important to keep in mind that the network is a shared resource; thus,
while Section 13.2 emphasizes customer contact operations, it frequently
references the network operations needed to serve a customer.
Many other essential operations are not discussed here. For instance,
one set of operations provides centralized support services to both customer-related operations and network operations. These operations exist
in most large businesses and include such functions as providing administrative services (for example, payroll, comptroller, and legal services);
operating motor vehicle pools; and managing real estate, inventory, and
materials. Other operations are covered in other areas of the book; for
example, Chapter 16 discusses the measurement of network performance
and customer satisfaction, and Section 17.3 discusses the development of
tariffs. Western Electric Company operations, while also essential to the
Bell System, are beyond the scope of this book.

13.2 CUSTOMER-RELATED OPERATIONS
This section describes Bell System operations that are directly related to
customer service. Because Bell System customers are so diverse in their
needs and expectations, customers with similar needs are grouped into
market segments, and operations are often adapted to these segments.
Thus, a residence customer with simple service needs would be handled
differently from a business customer with complex and sophisticated communications needs.

13.2.1 PROVISION OF SERVICE TO THE CUSTOMER
The service-provisioning process begins with the first contact between the
customer and the telephone company representative to negotiate service.
It ends with the satisfactory delivery of a product or service, timely billing for the service provided, and updating of all related records. Figure

574

Operations

Part 4

13-1 shows the major functions of the provisioning process and the
sequence in which they are performed.

c

u
S
T

o
M
E
R

Figure 13-1. Service-provisioning operations - functional flow.

Order Negotiation and Marketing
The first phase of provisioning is order negotiation, that is, the negotiation of services through face-to-face contact or telephone contact initiated
by either the customer, a Bell operating company (BOC), or AT&T Long
Lines.
A service representative handles most customer-initiated requests either
in a retail sales environment or over the telephone in a local service
center. The service representative and the customer discuss the types of
equipment,2 services, and features available and agree on what is to be
provided. To complete the interaction, the service representative must
usually access the customer's current service and billing records if they
exist. For a new customer, the service representative must obtain directory listing information (for example, name, address, and type of listing)
and billing and credit information (for example, billing name and ['.ddress
and current employment status).
Customer-initiated requests for complex services, such as a private
branch exchange (PBX) and data and private-line services, involve marketing functions. The marketing organization also initiates contact with
selected accounts such as large business customers. It discusses problems

2 Under provisions of FCC Computer Inquiry II (see Section 17.4.3), effective January 1,
1983, Bell operating companies can provide only that customer-premises equipment (CPE)
in their inventory on that date. New CPE from AT&T must be furnished through a
separate subsidiary. Customers may also obtain CPE from non-Bell System vendors.
Neither the Bell operating companies nor Bell Laboratories can market or promote
products of the subsidiary. CPE that is part of the embedded base of the Bell operating
companies, that is, installed or in inventory on January 1, 1983, will be transferred to the
subsidiary no later than the time of divestiture.

Chap. 13

Overview of Telephone Company
Operations

575

and concerns with the customer's decisionmakers and confirms the customer's interest in having the Bell System study and propose a communications system that would best resolve these problems. The customer
makes a further commitment to provide the necessary information and
resources to aid the marketing team during this, process. Once the marketing representative secures this commitment, the sales support function
helps the marketing team design a telecommunications system tailored to
the customer's needs. Sales support organizations in the BOC help to
acquire the necessary data (number of calls handled, peak hours, number
of employees, etc.) to understand the customer's needs in more depth.
For example, specific engineering centers dedicated to marketing may
assist the sales support function by providing traffic engineering studies.
These studies identify system usage patterns. Engineering centers may
also provide presale design assistance to determine the customer's unique
telecommunications requirements. Any subsequent negotiation that
results from this activity (such as a change in the order) is also the
responsibility of the marketing organization.
When the marketing organization, a service center, or a retail sales
location completes a negotiation with a customer, a written service order
implements the agreement. The service order, which describes the equipment or service to be provided and contains customer information,
authorizes the various departments to do the work needed to complete
the order. Customer information may include items such as the customer's name and address and billing name and address.

Order Generation
Service orders reflecting the customer's needs are formatted and entered
into computer-based service-order processing systems that edit and validate the orders, distribute the order information to the appropriate work
groups, and track the overall order process. Since processing of service
orders involves the interaction of several departments within an operating company, tracking and coordinating the actions of all the work
groups involved ensures that the service provisioning occurs by the
promised date. The service-order processing system supports this tracking function by receiving status reports and reporting when the promised
date is in jeopardy because other critical dates have not been met. Provisioning con.trol begins at order generation (shown in Figure 13-1), and it
continues until the order is completed.
The service-order processing system also maintains a pending-order
file during provisioning to keep track of changes that occur as the order
is being processed. For example, the customer may want to change the
services ordered before the installation date, or a problem in the assigned
facilities may require a new assignment. Incorporating these changes

576

Operations

Part 4

into the pending-order file ensures that the order information is accurate
and current.
Network and Premises Equipment Engineering
This phase of service provisioning is unnecessary for certain orders, and
when it is needed, the details may vary. A fundamental portion of network engineering-assignment-involves choosing a specific local loop
and other items needed to fulfill the service order from inventories of
available network equipment. For complex orders (for services such as tie
lines, foreign exchange lines, and centrex), network engineering provides
a detailed design for the service. (Section 13.3 describes these functions
in more detail.) During this phase, assignment information is added to
the service order.
Complex orders may require the engineering work group to design
customer-premises equipment and/ or circuit equipment, make cost estimates, order equipment and special software, and draw up installation
specifications. These activities may begin with support of the initial
marketing contact and may continue until the service is satisfactorily
provided.
Installation
The hands-on work required to provide service may include central office
work to connect a loop to the switching system, loop work to crossconnect cable pairs in the outside plant between the central office and the
customer's premises, and premises work to install the terminal equipment. The premises work may range from installing a simple telephone
set (which often takes less than an hour) to installing a complex customer
switching system (which may take several weeks).
Since modular telephone sets have become available, the customer has
become more involved in the installation process. Modular sets have
plugs that allow easy installation; the customer can pick up the sets at a
retail sales location, take them home, and plug them into standard jacks,
thus reducing the need for premises visits by an installer.
Order Completion
When the service has been installed and tested and the customer verifies
that it is in good working order, the service order is closed out, and the
billing process is initiated. Any customer training required on the use of
the service begins at this time.
Copies of the completed service order are forwarded to various organizations. Service representatives must have records in order to respond to
future calls from the customer regarding service. The maintenance

Chap. 13

Overview of Telephone Company
Operations

577

organization requires records of the type of service and central office connections to respond to repair calls. The Revenue Accounting Office uses
service-order information for the billing process. White pages and Yellow Pages listings are prepared by a directory services organization from
service-order information, and delivery of directories is arranged. Listing
information must also be provided to the operator-services organization
for the directory assistance function. The activities of these organizations
are described in subsequent sections.

13.2.2 SERVICE ADMINISTRATION
The functions needed to sustain a service once an order is completed
depend on the type of service. While all services require billing and
recordkeeping functions, such functions are much more complicated for
sophisticated business-customer services than for residence-customer services, and some administrative functions are unique to a service type (for
example, collecting coins from public telephones).
The rest of Section 13.2.2 discusses billing and customer switching system administration. Sections 13.2.4 and 13.2.5 discuss service administration functions unique to Public Communications Services and Directory
S~rvices, respectively. Section 13.3 discusses network administration.

Billing and Resolution of Customer Billing Inquiries
The Revenue Accounting Office (RAO) is the operations center responsible for accumulating and processing billing information and for preparing bills. Bell System bills generally contain two types of charges: I-time
charges or a recurring flat monthly charge for equipment and services,
and usage-sensitive charges based on the customer's use of the Bell System network. Information for the non-usage-sensitive portion of the bill
comes from the service order, which specifies the services and equipment
being provided to the customer. Information for the usage-sensitive portion comes from devices that record network usage, such as registers, and
message accounting systems (see Section 10.5).
Billing functions include:
• reading and processing accounting records (primarily on magnetic
tape) that identify calling and called telephone numbers, date, time of
day, and duration of a direct distance dialing call.
• rating the call by calculating the distance between calling and called
end offices and applying the appropriate rate based on distance, date,
time of day, and call duration to determine the charge for the call.
Operator-handled calls and Wide Area Telecommunications Services
calls require specialized processing.

578

Operations

Part 4

• posting the charged calls to the customer's account to accumulate
charges during the billing period.
• incorporating the usage-sensitive and flat-rate charges into a customer's monthly bill.
Customers may pay their bills by mail, at public payment offices, or
through agents such as banks. The RAO credits accounts on receipt of
payment and keeps a record of accounts with unpaid balances. Service
representatives receive lists of delinquent accounts and attempt to collect
un paid balances from customers.
Service representatives also generally handle customer billing
inqUirIes. These inquiries may concern toll charges, charges for
message-rate service, the balance due, or duplicate bill requests. The service representative investigates the inquiry and attempts to reach an
agreement with the customer. An investigation may involve detailed
analysis of the service being provided, current and past bills, and customer calling patterns. If an agreement modifies a bill, the RAO receives
notification for billing record adjustment. (Figure 15-4 is a functional
flow diagram of the Residence Customer Billing Inquiry Process.) For the
larger business accounts, billing inquiries that are directly related to the
types of services ordered are referred to the marketing organization for
resolution.

Administration of Customer Switching Systems
Typically, service provided to most residence and small business customers remains unchanged for long periods of time and then changes only
because of a move, which requires a new installation. In contrast, for
large business customers with more complex communications requirements, demands change frequently. These demands are often met
through the administration of the customer switching system (a PBX, for
example).3 Customer switching systems are designed and installed to
meet current needs with some margin for growth, so the two major
postinstallation activities associated with such systems are rearranging the
system and capacity analysis.
A company reorganization, for example, could require the reassignment of telephone extension numbers. The older customer switching systems generally required a premises visit to make the cross-connections

3 As discussed in footnote 2, limitations on the provision of customer-premises equipment
were placed on Bell operating companies effective January I, 1983.

Chap. 13

Overview of Telephone Company
Operations

579

needed to implement the change. With the newer, computer-based systems, clerks in centralized maintenance centers can perform rearrangements remotely by transmitting the appropriate coding information to the
customer switching system over a dial-up data link.
Most rearrangements are handled by the service-order process, but a
developing trend is to have customers become more involved in
configuring their systems. For example, the customer may enter extension number and feature changes directly (as described in Section 11.2),
reducing BOC involvement in short-term rearrangements and, therefore,
reducing the time required to change a configuration. Equipment
changes still require dispatching a craftsperson to the premises.
Periodic analysis of the load on a customer switching system determines whether the system is adequately handling it. The initiative for
such a study may come from the marketing department or from the customer. When a capacity study is desired, a count of working customer
stations is obtained and the load on the trunks to the central office is
measured. Traffic information required for capacity analyses of
computer-based customer switching systems is provided remotely by centralized operations systems at BOC locations. Such analyses may result in
a recommendation from the marketing organization that the customer
increase or decrease system capacity.

13.2.3 CUSTOMER-SERVICE MAINTENANCE OPERATIONS
Maintenance is divided into preventive maintenance and corrective
maintenance. Preventive maintenance includes routine procedures to detect
potential trouble conditions before they affect service. Corrective maintenance involves several separate, consecutive functions. These are
described below and illustrated in Figure 13-2.
1)

Trouble detection - recognizing that a trouble condition exists.
Customers usually detect troubles in station equipment and some
troubles in loops and report them to the telephone company repair
service.

2)

Trouble notification - alerting craft personnel to the existence and
severity of a trouble condition so that corrective action may begin.
Once the repair service attendant who is informed of the trouble
condition has the associated information, craft personnel are
notified.

3)

Trouble verification - determining if a reported trouble condition
still exists. An interval may occur between trouble notification and
the start of the trouble-location function. Since experience has
shown that many trouble indications are transient, correcting
verified troubles receives first priority.

r+

TROUBLE
DETECTION

~

TROUBLE
NOTIFICATION

r----..

TROUBLE
VERIFICATION

r---+

TROUBLE
LOCATION

..... .....

CENTRAL
OFFICE
REPAIR

r--.

...... ......

OUTSIDE
PLANT
REPAIR

I
I

SERVICE
I
VERIFICATION
I

4

PREMISES
REPAIR

~

Figure 13-2. Customer-service maintenance operations-functional flow.

Overview of Telephone Company
Operations

Chap. 13

581

4)

Trouble location - determining whether the trouble is in central
office, loop, or premises equipment4 so that the appropriate craft
force can correct the trouble. This is often the most difficult and
time-consuming step in the maintenance process.

5)

Trouble repair - on-site repair of the defective unit is sometimes
required. In other cases, the defective unit is replaced with a spare
to correct the problem.

6)

Service verification - after repair is complete, the craftsperson
verifies clearance of the trouble condition.

The sequence of events for a typical customer-detected trouble is:
1)

A customer reports a trouble to a BOC repair attendant.

2)

The attendant asks the customer for the affected telephone number
and a description of the trouble. The repair service attendant is
supported by an operations system that provides information in
real time, such as:
a)

customer name and address.

b)

service status (for example, whether service is disconnected, the telephone number is nonworking, or the
number is affiliated with a telephone answering service).

c)

the date of the last trouble and the number of previous
trouble reports if the current one is not the first.

d)

an appointment time that can be offered to the customer if
a repair visit is necessary. The appointment time takes
into account both the backlog of trouble reports awaiting
dispatch and the size of the craft work force responsible
for the corresponding .repair coverage.

e)

results of any automatic tests (for example, shorted or
open line or receiver off-hook).

f)

information on cable and other equipment failures that are
affecting the particular customer's circuit.

Using this information, the attendant talks with the customer in an
attempt to identify the cause of the trouble. The attendant then
gives the customer a repair commitment time and generates a
trouble report.

4 In the case of special services, troubles may be traced to interoffice facilities.

582

Operations

Part 4

3)

The trouble report is reviewed, and if needed, additional tests are
made to determine the location of the trouble so that the appropriate work force (central office, loop, or premises) can correct it.
(Section 13.3.3 describes loop and central office repair.) Premises
repair generally requires dispatching a craftsperson.

4)

The typical repair sequence is completed when the craftsperson
verifies that the trouble condition has been corrected, and the customer has been notified and is satisfied.

Current trends in maintenance operations include more customer
involvement and remote testing and maintenance. Modular sets have
made it possible for the customer to unplug a defective set and bring it to
a customer service center for immediate replacement. As of May 1981, 51
percent of 11 millions home phones experiencing troubles in a I-year
period were being brought in for repair in association with a Defective
Equipment Replacement Program (DERP). Troubles range from a
burned-out bulb in a PRINCESS telephone to a damaged cord and a defective dial. Since 90 percent of the customers who participated in DERP
indicated that they would do so again, DERP participation could increase
from 51 percent to 80 percent without any additional incentive as more
customers are referred to the program. Having customers bring defective
phones to customer service centers reduces the need for craftspeople to
visit customers' premises and is often more convenient and results in
faster problem resolution.
Remote testing of computer-based customer switching systems (made
possible by operations systems) helps in diagnosing faults and eliminating false dispatches. It also allows routine maintenance to be performed
remotely so that troubles can be detected and corrected before they affect
service.
13.2.4 OPERATIONS RELATED TO PUBLIC TELEPHONE SERVICE

Because the general public has access to the equipment, operations related
to public telephone service differ from those associated with residence
and business customers.
Provisioning
A specialized sales force handles sales of public telephones. As with
other services, a service order authorizes the installation of a public telephone. A unique part of the installation may be the need to order and
5 Out of a 100 million installed base.

Chap. 13

Overview of Telephone Company
Operations

583

install telephone booths. Booth installation is often subcontracted outside
the Bell System.

Coin Collecting and Counting
Coins must be collected from public telephone stations before the boxes
are too full to accept any more. However, if collection is scheduled too
often, collection costs will increase. Therefore, computer-based operations systems aid collection by preparing lists of coin boxes that are candidates for collection, taking into account location and projected activity.
The coin collection organization collects the coins, counts them, and
enters the collection data (for example, time, amount, location) into the
operations system. Discrepancies between actual and expected revenue
are reported to a security group that investigates them and reports potential security problems.
, Routine station inspections are also performed during collection, and
out-of-service or hazardous conditions are reported immediately to the
maintenance force. In addition, expanded collection duties require the
collector to replace the directory if it is missing, outdated, or damaged.
In some cases, directory replacement can be contracted outside the Bell
System at lower cost.

Maintenance
A separate repair force is responsible for maintaining coin telephones.
The public, Traffic Service Position System6 operators, collection forces, or
booth-cleaning personnel may report troubles. In addition, suspected
trouble conditions are referred to the maintenance group when analysis
of ongoing collection reports and coin deposit refund data indicates
abnormalities. The maintenance group analyzes the trouble information
received, dispatches the proper repair people when necessary, and submits booth repair orders to outside contractors as required.
The repair people dispatched to a public telephone site also perform
other functions during their visits. They will replace directories if
required and notify the collection force to make a special collection if
they find a full coin box.
Public telephones must operate in a harsh environment. Heavy customer use and occasional misuse or abuse, exposure to the weather, and
exposure to accidental damage (for example, being struck by an automobile) all contribute to the cost of maintenance. Vandalism is not a major

6 Section 10.4.1 describes Traffic Service Position Systems.

584

Operations

Part 4

problem, although some sites may suffer a high incidence. Total maintenance expense amounts to more than $250 million a year.

13.2.5 DIRECTORY SERVICE
In 1980, the Bell System published roughly 2130 different directories.
Directory operations include white pages compilation, directory advertising sales, Yellow Pages production, directory printing, and directory
delivery. The directory business is largely unregulated, and telephone
company operations have evolved in different ways to take advantage of
local opportunities. Operations systems, for example, may be supplied by
vendors. Outside contractors may provide support to some portions of
directory operations such as directory advertising sales.
Service-order information is used to add, delete, or change the listings
in the white pages. Usually, the updated listing is stored and maintained
in an operations system. Like service-order activity, this updating activity
goes on daily throughout the year. Directories for a telephone company
are published annually in a staggered manner, so that during any given
week, directories for some areas are being prepared for the printer. Publishing the directory involves extracting and photocomposing listings
from a data base.
Often, directory assistance operators use records from the white pages
support system as a source of listings. Directory systems can produce
paper or microfiche copies of listings for manual directory assistance
operations or magnetic tape for computerized operations.
A major portion of directory work is the production of Yellow Pages
directories. Annually, several months before each directory publication,
sales operations for directory advertising will begin. A sales campaign
starts with contracts containing the previous year's advertising together
with any changes resulting from service-order activity. Sales people
either visit or telephone directory advertisers and negotiate the details of
advertisements with them. Display advertisements and art work are produced separately from the listings.
Yellow Pages production operations combine the directory advertising
with listings to produce complete pages. An operations system maintains
Yellow Pages listings in much the same way as white pages listings are
maintained. Listings are photocQI!1posed according to heading, and pages
are laid out so that listings and advertising for each heading appear
together.
Printing and distributing telephone directories is an enormous task.
In 1980, the Bell System distributed about 275 million copies. Directory
printing is usually done under contract by large commercial printers.
Although Directory Service personnel maintain the delivery records,
agents are hired to deliver the directories. The Directory Service organization provides the delivery agents with delivery route lists specifying

Chap. 13

Overview of Telephone Company
Operations

585

the number of directories to be delivered to each address (usually based
on the number of telephones).

13.3 NETWORK-RELATED OPERATIONS
Like customer-related operations, network operations consists of the same
three broad functional areas: provisioning, administration, and maintenance. The following sections discuss each of these areas and describe
many activities that make up network operations and the relationships
among them.
13.3.1 PROVISIONING

Network provisioning provides the types and quantities of network elements in the configurations needed to ensure economical, high-quality
service. Network-provisioning functions can be grouped into three
categories. Figure 13-3 shows how they are related in terms of the lead
time each requires.

• Planning - fundamental (long-term) planning for changes and growth
in network structure and forecasting of the quantities and types of
network elements required to provide service; current (short-term)
planning to revise forecasts to reflect the actual evolution of the network structure.

• Engineering -

specifying network elements and their configuration,
accounting for unplanned, near-term needs.

• Implementation -

installing telecommunications network components
in response to the specifications of the long-term and near-term planning activities.

Planning. Starting with long-range forecasts of demands for existing and
new services, fundamental planning organizations develop plans for the
long-term evolution of the network. They evaluate the economics of
current alternatives and decisions and identify the sequence of projects
that should be implemented during the next 20 years (see Section 4.5).
The resulting view of the future network provides, in part, the basis for
shorter-range forecasting activities and provides information to be used
in the ongoing engineering and implementation of the network.
Current planning is more precise than fundamental planning. It usually provides estimates of network elements that will be required 4 to 6
years in the future based on historical traffic load data, growth projections, and an understanding of the present and projected structure of the
network. The results of current planning are provided to other planning
organizations. For example, they are the basis for engineering activities

Operations

586

Part 4

that determine network changes needed to ensure timely and highquality service.
Engineering. The objective of the engineering function is to specify network components and manage network investments in accordance with
the plans and forecasts developed in the planning organizations.
Engineering also takes into account near-term needs not fully anticipated
by planning. Some specific objectives of the engineering function are:
• to determine the appropriate types and quantities of network elements
• to configure the network elements
• to design central office equipment and interoffice facilities and write
specifications for the facilities and equipment
• to evaluate new products, technology, and services to be implemented
in the network
• to develop plans for implementing new equipment and services.
PAST
(LEAD-TIME REQUIREMENT)

PRESENT

,....---------------....A-----------'r-l
P r------,
L
FOREA
CAST
N
N '--_--J
I
N
G

,

INITIATED

FUNDAMENTAL PLAN

~I

INITIATED
4-6

BY:F~:E

CURRENT PLAN

... 1

....-------------I...~·I

- - - - - - - - - - - - - - - - - - - - - ,1
E
N
G

I
N
E
E

R
I

N
G

2-5
YEARS I- - - BEFORE
ENGINEERING STARTED ON: ,
• NEW SWITCHING OFFICES
1
• NEW FACILITY ROUTES
• MAJOR UPGRADES OF
SWITCHING OR FACILITIES'

.1

I

INITIATED

2
YEARS
BEFORE

I

~
~

M
E
N
T

~

6
N

.,1

I------.t
ENGINEERING ..... "
STARTED ON:

FACILITY }
EQUIPMENT
GROWTH

1
L
_-1

Figure 13-3. Network-provisioning activities and lead times required by each.

Implementation. The implementation aspect of provIsIOning can be
divided into two parts: establishing network capacity and servicing the
network 7 to maintain that capacity. Establishing network capacity means
7 Network servicing is traditionally considered an administration activity. However, it is so
closely coupled to the overall process of providing network capacity that it is discussed in
the provisioning section.

Chap. 13

Overview of Telephone Company
Operations

587

installing additional equipment and facilities in response to the
specifications of the planning and engineering organizations. Following
installation, the new equipment and facilities are tested and made available for service as added capacity.
The objective of network-servicing implementation activities is to
meet service demands by rearranging or connecting existing equipment
and facilities. Service demands are originated by customers and by internal operating company activities (such as estimations of short-term trunk
group size changes or routing alterations needed to resolve identified network capacity problems). For example, some types of network servicing
may be done on a seasonal basis: It may be necessary to reconfigure
equipment and facilities to provide more trunk capacity at a ski resort
during the winter. In the summer, the configuration can be rearranged
to provide trunk capacity elsewhere. Network-servicing activities can
also be stimulated indirectly by rate and tariff changes requiring the
installation or modification of billing equipment.
Provisioning the telecommunications network is a continual, complex
set of interrelated activities. It is the performance of various disjointed
tasks by separate organizations in several disciplines and the combination
of the results of these tasks into a single plan.
Trunk, transmission facility, and switching equipment provisioning;
operator-services provisioning; and common-systems provisioning are
examined more closely in the following paragraphs.

Trunk Provisioning
The objective of trunk provisioning is to ensure that the proper numbers
of trunks are provided where and when they are needed. Each stage of
this provisioning process is coordinated with the planning, engineering,
and resource implementation activities performed for facility provisioning (described later).
Long-range planning for trunk provisioning estimates how a portion
of the traffic network-usually a metropolitan network or a numbering
plan area (see Section 4.3)-will be configured 5 to 20 years in the future.
(In this context, long-range planning does not include facility planning,
which determines geographic routes and types of facilities to carry
trunks.) Planning the topology of the network is based, first, on projections of traffic loads. This topology includes the numbers, types, and
locations of the switching systems and the homing arrangements (see
Section 4.2.1) whereby final trunk groups are established between lower
and higher levels of the hierarchy. Then, the sizes of the other trunk
groups in the network are estimated using the available projections of
traffic load between points of the network. The resulting long-range plan
embodies various traffic routing rules, such as whether local and toll
traffic will flow through the same tandems in a metropolitan network and
what sequences of alternate routes are to be used. Such planning does

588

Operations

Part 4

not involve commitments to spend money; its purpose is to ensure that
the long-term consequences of current decisions are foreseen and that the
evolution of the network proceeds smoothly and economically.
The current planning8 interval, in contrast, extends from about 1 year
to 4 or 6 years in the future. Six years is the approximate lead time
required to establish a new switching location, including acquisition of
land and construction of a building. The trunk forecast for each year
specifies the circuits that will actually be needed for the next year's busy
season to meet network objectives. The forecasting process prescribes the
number of trunks required between switching systems. Two different
approaches to trunk forecasting are in use: the trunk-based method and the
point-to-point method (see Section 5.5).
Based on a careful forecast of traffic loads for each year, the current
planning process determines the traffic routing rules for the network and
increasingly detailed decisions about the functions of all the switching
systems and lays out the trunk traffic network according to a detailed set
of engineering rules. Given the trunk forecast and current facility and
equipment plans, engineering groups generate work orders, and requisitions for the installation of the appropriate equipment and facilities are
generated. The equipment and facilities are then installed and tested by
installation groups.
When the required transmission facilities have been built, the
appropriate circuits have been established, and the busy season has
arrived, it is still unlikely that the network as engineered will exactly
match the offered traffic. Thus, measurements of actual usage and counts
of calls offered and overflowing are used to determine where to add or
subtract circuits to meet the actual demand. A few trunks are generally
removed from one group and added to another in various plac~s, so that
all final groups meet their objective of I-percent blocking. The activity of
determining what rearrangements to make is known as trunk servicing.
Transmission Facility and Switching Equipment Provisioning
The aim of transmission facility and switching equipment provisioning is
to ensure that transmission facilities and switching equipment are available to meet the growing demand for communications services at the
lowest possible long-term cost. Each stage of this provisioning process
functions in coordination with the planning, engineering, and resource
implementation activities performed for trunk provisioning just
described.
Wire center studies are a good example of such provisioning activities.
New wire centers must be planned well in advance to meet the growing
8 Section 14.2.1 describes current planning in more detail.

Chap. 13

Overview of Telephone Company
Operations

589

demand for service. Typically, several alternative locations for the wire
center may be proposed, and the serving area boundaries for each proposed location are determined for a particular time in the future. Then,
year-by-year plans for establishing each alternative are developed. Each
of the plans is actually a sequence of placements, removals, and rearrangements of facilities. The construction activity and capital expenditures of the plans are compared. Next, a detailed report of the most
economical wire center plan (including customer assignments, loop costs,
and trunk costs) is prepared. This plan must also be compatible with
other factors, such as the availability of land for new buildings or floor
space in existing buildings. Last, detailed growth patterns are developed
by considering timing of cable additions, structural constraints, timing of
transitions from previous methods of operation to new methods, sizing of
central office equipment subject to economic constraint, and trunk network sizing. In this way, a comprehensive plan is produced for the
efficient and economical coordinated growth of cable and switching
equipment in the exchange area.
The engineering and implementation stages of transmission facility
and switching equipment provisioning were covered in the section on
trunk provisioning because they use information from trunk forecasts.

Operator-Services Provisioning
The operator-services facility administration and force management functions (see Section 13.3.2) provide the data that are used for operatorservices planning, engineering, and implementation. Operator-services
facilities engineering must provide adequate switching capacity and the
proper number of operator positions to meet expected demand. As in all
other operating company engineering projects, the goal is to specify
equipment quantities to minimize capital and installation expense while
meeting service objectives.
Operator-services switching systems are planned and engineered
under assumptions much like those for local and toll switching systems.
However, position engineering (the number of operator equipment positions) and operator staffing, or forcing (the number of operators), employ a
blocked-calIs-delayed queuing model, normally the Erlang C model (see
Section 5.3.2). Operator-services traffic engineering is based on traffic
measurements. These data are used to determine if service is adequate, to
evaluate operator-services efficiency, and to provide engineering data
used in planning for future needs. The primary traffic service measurements for operator systems are .speed in answering the customer and
efficiency in serving the request. (Efficiency is usually measured by the
average time spent on a request.)

590

Operations

Part 4

Operator-services implementation consists of those activities through
which the operator force and its support systems are physically established. This latter activity entails installing operator-services equipment
and facilities to provide sufficient traffic capacity to meet future demand.
Efforts to meet operating force requirements today and in the future
include automation, consolidation, and control of demand. Computerbased systems such as the Automatic Intercept System (see Section 10.4.2)
and the Traffic Service Position System (see Section 10.4.1 and Figure
10-16) have been developed to automate portions of the operator's job to
decrease the work time per call.
Consolidation provides three benefits: It accommodates the large
traffic volume needed to justify the initial cost of computer-based systems;
it raises efficiency by substituting large groups of servers (operators) for
dispersed smaller groups (therefore, service levels can be maintained
while the productivity of the operators increases); and fewer operators
are needed to provide night coverage.
The major approach used to control demand has been to charge higher
rates for services requiring operator assistance. Another way is to
improve network performance, thereby reducing traffic from customers
having difficulty placing their calls.
Common-Systems Provisioning
Common-systems9 provisioning ensures that adequate power, interconnection, and environmental support are available for switching equipment and transmission facilities over a long-term service lifetime (about
20 to 30 years).
Common-systems provisioning for a particular network equipment
location includes plans for: power, distributing frames, the cable entrance
facility, equipment space, and cable pathways. Figure 13-4 shows typical
common-systems provisioning activities. (For discussion purposes, the
activities are assigned step numbers that correspond to the circled
numbers on the figure.)

Step 1 -

Common-systems planners receive facility and equipment plan
proposals from fundamental network planners. Plans for common systems are usually specified after the plans (or major alternatives) for facilities and equipment have been defined.

Step 2 -

Common-systems planners consult with building planners and
equipment and facility engineers about feasible alternatives. Common
systems support a wide variety of transmission and switching equipment.

9 Chapter 12 discusses common systems.

Figure 13-4. Common-systems provisioning activities.

At any time, much of this equipment is undergoing modernization and
growth; hence, common-systems planning requires careful analysis of
forecast data and plans from other areas.

Step 3 - Common-systems planners generate a common-systems alternatives study that includes economic evaluations. Typically, alternative
plans for common systems involve varying amounts of rearrangement,
addition, or replacement of existing capacity. Alternatives can be compared using present-worth evaluation of associated capital and expense
items (see Section 18.3). Selection of an alternative is not based on
common-systems economics alone, but on overall costs for various equipment and facilities alternatives.
Step 4 -

Common-systems planners consult with fundamental network
planners who generate a fundamental plan that includes commonsystems plans. The common-systems plan summaries form a section of
the fundamental plan for a telephone equipment building.

Step 5 -

Common-systems planners develop a comprehensive plan that
provides engineering details for the selected alternative. The comprehensive plan has a separate section for each of the common systems. It constitutes a detailed strategy by which common systems can be evolved
from their existing condition to some planned "ultimate" configuration at
the end of the planned service lifetime.
591

592

Operations

Part 4

Step 6 - Implementation of common-systems plans is the responsibility
of engineering organizations such as equipment engineering. Implementation engineers place orders with suppliers for common-systems equipment and arrange for its installation. When implementation cannot follow the comprehensive plan, implementation engineering may request a
variance-an amendment to the plan made after approval by a commonsystems planner. For example, a comprehensive plan may have been
established for a central office, but added service requirements resulting
from an unexpected increase in business activity may require
equipment/facility upgrading that was not included in the original
comprehensive plan.
13.3.2 NETWORK ADMINISTRATION
Network administration consists of ensuring efficient use of existing and
planned components of the telecommunications network to provide preestablished levels of service to the customer. A variety of organizational
structures may exist for performing the network administration job, but
they have the same primary role-responsibility for the overall quality of
service given by a specific portion of the telephone network.
The responsibilities for network administration are divided among
seven major functional areas:
• data administration
• operator-services administration
• equipment utilization
• office status evaluation
• service problem analysis and corrective action
• transition management
• network management.
Data Administration
Data administration includes scheduling, recording, posting, and checking the validity of traffic and equipment data required to administer,
evaluate, and engineer the switching system(s) and the associated trunking network properly. These data are generally obtained using
automated data systems,10 although manual register reading and recording on film may also be used. Data administration also involves reviewing the operation of installed measuring devices, establishing controls to
10 These systems are part of the Total Network Data System (TNDS) described in Section
14.3.1.

Chap. 13

Overview of Telephone Company
Operations

593

ensure proper data collection, analyzing trouble-detection reports, and
resolving any problem affecting data collection.

Operator-Services Administration
In the area of operator-services administration, two functions meet daily
operating requirements. The first function, ensuring that sufficient operator facilities are available and working properly, is called facility administration. The second function, ensuring that sufficient operators are available, is called force management. Because of the high costs involved in
providing operator services (in terms of equipment, facilities, and operator work force), both functions are closely monitored by Bell operating
company management.
Both facility administration and force management start with the collection of traffic-usage and call-volume data from operator-services equipment and facilities (such as a Traffic Service Position System). Some of
the traffic data and measurements are used by maintenance forces to
maintain operator-services equipment and trunks. Operator services does
not have a separate maintenance force but relies on regular central office
maintenance forces.
Operator-services facility administration consists of analyzing volume
and traffic-usage data in order to determine how well equipment is being
used, to find potential maintenance problems, and to ensure proper load
balance between operator offices. Facility administration also provides
the data used for operator-services planning and engineering.
Operator force management is the continuous job of providing
enough-but not too many-operators to meet established service levels.
Operators work "tours" of from 5-% to 8-% hours. Given the volume
(number of calls) and usage datal! (the types of calls), a forecast of operator requirements is made by quarter-hours for the week 2 weeks ahead.
These requirements are used to create a schedule of available tours and to
allocate the tours to the administrative units that make up the serving
team. The allocated schedule tells each manager how many operators to
provide for each quarter-hour and which tours the operators are to work.
Individual operators are then assigned to specific tours. Each day, a force
manager monitors how well the planning anticipated the actual demand
and makes adjustments as required. The adjustments take the form of
projecting demand from intraday trends and calling out additional operators to meet unexpected demand or releasing an appropriate number of
operators if demand is lower than anticipated. Fine tuning is practiced
by adjusting lunches and reliefs and by rescheduling training or clerical
functions to counter peak or slack periods.

11 Usage data are similar to data used in facility administration.

594

Operations

Part 4

Equipment Utilization
This aspect of network administration ensures that installed equipment is
being utilized in the best way possible and that forecasts of equipment
and trunk additions are consistent with projections of future capacity
requirements. The administrator assesses how much each piece of equipment is being used (loading), evaluates the distribution of that use across
components of the same type (balancing), and maintains up-to-date equipment, assignment, trunk group, and routing records. A variety of inputs,
such as load-balance 12 data, forecasts, and engineered office capacities, are
used to assign equipment and facilities to maintain objective levels of service as demand changes.

Office Status Evaluation
Another network administration responsibility is the daily analysis of the
switching office status, including the integrated review of key data items
on customer usage (CCS 13 per main station, call rates, holding times,
etc.),14 equipment loads, and traffic volumes. Overall office and component group capacities must be determined for daily review of load
versus capacity. Growth sizing and scheduling are monitored to ensure
that sufficient switching capacity will be available to meet forecast
demand.

Service Problem Analysis and Corrective Action
Network administration encompasses responsibility for the identification
(either through detection or reports of trouble), investigation, and resolution of service problems through daily surveillance of central office
equipment and connecting trunk groups. A service problem is defined as
any condition in which established service objectives are not met, with or
without justification.
Some specific activities in this area include:
diagnosing causes of service problems by evaluating service impairment indicators (such as excessive dial-tone delay) and coordinating
with network service organizations to resolve problems
• coordinating with corrective action plans and providing frequent
evaluation and updates of predicted service problems.

12 Section 5.3.4 discusses load balancing.
13 Hundred call seconds.
14 Sections 4.4.1 and 5.2.1 discuss these concepts.

Chap. 13

Overview of Telephone Company
Operations

595

Transition Management
Transition management covers the analysis of plans for equipment additions, replacements, removals, and/or rearrangements. The administrator
must evaluate the impact of this type of activity on service and ensure
that procedures for transition will result in the desired equipment
configurations with minimal equipment outages and service deterioration.
Determining the maximum allowable quantity of equipment (by type and
group) that could be removed from service for maintenance and/or transition activity during additions, rearrangements, or replacements is also a
network administration responsibility. During transition, monitoring
must be performed for the customer load, overall service, equipment
outages, and the trunk and routing network to ensure that they adhere to
the transition plan. Monitoring is also done to avoid, or at least limit,
adverse service effects of the transition.

Network Management
As mentioned in Section 5.6, network management is the function that
keeps the network operating near maximum efficiency (as defined by
completed messages per unit of time) when unusual traffic patterns or
equipment failures would otherwise cause network congestion and
inefficiency. Network management involves the use of controls 15 to alter
the normal network routing structure to respond to overloads or to anticipate them. The trend has been to automate these control actions as much
as possible and to supplement the automatic controls with centralized
computer-based systems. These systems provide status displays and
reports that allow network management personnel to make decisions concerning when to implement manual controls and which controls are most
appropriate. Figure 13-5 illustrates the type of information shown on a
status display used by network management personnel.
A Network Management Center may be local, regional, or national.
The local Network Management Center has responsibility for up to
forty-eight switching offices and is supported by the Engineering and
Administrative
Data
Acquisition
System/Network
Management
(EADAS/NM).16 There are eight regional Network Management Centers
in the United States. Each is responsible for a specified large geographic
area encompassing the control jurisdictions of several local centers and is
also supported by EADAS / NM. The national Network Operations
Center, located in Bedminster, New Jersey, is supported by the Network
Operations Center System and is responsible for coordinating interregional network management controls.
15 Section 5.6 discusses network management controls.
16 EADAS/NM is part of TNDS, which is described in Section 14.3.1.

CCIS
STATUS

C

\I..-----TRUNK GROUP STATUS _ _ _ _ _- ' 1 ' - - - SWITC~i~~uS;STEM--1

Figure 13-5. Typical information on a wallboard status display for
network management. The portion of a larg e wall display shown uses
colored indicators for various status information . At the top, status of signal
transfer points and other elements of the common-channel int eroffice
signaling network is shown . Trunk groups are in co lumns by region (only
three are shown) and grouped vertically according to where (which region)
the group terminates. Associated indicators show whether trunk groups are
in a normal or overload condition and what controls are in effect. At the right
are toll switching systems, with colored indicators for measures of
congestion and the status of controls implemented on traffic t o or from each
system .

Although day-to-day network management is handled best by
automatic controls, manual network management controls are also needed
to handle unusual situations requiring flexibility and human judgment.
A fundamental network management function performed by people is
preplanning. Preplanning develops a plan of controls for handling anticipated overloads such as those that occur, for example, on Mother's Day
(intertoll congestion) or as a result of a localized natural disaster (focused
overload).!7
17 Section 5.6 discusses overloads.
596

Chap. 13

Overview of Telephone Company
Operations

597

13.3.3 MAINTENANCE OF THE NETWORK
As discussed in Section 13.1, maintenance involves both preventing
potential troubles and correcting existing ones. Preventive maintenance,
as noted in Section 13.2.3, includes routine procedures, such as testing, to
detect and eliminate potential trouble conditions before they affect service. It also includes such activities as lubrication, cleaning, and adjusting of equipment.
On a functional level, corrective maintenance procedures for customer
and network operations are very similar and involve the same sequence
of events. However, unlike customer service troubles, most network
troubles (for example, switching system or transmission facility troubles)
are detected automatically.
Every telecommunications system is planned and designed to achieve
an economic balance between system reliability, preventive maintenance,
and corrective maintenance. A system can be designed to have very high
reliability and, thus, require very little maintenance. However, a less
reliable system-that is, a system that is less expensive but requires more
maintenance support-might be more economical. Also, maintenance
planning organizations at Bell Laboratories interact with various development areas during the design stages for new equipment and maintenance
systems. This interaction ensures that new network components are
designed with ease of maintenance in mind and are introduced with a
viable maintenance plan.

Network Service Centers
Systems for switching system maintenance, trunk maintenance, and carrier system maintenance are each designed to support one portion of the
telecommunications network. To ensure that the overall network provides good service, Network Service Centers have been established.
Their major inputs are reports of troubles encountered by operators,
either directly or through customer reports, and reports from other Bell
System employees. The reports are analyzed for patterns pointing to
specific problems in the network (such as a large number of reports
resulting from attempts to reach a certain location). These problems can
be referred to the local repair forces. To aid in the analysis of trouble
reports, computer-based support systems18 have been developed to sort,
format, forward, and examine trouble reports from the entire country for
standard errors. These systems also manipulate data for analysis and
return suspected trouble patterns to Network Service Centers in each
operating company.

18 Such as the Network Operations Trouble Information System (NOnS), the Network
Service Center System (NSCS), and the Automated Trouble Reporting System (ATRS).

Operations

598

Part 4

Figure 13-6 shows the relationship between Network Service Centers
and the loop, switching system, trunk, carrier system, and special-services
circuit maintenance functions. The following sections discuss each of
these functions in more detail.

NETWORK SERVICE CENTER
• OVERALL SERVICE MONITORING
• CENTRALIZED TROUBLE PATTERNING

........
LOOP

SWITCHING SYSTEM

CARRIER SYSTEM

TRUNK

....

SPECIAL-SERVICES
CIRCUIT

• TROUBLE CORRECTION
• PREVENTATIVE MAINTENANCE
• TESTING

Figure 13-6. Network maintenance functions_

Loop Maintenance
The customer is the major source of trouble reports 19 in loop equipment.
Because a customer location is permanently associated with a particular
loop, the customer will always be affected by a loop trouble and will be
aware of the trouble until it is corrected. Therefore, prompt correction of
loop problems is extremely important.
In loop maintenance development, the emphasis has been on improving the efficiency of craft forces through mechanized aids and better testing facilities. Surveillance systems have been developed for automatic
line-insulation testing and for cable-pressure monitoring to detect incipient trouble conditions before customer service is affected. However, once
a customer loop problem has been identified and located by the various
19 Section 13.2.3 describes the operations involved in processing a customer trouble report.

Chap. 13

Overview of Telephone Company
Operations

599

systems in use, it is still necessary, in many cases, for a repair person with
a portable test set to go to the main distributing frame or out on a cable
route to pinpoint the specific trouble location.

Switching System Maintenance
In most central office buildings, craft personnel perform a variety of tasks
including maintenance, rearrangements, and cross-connects (see Section
12.3) on frames. In very large buildings with several switching systems,
however, the assignment of craft personnel dedicated only to maintenance tasks is justified. In such cases, these same people may have
maintenance responsibilities for transmission and signaling equipment
located in the building as well. In contrast, for small switching offices
that are not staffed for one or more shifts per day, system alarms indicating trouble conditions or failures are transmitted to a centralized location
to initiate maintenance action.
The maintenance procedures associated with electromechanical and
electronic switching systems are substantially different. Because they
experience minor trouble conditions more frequently, electromechanical
switching systems require more extensive periodic maintenance. Electronic switching systems are relatively trouble free, but when they do
fail, the trouble symptoms are usually more difficult to diagnose, and
some failures can affect a large number of customers.
Because of their basic reliability and because their maintenance is
automated, electronic switching systems do not afford the craft force
much practice on difficult trouble analysis and location tasks. The
number of highly skilled craftspeople capable of diagnosing certain electronic switching system troubles is limited, making centralized maintenance and self-diagnostics a necessity. For electromechanical systems, the
reasons for centralization have not been so compelling.
With centralized maintenance for switching systems, the majority of
the maintenance functiqns are performed remotely from a suitably
equipped Switching Control Center (SCC).20 Regularly scheduled on-site
maintenance for individual central offices can often be reduced, and
necessary on-site work can be performed on a dispatch basis from a central pool.

Trunk Maintenance
Before direct distance dialing became available, a customer making a
long-distance call had to speak to an operator first. The operator would
then set up the connection and speak to operators at the other end. If a
20 Section 14.3.2 describes the Switching Control Center System, which supports SCCs for
electronic offices.

600

Operations

Part 4

particular trunk was not operating properly, the operator could identify it
and report it. Today, most connections are established without the aid of
an operator, and the particular trunks used by a customer are not
identifiable once the customer hangs up even if the customer reports the
trouble immediately. Consequently, an important source of trouble
reports on trunks has been lost. As a result, trunk maintenance today
depends primarily on routine testing of trunks, generally done automatically by one of several computer-based systems. 21

Carrier System Maintenance
Carrier system maintenance is handled quite differently from trunk
maintenance. Combining many circuits on a single carrier facility necessitates the use of sophisticated alarm and surveillance techniques to
ensure that service to a large number of customers is not impaired if a
failure occurs. In addition, periodic testing and maintenance are usually
performed to ensure that the carrier systems are operating properly.
However, with newer, more reliable solid-state systems, the trend is
toward reducing the amount and frequency of periodic maintenance.
In analog carrier systems, alarms indicating trouble conditions are
activated when control and operating signals22 fall outside a predetermined range. The alarm signals are displayed in the central maintenance
location responsible for that system. In addition, certain analog systems
are monitored at intermediate repeater points. In the event of a failure in
a repeater station, an alarm 23 that identifies the station is transmitted back
to the central maintenance location. The central operator can then
request, via electrical signals (telemetry), an examination of the station
for a limited amount of detailed information about the failure.
Maintenance operations for digital carrier systems are generally centralized within a geographic region. A computer-based system 24 has been
developed to support the trouble identification and location functions. In
addition, a new generation of maintenance features is being implemented
to provide automatic equipment surveillance alarms and automatic backup lines (a spare maintenance line between repeaters where the failure
occurred) within a carrier system. Thus, for digital systems, continuous

21 The Centralized Automatic Reporting on Trunks (CAROT) system is an example.
22 These are also called pilot tones and are located between standard channel frequency
allocations.
23 The Telecommunications Alarm Surveillance and Control (TASC) System.
24 The T-Carrier Administration System (TCAS).

Chap. 13

Overview of Telephone Company
Operations

601

monitoring of system performance has generally replaced periodic
maintenance.
A major difference in the monitoring of analog versus digital carrier
systems is that a direct measurement of system performance is available
for the digital systems. The performance of a digital carrier system can
be monitored by computer-based operations systems to measure the bit
error rate 25 while the carrier system is in service. A report can be made
when the quality becomes degraded. With analog carrier systems, performance is determined indirectly by measuring a number of impairments
such as random noise and impulse noise.

Special-Services Circuit Maintenance
Special services require circuit layout, transmission, or signaling features
in combinations that are not found in the PSTN. (Chapter 3 gives some
examples.) These configurations present a particular maintenance problem in that diverse maintenance abilities are required to handle the particular characteristics of each configuration.
One important concept in special-services maintenance is the use of
consolidated equipment arrangements. For example, the voice-frequency
facility terminal concept, presented in Section 9.2.2, consolidates
transmission, signaling, and circuit maintenance access functions in one
equipment assembly, eliminating the need for intermediate distributing
frames and complex office wiring. The design and structure of such
arrangements incorporate maintenance objectives from the start. The use
of functional plug-in units makes the initial engineering and subsequent
maintenance of the circuits much simpler.
Another major concept is the use of computer-based systems26 through
which personnel can access and perform sophisticated tests on most types
of special-services circuits. In particular, with these operations systems, a
special-services circuit can be tested by one person, eliminating the time
and expense associated with providing personnel at both ends of the circuit to be tested. One-person testing is a Bell System goal attainable
through the use of the current technology. Remote digital access for testing of special-services circuits is provided by the Digital Access and
Cross-Connect System (DACS), discussed in Section 9.4.3. 27

25 A measurement of the number of bits that are perceived to be incorrect at the receiving
end.
26 An example is the Switched Access Remote Test System (SARTS).
27 Martz and Osofsky 1982 discusses the impact of DACS on operations.

602

Operations

Part 4

Conclusion
The activities discussed in the network operations section of this chapter
are crucial to keeping the network operating at acceptable standards so
that customers are provided with consistently high-quality service. However, while the provisioning and administrative activities are completely
transparent to customers, customers may interact with maintenance personnel to some degree (for example, when there is trouble on a line).
Maintenance operations of the future will detect service degradation for
repair before a service-affecting trouble becomes apparent to the customer. This improved detection will result from increased automation of
maintenance activities.

AUTHORS
B. R. Barrall

J.

A.
L. L.
B. S.
J. A.

DeMaio
Desmond
Eldred
Fitzgerald

E. J. Kovac
R. E. Mallon

H. I. Rothrock
P. F. Wainwright

14
Computer-Based Systems
for Operations

14.1 INTRODUCTION
In the mid-1960s, computer technology had developed to the point where
it could be applied to operations in many ways other than in traditional
accounting functions. Over the years, the term operations systems has
come into use to cover these applications.
Operations systems have a variety of objectives and corresponding
benefits. Sometimes, the primary objective is to provide analysis support
for Bell operating company (BOC) personnel, and the benefit is a more
thorough, accurate, and timely analysis capability. Most systems contain
a data base, and in some cases, mechanization of the data base is the primary objective, relieving telephone company personnel of the need to
maintain certain records manually. Both of these benefits provide
expense savings. In many cases, the main benefit of the system is in
reducing capital investment for additional equipment by providing more
accurate records of what capital equipment is already available to meet a
current need. 1 A related benefit provided by some systems is assistance in
planning the use of new capital investment to meet the various equipment needs in a timely, cost-effective manner. In addition to mechanizing manual tasks, operations systems facilitate the introduction of network capabilities and services, some of which might not be feasible if
planning, administration, and maintenance were manual functions.
Operations systems have yielded a considerable reduction in clerical
and recordkeeping personnel. This is partly due to the mechanization of
tasks and partly due to personnel economies resulting from the centralization of work groups that some operations systems allow. These systems
1 This type of benefit is referred to as equipment and facility recovery.

603

604

Operations

Part 4

relieve employees of tedious and repetitive clerical tasks, freeing them for
functions that require judgment and personal interaction, such as:
• establishing an acceptable level for the risk that an inventory of spares
will be depleted
• forecasting growth in number of subscribers in a particular area and
in their calling characteristics
• shipping, receiving, and inserting plug-in units.
Table 14-1 shows the growth expected in the use of mechanized systems between 1981 and 1985. The BaCs use over 400 such systems. The
development of these systems involves Bell Laboratories, AT&T, Western
Electric, the BaCs} and non-Bell System organizations, such as
Tymshare, McDonnell Douglas Automation, and National CSS.
Most operations systems are designed to run on computer systems
owned or rented by the operating telephone companies. Most of the
smaller systems are designed for Digital Equipment Corporation or
Hewlett-Packard computers; most of the large ones are designed for
1BM3-compatible or UN1VAC4 computers. A small number of operations
systems run on central, time-sharing systems. Access to the time-sharing
systems is usually obtained through dial-up, switched facilities.

TABLE 14-1
MECHANIZATION OF HOC OPERATIONS
1981

Employees (thousands)
Terminals (thousands)
Minicomputers
Large mainframe computers
Terminals per employee

841
114
4200
315
0.14

1985
(Est.)

Growth
(%)

886
180

5
60

5800

40

500
0.20

NOTE: The 1985 estimate does not account for the effects of divestiture on operations.

2 Section 14.5 discusses the roles of these organizations.
3 Registered trademark of IBM Corporation.
4 Registered trademark of Sperry Corporation.

60
50

Chap. 14

Computer-Based
Systems for Operations

605

Sections 14.2 through 14.4 describe a few representative systems to
give a general idea of their functions, operations, and benefits. The systems described fall into three categories:
• recordkeeping and order processing
• equipment surveillance, maintenance,

administrati~n,

and control

• planning and engineering.
Section 14.5 discusses development and support of operations systems.

14.2 RECORDKEEPING AND ORDER PROCESSING
Some of the largest scale operations systems are used for recordkeeping
and order processing, massive tasks for operating telephone companies.
The following sections describe three such systems. The Trunks
Integrated Records Keeping System and the Plug-In Inventory Control
System support network operations related to growth and change in the
network by providing an accurate record of circuits and components in
use and available for use. The Premises Information System supports
customer-related operations by providing rapid access to information concerning equipment located on a customer's premises.
14.2.1 TRUNKS INTEGRATED RECORDS KEEPING SYSTEM (TIRKS)

The Role of TIRKS in Circuit Provisioning
In recent years, the scope and complexity of circuit prOVIsIOning has
increased substantially. Large growth and huge Bell System investment
in equipment and facilities have required a typical operating company to
administer millions of pieces of equipment, facilities, and circuits and
process thousands of orders. The complexity and interdependence of
records needed by the network has increased because of technological
advances, such as digital switching systems, and intelligent network elements, such as the Digital Access and Cross-Connect System (see Section
9.4.3). As a result, manual recordkeeping systems have become inadequate. TIRKS was developed to mechanize the circuit-provisioning process. It is deployed in eighteen BOCs to mechanize two aspects of circuit
provisioning: daily circuit provisioning and current planning (see
Figure 14-1).5

5 Section 13.3.1 discusses provisioning for network operations.

BUSINESS
SERVICE
CENTERS

CIRCUIT
ADMINISTRATION
CENTERS

DAILY
PROVISIONING
PROCESS

CIRCUIT ORDER
MONITORING
CIRCUIT DESIGN
AND ASSIGNMENT

CURRENT PLANNING PROCESS
•

FACILITY AND EQUIPMENT
PLANNING

•

EQUIPMENT AND FACILITY
SHORTAGES

Figure 14-1. TIRKS circuit-provisioning process.
(Section
14.3.1
discusses the functions of Circuit Administration Centers and Network
Operations Centers.)

Chap. 14

Computer-Based
Systems for Operations

607

Daily provIsIoning consists of processing orders to satisfy customer
needs for special-services circuits and processing orders initiated for message trunks and carrier systems for the public switched telephone network (PSTN). The daily circuit-provisioning process begins at various
operations centers 6 and flows to Circuit Provision Centers (CPCs), which track
the orders, design the circuits, and assign the components using TIRKS.
The CPC prepares work packages and distributes them to the technicians
working in the field who are responsible for implementing the work
packages and for reporting completion of their work.
Current planning determines the equipment and facility requirements
for future new circuits. The current-planning process apportions forecasts for circuits among the circuit designs planned for new circuits. For
example, given a variety of designs for a particular type 7 of circuit, the
number of circuits of that type forecast will be divided among the possible designs.
The same designs are used in both daily provisioning and current
planning, and all the operations are accomplished using TIRKS.
Components
TIRKS comprises five major interacting component systems: the circuit
order control (COC) system, the equipment (El) system, the facility (Ft)
system, the circuit (Cl) system, and the Facility and Equipment Planning System (FEPS). Figure 14-2 shows the relationship between cac
and C1, E1, and Fl.
cac controls telephone company message trunk, special-services, and
carrier system orders by tracking critical dates along the life cycle of an
order as it flows from the source to the Circuit Provision Center and on
to the field forces. It produces daily, scheduled, unscheduled, and ondemand reports to provide management with the current status of all circuit orders. It also provides data to other TIRKS component systems to
update the assignment status of equipment, facilities, and circuits as
orders are processed.
C1 is the heart of TIRKS. Using basic facility information (such as
location and type), it automatically determines the types of equipment
required for a given circuit (circuit design), assigns the equipment and
facilities needed, determines levels at the various transmission level points
(see Section 8.6.1) on the circuit, specifies the test requirements, and
establishes circuit records for the circuits. When the automated design
process is complete, C1 reformats the circuit records into work packages

6 Operating company work centers. See Sections 13.1 and 15.2.1.
7 Data circuits, foreign exchange lines, and message trunk circuits are examples of different
types of circuits.

COC
ORDER TRACKING
REPORTS
ASSIGNMENT STATUS OF
INVENTORY AND CIRCUITS

C1
COMPLETIONS

CIRCUIT DESIGN AND ASSIGNMENT 14-_IN_V_EN_T_O_RY--I~
INFORMATION

F1
FACILITIES

REPORTS

PREPARATION OF WORK PACKAGES

REQUESTS FOR
DATA

FIELD

Figure 14-2. Interactions between TIRKS components in daily provisioning.

needed for the installation of the circuit and distributes them to the field
forces at various work locations. Once the circuit record is established,
Cl maintains the record for future additions or changes.
El and Fl are the two major inventory component systems in TIRKS.
El contains telephone company equipment inventory and assignment
records and pending equipment orders. The records indicate the number
of various equipment types that are spares (that is, available for assignment) and circuit identification for equipment already assigned. Fl contains telephone company cable and carrier (that is, transmission facility)
inventory and assignment records. Cl accesses both the El equipment
inventory and the Fl cable and carrier inventory.

608

Computer-Based
Systems for Operations

Chap. 14

609

FEPS supports the current planning process (see Figure 14-3). The
planning process determines the transmission facilities and equipment
that will be required for new services, which can then be compared to
inventory to determine possible shortages. FEPS helps planners use the
information in the E1, F1, and C1 data bases along with forecasts of
future growth to allocate existing inventories efficiently, to determine
future facility and equipment requirements, and to update planning
designs.

C1

PLANNING
DESIGNS
INPUT

PLANNED
INVENTORY

EXISTING
INVENTORY

PLANNING
DESIGNS
UPDATE

FEPS
SUPPORTS CURRENT PLANNING

FACILITY AND
EQUIPMENT
REQUIREMENTS

PREDICTIONS
OF SHORTAGES

ALLOCATIONS
OF EXISTING
INVENTORIES

Figure 14-3. Role of FEPS in current planning.

Interfaces
TIRKS is the master recordkeeping system for the network. Consequently, it must interface with many other operations systems, processes,
and various new intelligent network elements that rely on TIRKS for
records. A number of these interfaces are provided by the extensive

610

Operations

Part 4

reports that are available from TIRKS; others are on-line. These interfaces are being developed in accordance with an overall operations plan
(see Sections 14.2.2 and 15.4).

Implementation
TIRKS is an on-line system that uses IBM 370-compatible hardware and
direct-access storage devices. Configuration planning is required to determine the computer system equipment needed to support TIRKS. The precise hardware configuration for a given operating company depends on
the company's inventory universe and the volume of activity affecting
that inventory.
Cathode-ray tube (CRT) video display terminals and teletypewriters
(TTYs) are used to enter and retrieve data. On-line information inquiries
and data-base changes use Teletype Corporation DATASPEED Model 40
display monitors (or the equivalent) for input and output. Reports are
automatically distributed to teletypewriters and medium- or high-speed
printers.
Conversion from a large, manual data base to TIRKS is a large undertaking and may take several years in an operating company. TIRKS
conversion systems have been designed to perform a number of interim
functions. For example, they prepare the data base for implementing the
fully automated system and provide early benefits in capital savings as a
result of finding unused equipment and facilities and increasing their
utilization. 8

Benefits
TIRKS benefits BOCs in three areas:

• Improved service to customers. Through its single, integrated data base,
TIRKS serves the total circuit-provisioning process-from initial planning of circuit components through revisions to the eventual retirement of components-which results in more accurate, efficient, and
timely circuit provisioning. Through its order control procedures,
TIRKS can track critical dates and alert management when completions of orders may be in jeopardy.
• Capital and expense savings. Capital savings result from improved planning and more efficient use of equipment and facilities. Expense savings result from the mechanization of Circuit Provision Center operations. TIRKS also eliminates the need to spend time verifying data
8 The ratio of working equipment to working-plus-spare equipment.

Chap. 14

Computer-Based
Systems for Operations

611

from inadequate records. The work force is used more efficiently
because TIRKS issues more accurate work orders.

• Better management control. TIRKS provides an integrated records base
with consistent terminology and format and can generate detailed and
summary reports on demand and on a scheduled basis for different
operating company levels.
14.2.2 PLUG-IN INVENTORY CONTROL SYSTEM (PICS)

The Bell System spends over $1 billion per month to add and replace
equipment. A large portion of this capital investment is for switching
and transmission systems. The physical design of these systems has
changed with time, as discussed in Chapters 9, 10, and 11. Early systems,
such as the No.5 Crossbar and the 701 private branch exchange (PBX),
were configured in large racks and "hardwired" in place. During the
mid-1960s, electronic switching systems and transmission systems, such as
T-carrier systems, that use large quantities of plug-in equipment were
introduced. Little attention was paid to inventory control because of the
rate at which growth was occurring. As a result, most central offices
accumulated significant numbers of spare plug-in units, the existence of
which was unknown to equipment engineers responsible for ordering
maintenance and growth spares. This "lost" spare equipment was generally scattered throughout central offices, often unused even by resident
technicians. New equipment was often requested and ordered from suppliers when the very plug-in equipment needed was nearby. It became
evident that the effective management of inventories of this size required
the use of a mechanized operations system, and the development of PICS
began.
PICS assists operating telephone company personnel with inventory
management and materials management. Inventory managers establish corporate policies for the types of equipment and for equipment utilization,
assist engineering organizations in introducing new types of equipment
while phasing out older types, and set utilization goals that balance service objectives (see Section 5.2.2 and Chapter 16) and carrying charges on
spare equipment. Materials managers work to achieve utilization goals.
They acquire spare equipment for growth and maintenance purposes
(growth spares are used to provide additional service such as the expansion
of a switchi!lg system; maintenance spares serve as replacements for equipment that has failed while in \!se). They also administer a hierarchy of
locations used for storing spare equipment, ranging from large
warehouses to small, unattended transmission system locations.
The following paragraphs describe a system that administers all types
of central office equipment-PICS with Detailed Continuing Property
Records (PICS/DCPR).

612

Operations

Part 4

PICS/DCPR
About 50 million plug-in units are in use in the Bell System. From the
viewpoint of PICS, these plug-in units represent about 100,000 different
types of equipment.
During the design of PICS, it also became clear that a detailed investment record of equipment by categories would be needed for regulatory
purposes. It was quickly recognized that an equipment inventory was a
prerequisite for a detailed investment or property record. The DCPR portion of PICSjDCPR serves as a detailed investment data base supporting
telephone company accounting records for all types of central office
plug-in and "hardwired" equipment.

Implementation. PICSjDCPR provides software, data bases, administrative procedures, and workflows to accomplish its goals of increasing utilization, decreasing manual effort, and providing a detailed supporting
record for a telephone company's investment. First, PICS jDCPR establishes two new functional entities in each company: the Plug-In Administrator (PIA) and the central stock. The PIA becomes the materials manager,
responsible for acquiring equipment, distributing it as needed to field
locations, repairing it, and accounting for it. To assist the PIA, the
central stock is created. The central stock is a warehouse where spare
equipment is consolidated and managed. To implement PICSjDCPR, a
telephone company must create a new management-level job (the PIA),
provide a clerical staff, set up a central stock, and provide a computer
environment for the system software. The PIA then recalls all excess
plug-in units. Som~, but not all, units are returned to the central stock.
The PIA then conducts complete physical inventories of the central stock
and all field locations, returning excess spare equipment from the field to
the central stock. At this point, the plug-in inventory is established, and
the PIA uses on-line terminal programs to order, receipt, bill, move,
repair, retire, and track equipment. A mechanized link between the PIA
and the central stock provides warehouse personnel with instructions for
shipping or receiving stock from the field.
As part of the inventory process, cost data based on average yearly
prices are supplied. These data make up the DCPR for plug-in equipment. The data are supplied by equipment engineers for both plug-in
and hardwired equipment on a going-forward basis9 as it is purchased. A
portion of the system provides an annual reconciliation between the
DCPR and the company books to ensure that the DCPR remains accurate.

9 Because of difficulties in identifying and pricing older hardwired equipment, no physical
inventory occurs. Equipment details are captured on a going-forward basis from the start
up of PICS/DCPR.

Chap. 14

Com puter-Based
Systems for Operations

613

Subsystems. PICS/DCPR software comprises five subsystems: the plugin inventory subsystem, the inventory management subsystem, the
plug-in DCPR subsystem, the hardwired DCPR subsystem, and the
reference file subsystem.
The plug-in inventory subsystem maintains order, repair, and inventory
records for all types of plug-in equipment. It spans seven processing
areas:
• acquisition (order control, logging, cancellation, equipment receipts)
• movement (between locations; connections and disconnections, for
example, from maintenance spare to growth spare)
• maintenance activity (logging defectives, tracking repair orders, junking plug-in units)
• circuit order control (to interface with circuit-provisioning systems
like TIRKS)
• inventory adjustments (to compensate for shortages and surpluses)
• "back-order" control (to track items on back order)
• miscellaneous (scan/search of inventory data base; printing of shipping information at central stock, etc.).
The inventory management subsystem provides the PIA with mechanized
processes to assist in these tasks:
• recommending levels for maintenance spares based on the rate at
which equipment is returned because of failure
• recommending when orders should be made based on historical
demand, known scheduled growth, or both
• monitoring supplier performance so that proper order lead time may
be taken into account
• reporting on equipment overstock or understock by location and
equipment type.
Specific models and algorithms exist for these activities for both central stock and field locations. At the PIA's request, the subsystem will
shift stock from overstocked locations to those that are understocked or
back-ordered.
The plug-in DCPR subsystem provides the processes required to maintain investment records for plug-in units, including:
• daily processing (loading cost data from purchases, updating investment as a result of movement, adjusting for shortages and surpluses)

614

Operations

Part 4

• billing adjustments and corrections
• monthly processing (to reconcile plug-in equipment billing between
the DCPR and accounting, to notify accounting of changes in investment because of plug-in equipment movement and retirements)
• annual processing (to reconcile the DCPR with the company books)
• miscellaneous (on-line display/search capability, auditing records for
tracking the source of investment change, etc.).
The hardwired DCPR subsystem maintains detailed accounting records
for hardwired central office equipment. Processing is analogous to that of
the plug-in DCPR subsystem, except that it is equipment engineers,
instead of the PIA organization, who supply the input. Because of
difficulties involved in identifying and pricing older hardwired equipment, no physical inventory occurs. Equipment details-records of the
various parts of larger systems, such as an electronic switching systemare captured on a going-forward basis from the startup of PICS/DCPR.
Investment for prior years is maintained on an undetailed basis; that is,
for equipment in existence before the startup of PIes /DCPR, total equipment investment is collected together by year for recordkeeping purposes. Another distinction between the plug-in and hardwired DCPR
concerns accounting rules. Hardwired equipment may not be moved
from location to location without a complicated accounting process.
Plug-in units, by nature of their mobility, were exempted from this
process by the Federal Communications Commission (FCC) by special
agreement.
The reference file subsystem provides and maintains reference data used
by all the other subsystems, including:
• equipment reference data, such as the J-number and list number from
the J-specification for the equipment, common-language equipment
identification codes, and availability designations 10
• location reference data, such as address, telephone number, and
common-language location identification code designation-a lO-character
code name that identifies any physical location within the Bell System
• DCPR reference data, such as account code and equipment category
number (for both plug-in and hardwired equipment)

10 J-numbers and the list numbers associated with them identify equipment. The nature of
systems or equipment units is described in J-specifications, which also explain
applications of equipment units or systems. Common-language equipment identification
codes identify equipment by family (for example, T-carrier), subfamily (for example,
channel units), and type (for example, a specific channel unit).

Chap. 14

Computer-Based
Systems for Operations

615

• on-line security data that prevents movement of equipment between
unauthorized locations
• audit reference data, such as the date and time a data base was
updated and the terminal that originated the update. (These data
track changes made to all reference data bases.)
PICS jDCPR subsystems are developed and maintained by Bell Laboratories. Both on-line and batch programs are contained in PICSjDCPR.
Software packages containing executable program code and ancillary data
needed to run the system are periodically released to the telephone companies. PICSjDCPR runs on IBM-compatible computing equipment
operating with the IBM Information Management System data-base
manager with teleprocessing options.

Interfaces. PICS jDCPR is an interdepartmental system. Interfaces
exchange data with organizations within a company in the areas of property and cost accounting, division of revenues, depreciation accounting,
and service cost accounting. Interfaces have also been provided to TIRKS
and other circuit-provisioning systems, to Bell Laboratories for the ongoing update of reference data, and to other BOCs that jointly own equipment. Initially, the interface between PICSjDCPR and TIRKS was
designed for manual operation, where plug-in requirements are forwarded to the PIA concurrent with the issuance of a TIRKS work order.
Ultimately, the process will be automated, and the systems will interact
with each other so that the specific plug-in equipment item will be
recorded on the work order before issuance to the field.
Benefits. All of the Bell System has now implemented PICSjDCPR.
Significant economic benefits arise as a result of reduced capital investment needs as spare equipment is located and effectively reused.
Obsolete equipment is junked and eliminated from a company's investment base. These actions increase the utilization of plug-in equipment.
Many companies have improved their utilization from approximately 70
percent to 90 percent or more with PICSjDCPR. By mechanizing many
actions that were previously done manually, staff has been either reduced
or made more productive. Secondary benefits include faster provision of
equipment, a basis for asset verification (the DCPR), and improved record
accuracy.
Various economic modeling techniques allow estimation of
PICSjDCPR's economic benefits. The total benefit over the period 1969
through 1990 is estimated at $1.1 billion. l l (This figure includes Bell
Laboratories development costs.)
11 Cumulative discounted funds flow (see Section 18.3.3) based on a January 1980 study.

616

Operations

Part 4

14.2.3 PREMISES INFORMATION SYSTEM (PREMIS)
Each BOC has thousands of service representatives. One of their functions
is to respond to requests for telephone service from residence customers.
To ensure customer satisfaction, each transaction must be completed as
quickly and as accurately as possible. Speed and accuracy are also important to an operating company since a faster, more accurate transaction
means lower costs for both customers and the company.
Developed by Bell Laboratories in cooperation with AT&T, PREMIS
provides the service representative with fast, convenient access to the
information needed to respond to service requests. The need for a
PREMIS-like system grew in the mid-1970s when efforts were being
directed toward reducing operating costs by decreasing residence installation visits. The installation of modular jacks in residences and the institution of PhoneCenter Stores, where customers could pick up sets, were
both part of this effort. As a result, the need for maintaining records of
jacking arrangements 12 arose and led to the development of PREMIS.
PREMIS was also developed in response to the need for address
standardization-a need that has grown as Bell operating companies
mechanize operations. Computer-based systems are used by the different
departments of a Bell operating company in processing the customer service order prepared by the service representative. Mechanization requires
that the customer address used in records be standardized, because the
service order (with its address information) feeds other downstream systems. Precision in the way that living units (such as houses or apartments) are represented-along with address standardization-promotes
the automatic reuse and reassignment of facilities. This benefit affords
savings to a Bell operating company by reducing the time and labor
needed for assignment and installation of outside plant facilities.
.
PREMIS has three mechanized data bases for use by service representatives in a telephone company:
• address data
• a credit file
• a list of available telephone numbers.
These data bases ensure greater accuracy than the paper records,
microfiche, and other information sources previously used.
PREMIS has one other major feature that serves the Loop Assignment
Center (LAC) rather than the service representative. This feature, called
PREMIS/ LAC, is an extension of the address data base and provides for
the storage of outside plant facility data at each address entry.
PREMIS/LAC is described later in this section.
12 The number of telephone jacks and their locations in a living unit.

Chap. 14

Com puter-Based
Systems for Operations

617

Customer Service Tasks and PREMIS Applications
PREMIS helps service representatives handle the following transactions:
• new service-a customer wishes to initiate service
• relocation of service-a customer wishes to disconnect service at the
existing address and initiate service, perhaps with changes, at a new
address. Service may exist or may have previously existed at the new
address for another customer.
These transactions involve several tasks: determining the customer's
correct address, negotiating service features, negotiating a service date,
checking the customer's credit status, and selecting a telephone number.
Some of these tasks provide information needed for the preparation of
the service order that will be used by other departments in the telephone
company to provide service. PREMIS support for these service representative tasks is described below.

Determining the Customer's Correct Address - The major feature of PREMIS
is its address-related and address-keyable information. When an order
for new or relocated service is being taken, the customer gives the service
representative the new address, which the service representative then
keys into PREMIS using a video display terminal, as shown in Figure
14-4. Address-related information is stored in the address data base (see
Figure 14-5).
When an input request does not contain an accurate or complete
address, PREMIS displays information that can be used to query the customer. For example, in a transaction with a customer who requests service for a residence at 125 Main Street, PREMIS would prompt for more
information if it had data for both 125 East Main Street and 125 West
Main Street. This prompting capability ranges from a list of street names
that partially match the input to a list of apartment numbers for a specific
address. In each situation, PREMIS provides selected information
designed to direct the service representative's attention to the specific
problem with the input. 13
When the address matches information contained in the address
data base, PREMIS responds with the full address (community, state, and
zip code) and information about the geographic area that is needed on
the service order. This includes wire center, exchange area, tax area,
directory group, and the service features available for that area. This
same display will also include the existing or previous customer's name
and telephone number, the modular jacking arrangement at the address,
13 As a further example, for a request for service at a new residence, PREMIS would
compare the address provided by the customer to its data base of valid addresses (streets
and numbers).

Figure 14-4. Service representatives with video display terminals for PREMIS.

UN I VA C COMPUTER

LOOP ASSIGNMENT CENTER
ADDRESS DATA
OUTSIDE PLANT
DATA

CUSTOMER

RESIDENCE
SERVICE CENTER

AVAILABLE
TELEPHONE
NUMBERS

REQUEST FOR SERVICE
CREDIT FILE

DATA BASES

Figure 14-5. PRE MIS data bases.

ASSIGNMENT
DATA

Chap. 14

Computer-Based
Systems for Operations

619

and an indication of whether a connected outside plant loop from the
address back to the central office was left in place.
If the previous service at the address was discontinued, the reason for
disconnect and the disconnect date are also displayed. This helps identify customers who are trying to re-establish service at the same address
when previous service was disconnected for nonpayment.
Before PREMIS was developed, service representatives had to use multiple paper sources to obtain the data PREMIS now provides in the
address data base. Without PREMIS, the service representative typically
would obtain address information from a book called the Street Address
Guide (SAG); modular jack information from a microfiche file; other information from paper records (previous disconnects, for example); and some
information, such as available service features, from the service
representative's handbook.
Through its address-keyable data base, PRE MIS offers the following
advantages:
• increased address accuracy on orders for new service because a
mechanized SAG can be kept more accurate and up to date. This accuracy reduces installation delays and correction costs that ripple
through every other mechanized system fed by the service order.
• reduction in the number of installer visits. The modular jacking information available from PREMIS is more complete than the file of
disconnect service orders that it usually replaces. With PREMIS, the
service representative has jacking information available for a new customer at a given address whether or not the existing customer at that
address has placed an order to disconnect the service. The capability
of locating unnumbered addresses also helps reduce installer visits.
Formerly, installer visits were required to locate the terminal bOX.14
With PREMIS, if a new customer can supply the previous customer's
name or telephone number, a service representative may be able to
identify an unnumbered living unit and the records of its associated
facilities.

Negotiating Service Features - When a service representative uses PREMIS
to check an address, PREMIS also indicates the service features that can
be sold at that address, providing useful information for discussing these
with a customer.
Negotiating a Service Date - PREMIS also indicates whether the outside
plant loop back to the central office has been left in place. IS If so, an
14 Outside plant where pairs of cables are accessible by installers.
15 When outside plant has been left in place, more detailed information is stored in the
address data base, although the service representative does not use this information.
This information is used by the Loop Assignment Center.

620

Operations

Part 4

installer will not be needed to connect service, and an earlier installation
date can be offered to the customer. Thus, additional revenue is generated for each additional day of service, and customer satisfaction is
increased.

Checking a Customer's Credit Status - Although a service representative can
use the address-keyable data base to determine if service has been disconnected at a residence for nonpayment, PREMIS also maintains a namekeyable file of customers with outstanding debts to the telephone company, whose service has been disconnected. The service representative
enters the new customer's name into PREMIS for comparison with the
credit file. If there is a customer in the credit file with the same name,
information on this customer is displayed, including the address where
the disconnect was made and the customer's social security number (for
identification). Thus, at the time that the order is taken, the service
representative can determine if the new customer has an outstanding
debt and can require payment of the past bill and a deposit before installing service. Before PREMIS was developed, many telephone companies
maintained a paper file of disconnect service orders. All new service
orders were checked against this file, but often not until service had
already been established. Although the elimination of the paper file provides clerical savings, prevention of potential loss of revenue is the major
benefit.
Selecting a Telephone Number - PREMIS contains a file of available telephone numbers, from which the service representative requests a telephone number for a specific address. Formerly, each day, clerks in a
central location prepared paper lists of available telephone numbers and
distributed them to service representatives.
Not only does PREMIS reduce the errors that result when paper
records are used for recording and assigning telephone numbers, but it
also reduces the total number of telephone numbers that need to be available in an area. With PREMIS, many residence service centers or service
representatives can access the same centralized list. Before PREMIS,
different lists of available telephone numbers were prepared for each
location to avoid assigning the same telephone number to different
customers.
PREMIS reads the available telephone numbers from a magnetic tape
supplied by the Computer System for Mainframe Operations
(COSMOS)-a computer system developed by Bell Laboratories that stores
the full inventory of telephone numbers. For locations where COSMOS
is not used, the numbers can be input by a clerk using a maintenance
transaction.

Chap. 14

Computer-Based
Systems for Operations

621

PREMIS/LAC Data
Another major feature of PREMIS is the mechanized dedicated plant assignment card (DPAC) capability used by the Loop Assignment Center. Each
operating company LAC maintains records of addresses where the outside
plant loop facilities (for example, cable pairs and terminal boxes) are
dedicated (left permanently in place). These records are organized and
accessed by address; thus, the DPAC data can be added to the PRE MIS
address data base, and the paper DPAC file can be eliminated. Outside
plant data may be maintained in PREMIS for a line that is currently
working at an address or for a nonworking line where the outside plant
loop has been left in place for reuse by the next customer. This mechanized capability is more accurate than the paper records it replaces, and
it reduces both errors and the outside plant assigner's time. In promoting
the reuse of facilities, this feature also reduces the number of installation
visits.
System Characteristics
PREMIS is an on-line interactive system whose prime users are service
representatives interacting with customers. During peak hours on peak
days, thousands of service representatives may be interacting with the
system at the same time. Therefore, PREMIS has fast response-time
requirements and a particular need for simplicity and clarity of input and
output.
The PREMIS system has a centralized data base and uses a large mainframe computer-the UNIVAC 1100 series. Depending on the size of the
Bell operating company, either one or two computers will serve the
entire company. The terminal used by the service representative is a
DATASPEED teletypewriter Model 40/4 or 4540. The service representative who uses PREMIS is often accessing other systems as well, so
PREMIS is designed to share the terminal with these other systems via a
communication network. To free the PREMIS computer for heavy on-line
usage, update of the data base is done overnight (in batch) with selected
data from each day's service orders. The data are extracted from the telephone company's mechanized service order distribution system and supplied to PREMIS on a tape.
PREMIS was piloted in South Central Bell in January 1979, and by
mid -1983, it was being used by nine BOCs.

14.3 EQUIPMENT MAINTENANCE, ADMINISTRATION,
AND CONTROL
The previous section described systems concerned with recordkeeping.
This section describes systems that gather and process data from operational network components (such as switching systems and transmission

622

Operations

Part 4

facilities) to improve the administration, maintenance, or control of the
network or its components. The systems in this category provide information to support a number of operating company functions, including:
• long-range and current planning for growth and rearrangement of the
network
• ordering and installing equipment and facilities
• adding, deleting, and rearranging circuits to meet changing traffic
demands
• managing the flow of traffic in the network to control overload
conditions
• monitoring the performance of transmission systems, switching systems, and trunks and facilitating maintenance by improving the
reporting and control of equipment problems.
The Total Network Data System, described in Section 14.3.1, is actually a large and complex set of coordinated systems. It supports a broad
range of operating company activities that depend on accurate traffic data.
The Switching Control Center System, discussed in Section 14.3.2, supports surveillance, maintenance, and control activities for network switching systems and transmission terminal equipment. This system illustrates
the advantages of combining a computer-based system with the concept
of centralized maintenance.
14.3.1 TOTAL NETWORK DATA SYSTEM (TNDS)
The Bell System communications network contains over 11,000 switching
systems that are connected by over 7.5 million trunks arranged in 400,000
separate trunk groups. Both the general growth of telecommunications
and the replacement of older switching and transmission systems by
newer, more versatile ones necessitate a construction program for the Bell
System that exceeds $3.5 billion each year just for the trunks of the PSTN.
Because such large capital investment is involved, adequate planning for
the future growth of the trunking and switching networks requires
significant attention to traffic data collection, administration, and
engineering. Measurements of current traffic levels are the basis on
which current network performance is assessed and future growth is
planned. These measurements (see Chapter 5) are made in terms of both
trunk usage and switching system operation. For the average operating
company, these evaluation and planning functions can require the collection, processing, and distribution of over 50 million pieces of traffic data
per week.
TNDS is a family of operations systems that work together to mechanize data gathering and reporting. TNDS consists of both manual procedures and computer systems that provide operating company managers

Chap. 14

Com puter-Based
Systems for Operations

623

with comprehensive, timely, and accurate network information that helps
them to analyze network operation in several ways. TNDS supports
op€rations centers responsible for administration of the trunking network, network data collection, daily surveillance of the load on the
switching network, the utilization of equipment by the switching network, and the design of local and central office switching equipment to
meet future service demands.
TNDS Modules
TNDS comprises several major component systems or modules that operate
at various locations. Modules that collect and format traffic data typically
us€ dedicated minicomputers and are located at an operating company's
computation center, referred to as a minicomputer maintenance center. Special data lines link these dedicated minicomputers to associated measuring equipment or directly to the switching systems located at central
offices. Other TNDS modules generate engineering and administrative
reports on switching systems and on the trunking network of message
trunks that interconnects them. Most of these modules run on generalpurpose computers located in the operating companies, although some
run on AT&T computers centrally located for use by all the Bell operating
companies. Depending on their needs, operating companies may use all
of the TNDS modules or various combinations of them.
TNDS Functions
TNDS modules perform four basic processes: data acquisition, central
office equipment reporting, trunk network reporting, and system performance measurement. Figure 14-6 shows the overall flow of information
in TNDS between the systems that perform these processes, and
Table 14-2 summarizes the component systems.
Data Acquisition. Switching offices-both electronic and electrom€chanical-provide traffic data in terms of peg count, overflow, and
usage measurements. 16 In electromechanical offices, a specialized data collection device, called a traffic usage recorder, scans trunks and other switching components periodically (every 100 seconds) and counts how many
are busy. In electronic offices, the data are collected by the switching
system's central processor. As shown in Figure 14-6, these traffic data are
transmitted by most switching systems to the first of the TNDS
systems-the Engineering and Administrative Data Acquisition System
(EADAS). EADAS is the major data-collecting system and runs on a dedicated minicomputer at an operating company's Network Data Collection

16 Section 5.7.1 defines these traffic measurement parameters.

TRUNK NETWORK REPORTING SYSTEMS

r---------------.

I

I

r-------------I--~~

TSS

I

TFS

I
I
I

I

DATA
ACQUISITION SYSTEMS

I

r-------- -----.
r-----'
I

I

I

I

I
I
I

EADAS
ALTERNATIVE
SYSTEMS

I

L _______________

I
I

I

~

I

I
I

SYSTEM PERFORMANCE
MEASUREMENT SYSTEMS

CENTRAL OFFICE REPORTING SYSTEMS

I

r - - - - - - - - - - - - - - - - - , r---------,

SWITCHING
SYSTEMS

I
I

I

I

I

I
I

I I
I I
LBS

L __

5XB
COER

SPCS
COER

.LJ...

I
I
GSAR

II

I

I

I

I:·
:
I
I
L:: _______ I

I I:

I
I
I
I

II
ICAN

SONDS

I

~----~71~----~
I
L _____________ J IL _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ I
~

SELECTED DATA

FROM OTHER

TNDS SYSTEMS
••••••• PROCESSED INFORMATION
- - - - TRAFFIC DATA
RECORD BASE INFORMATION

=

Figure 14-6. Data flow among TNDS systems.

~

TABLE 14-2
TNDS AT A GLANCE
System

Acronym!
Abbreviation

Specific Function

Computer System

Mode of Operation

Engineering and Administrative
Data Acquisition System

EADAS

Collects traffic data from switching offices;
provides near real-time reports

BOC dedicated
minicomputer

On-line,
interactive

Engineering and Administrative
Data Acquisition System/Network
Management

EADAS/NM

Provides real-time surveillance and control of
trunk network

Long Lines/BOC
dedicated
minicomputer

On-line,
interactive

Traffic Data Administration System

TDAS

Formats and temporarily stores data for
downstream systems

BOC mainframe
computer

Batch, off-line

Common Update/Equipment

CU/EQ

Common data base shared by several systems

BOC mainframe
computer

Batch, off-line

Load Balance System

LBS

Measures customer traffic load on switching
machines; computes load balance index

BOC mainframe
computer

Batch, off-line

5XB COER

Monitors switching system service; measures
utilization; calculates capacity for No.5
Crossbar

BOC mainframe
computer

Batch, off-line

Stored-Program Control System
Central Office Equipment Reports

SPCS COER

Monitors switching system service; measures
utilization; calculates capacity for lESS, 2ESS,
and 3ESS switch offices

AT&T mainframe
computer

Batch and on-line
interactive

Individual Circuit Analysis

ICAN

Identifies abnormal load patterns on
individual circuits

BOC mainframe
computer

Batch

Small Office Network Data System

SONDS

Collects and processes central office and
trunking data for step-by-step offices

AT&T mainframe
computer

Batch and on-line
interactive

Trunk Servicing System

TSS

Computes trunk group traffic loads and current
trunk requirements

BOC mainframe
computer

Batch, off-line

Common Update Trunking

CU/TK

Common data base shared by trunking systems

BOC mainframe
computer

Batch, off-line

Trunk Forecasting System

TFS

Forecasts message trunk requirements for the
next 5 years

BOC mainframe
computer

Batch, off-line

Centralized System for Analysis
and Reporting

CSAR

Measures performance of TNDS

AT&T mainframe
computer

Batch and on-line
interactive

. No.5 Crossbar Central Office
Equipment Reports

General Function

Data acquisition

Central office reporting

Trunk network reporting

System performance
measurement

Chap. 14

Com puter-Based
Systems for Operations

627

Center. Each EADAS serves up to fifty switching offices. Some companies
use other data-collecting systems that are developed locally or supplied
by vendors. 17 Interfaces are provided so that measurement data gathered
on these systems can be processed downstream. Two large toll switching
systems-the 4ESS system and the No. 4A Crossbar 18 -collect their own
data and do not interact with EADAS or alternative data acquisition systems. They provide their data directly to the TNDS component systems
downstream from EADAS.
When EADAS receives traffic data from switching offices, it assembles
and summarizes the data for processing by the other downstream TNDS
systems, and it processes some of the data in near real time to provide
hourly and half-hourly reports for network administrators.
Network administrators use EADAS reports to determine the quality
of service (for example, in terms of dial-tone delay) and to identify
switching problems, such as failure to complete calls. EADAS also makes
additional real-time information available to these administrators by providing traffic data history that covers up to 48 hours. The data history
provides flexibility (via a module called NORGEN-Network Operations
Report Generator) so that administrators can tailor their requests for
information to determine critical quantities such as dial-tone delay, holding time for certain units of switching equipment, and overflow on
important trunk groups.
EADAS forwards traffic data to three other TNDS systems by data
links or magnetic tape (see Figure 14-6). One of these systems-the
Traffic Data Administration System (TDAS)-formats the traffic data for use
by most of the other downstream systems. The other two systems-the
Individual Circuit Analysis (lCAN) program and the Engineering and Administrative Data Acquisition System/Network Management (EADAS/NM)-use
data directly from EADAS. ICAN is one of the central office reporting
systems discussed in Central Office Reporting Process below.
EADAS/NM also receives data from those switching systems that do
not interface with EADAS. It monitors switching systems and trunk
groups designated by network managers and reports existing or anticipated congestion on a display board at operating company Network
Management Centers. Network management personnel use EADAS/NM to
analyze problems in near real time to determine their location and causes.
For problems that require national coordination, EADAS/NM provides
information to the AT&T Long Lines Network Operations Center at Bedminster, New Jersey. This center is supported by a Network Operations Center
System (NOCS), which performs functions similar to EADAS/NM on a
17 For simplicity, only the term EADAS will be used in describing these acquisition
functions. But the possible use of alternative systems should be understood.
18 No. 4A Crossbar Systems currently in service use a peripheral bus computer (PBC) to
collect data. This is an adjunct to the electronic translator system. (See Section 10.2.5.)

628

Operations

Part 4

national scale. Like EADAS, EADAS/NM uses dedicated minicomputers
to provide users with interactive real-time response and control.
TDAS, the second module in the data acquisition process, accepts data
from EADAS, local vendor systems, and large toll switching systems on a
weekly basis as magnetic tape. TDAS functions primarily as a warehouse
and distribution facility for the traffic data and runs as a batch system at
an operating company's computation center.
TDAS treats its data-acquisition job as a basic order/inventory system,
in which the traffic data collected represent the inventory, and traffic
measurement requests for specific data represent the orders. Traffic measurement requests, prepared manually and sent to TDAS by operating
company personnel, are stored in a master data base called common
update/ equipment (CD /EQ) that shares information with some of the other
central office reporting systems. This shared information for each central
office is necessary to ensure the correct association between recorded
traffic data and the switching or trunking elements being measured.
CD /EQ runs as a batch system in the same computer as TDAS. CU /EQ is
updated regularly with batch transactions to keep it current with changes
in the physical arrangements of the central office switching machines.
This ensures that recorded measurements are treated consistently in each
of the reporting systems that use CD / EQ records.
The traffic data (the inventory) processed through TDAS are matched
against the traffic measurement requests (the orders) stored in CU /EQ.
When the data necessary to fill an order have been received, a weekly
data summary (printed or on magnetic tape) is sent to the personnel
requesting it for use in preparation of an engineering or administrative
report. The availability of summarized traffic data to downstream systems
completes the TNDS data acquisition function. The remaining TNDS systems, shown in Figure 14-6 and described below, help managers analyze
the data that have been gathered.

Central Office Reporting Process. Five TNDS engineering and administrative systems provide operating company personnel with reports about
central office switching equipment. These component systems run either
as batch processes on operating company computers or interactively at a
centralized AT&T (mainframe) computer center (see Table 14-2). The systems are:
• the Load Balance System (LBS)
• the No.5 Crossbar Central Office Equipment Reports (5XB COER)
• the Stored-Program Control Systems Central Office Equipment Reports
(SPCS COER)

Chap. 14

Computer-Based
Systems for Operations

629

• the Individual Circuit Analysis (ICAN) program
• the Small·Office Network Data System (SONDS).
The first three systems receive their traffic data directly from the
Traffic Data Administration System (TDAS). ICAN receives data from
EADAS independently of TDAS and uses the CD /EQ data base for some
of its reference information. SONDS collects its own data from small
step-by-step offices independently of both EADAS and TDAS (as shown
in Figure 14-6).
LBS is a batch-executed system that helps assure the network administrator that traffic loads in each switching system are uniformly distributed. LBS analyzes the traffic data to establish traffic loads on each line
group of a switching system. 19 The resulting reports are used by the N etwork Administration Center to determine the lightly loaded line groups to
which new subscriber lines can be assigned. LBS also calculates load balance indices for each system and aggregates the results for the entire
operating company.
The 5XB COER and the SPCS COER systems provide information on
common-control switching equipment operation for different types of
switching systems. 5XB COER is a batch-executed system that runs on an
operating company's (mainframe) computer, while SPCS COER is an
interactive system that runs on a centralized AT&T mainframe computer.
Both analyze traffic data to determine how heavily various switching system components are used and to measure certain service parameters such
as dial-tone delay. Network administrators use these service and analysis
reports to monitor day-to-day switching system performance, diagnose
potential switching malfunctions, and help predict future service needs.
Traffic engineers rely on these equipment utilization reports to assess
switching office capacity and to forecast equipment requirements. To be
most useful, service and traffic load measurements must be made during
the busiest periods of the day and year. Both 5XB COER and SPCS COER
produce busy hour and busy season 20 reports to meet that demand.
The ICAN program detects electromechanical switching system equipment faults by identifying abnormal load patterns on individual circuits
within a circuit group. These faults, for example, include defective circuits that prevent customer calls from being completed. ICAN produces a
series of reports used by the Network Administration Center to analyze

19 A line group is a collection of subscriber lines that share a concentration module of the
switching system. Section 5.3.4 discusses load balancing, and Section 7.3 discusses the
concept of concentration.
20 Section 5.2.3 discusses these measurement periods.

630

Operations

Part 4

the individual circuits and to verify that such circuits are being correctly
associated with their respective groups.21
SaNDS differs from the other office equipment reporting systems in
that it performs a full range of data manipulation functions. It provides a
number of TNDS features economically for the smaller electromechanical
step-by-step offices. SaNDS collects traffic data directly from the offices
being measured, processes the data, and automatically distributes weekly,
monthly, exception, and on-demand reports to managers at the Network
Administration Centers via dial-up terminals. SaNDS runs on an interactive basis at a centralized AT&T mainframe computer.

Trunk Network Reporting Process. Shown in Figure 14-6 are three
TNDS systems that support trunk servicing and forecasting at the Circuit
Administration Center. They are all batch programs run on an operating
company's computer and include:
• Common updateltrunking (CU ITK)
• Trunk Servicing System (TSS)
• Trunk Forecasting System (TFS).
CU ITK is the data-base system that contains the trunking network
information (such as alternate routes) and other information (such as
trunk group identification and trunks in service) required by TSS and
TFS. Personnel in the Circuit Administration Center update the CU ITK
data base regularly to keep it current with changes in the physical
arrangements of the trunks and switching machines in the central offices.
This updating process includes maintaining office growth information
and a "common-language" (standard) circuit identification (see Section 14.2.2) of all circuits for individual switching machines. Such
maintenance is necessary to ensure that traffic data provided by the
Traffic Data Administration System (TDAS) will be correctly associated
with the proper trunking and switching configuration when it is processed by TSS and TFS.
TSS helps trunk administrators develop short-term plans and determine the number of circuits required in a trunk group. It processes traffic
data from TDAS and computes the offered load for each trunk group, that
is, the amount of traffic that would have been carried had the number of
circuits been large enough to handle the load without trunk blocking.
(Chapter 5 discusses these traffic-related parameters.) Using offered load
on a per-trunk-group basis, TSS calculates the number of trunks theoretically required to handle that traffic load at a designated grade of service.
21 Section 3.2 discusses trunks, special-services circuits, and trunk groups.

Chap. 14

Computer-Based
Systems for Operations

631

TSS produces weekly reports showing which trunk groups have too many
trunks (under-utilization) and which trunk groups have too few trunks
and are performing below the grade-of-service objective. Personnel in
the Circuit Administration Center will use this information to issue trunk
orders that either add or disconnect trunks.
The traffic load data computed by TSS are also used to support the
trunk forecasting function performed by TFS. These data, along with
information on the network configuration and forecasting parameters
stored in the CU /TK data base, are used in long-term construction planning for new trunks. TFS forecasts message trunk requirements for the
next 5 years as the fundamental input to the planning process that leads
to the provisioning of additional facilities.

System Performance Measurement Process. Traffic data flow through
TNDS provides operating company decisionmakers with timely and accurate information so that they can deal with short- and long-term issues.
Because of the complex steps and the number of systems involved in providing the various engineering and administrative reports, a separate
reporting system assesses TNDS performance and identifies potential
problem areas. The Centralized System for Analysis and Reporting (CSAR) is
designed to monitor and measure how well data are being processed
through TNDS. CSAR collects and analyzes data from other TNDS systems. (CSAR does not now analyze data from EADAS/NM, SaNDS, or
TFS.) CSAR provides operating company personnel at Network Data Collection Centers, Network Administration Centers, and Circuit Administration Centers with quantitative measures of the accuracy, timeliness, and
completeness of the TNDS data flow and the consistency of the TNDS
record bases. CSAR also furnishes enough information to locate and
identify a data collection problem, for example, if EADAS receives no
data for several days on some equipment component in a particular 5XB
COER office.
.
In addition to assisting each operating company in monitoring the
overall operation of TNDS, CSAR summarizes the results for that company as input to the TNDS Performance Measurement Plan (TPMP).
TPMP results are published monthly by AT&T.
CSAR runs as a centralized on-line interactive system at an AT&T
computer center. At the conclusion of each run of a TNDS system at the
BOC, data required by CSAR are placed into a special file. At the
appropriate time, these files are merged and transferred to the AT&T
computer center. CSAR performs the proper associations and analyzes
each system's results. Operating company managers access the resulting
information using dial-up terminals from their own work locations.
Reports can be arranged in a number of formats that provide details on
overall TNDS performance or individual system effectiveness. These

632

Operations

Part 4

reports also help to identify and resolve specific problems. Reports can
be arranged to reflect performance on a level as broad as a total company
or as specific as an individual trunk group or switching system. Such
reports enable managers to analyze TNDS operation on an end-to-end
basis, a process that would otherwise involve significant effort to summarize and correlate information across the various system boundaries, if
done manually.

Near-Term Evolution of TNDS
Since its inception in the late 1960s, TNDS has evolved into a large,
mature system meeting operating company needs for traffic data collection and processing. In its early stages, TNDS represented the first set of
tightly integrated operations systems and eliminated the need for the
extensive manual recording, tabulating, and summarizing that had
characterized the traffic measurement, collection, and analysis process
thus far.
New algorithms and better techniques are being developed to improve
the operational and reporting characteristics of TNDS and to keep pace
with advances in the state of technology and with the needs of its users.
Some of these improvements will mechanize the user interface to include
on-line access to the data base and processed results generated by the
various batch component systems. Because TNDS is an integrated system,
the systems-planning effort associated with the continual introduction of
new systems, features, and generic programs served as a forerunner of
the main Bell System operations planning activities. TNDS is now a key
element of the Total Network Operations Plan (see Chapter 15), and its
future evolution is being carefully coordinated through this plan.

14.3.2 SWITCHING CONTROL CENTER SYSTEM (SCCS)
The Switching Control Center concept evolved within the telephone companies starting in the late 1960s. The primary motivation was to centralize the administration, maintenance, and control of the lESS switching
system that was being rapidly deployed within the Bell System. By
grouping personnel performing the same and related functions, centralization results in more efficient use of people as well as other advantages
such as cross-training.
The lESS system, a Stored-Program Control System (SPCS), is far more
sophisticated than its electromechanical predecessors and requires
different training for maintenance personnel. Maintenance of lESS
switching equipment is supported by a master control center (MCC), a
frame of equipment in a lESS system with lamps to indicate the current
state of the office equipment and keys for operating controls (see Figure
14-7). Maintenance data are available at a maintenance teletype and are

Figure 14-7. lESS system master control center.

used by a technician to request diagnostics and to remove and restore
equipment units to service . Because the lESS system has a highly reliable
duplex nature, an office can operate unattended, either entirely or for
substantial parts of each day.
Unattended operation is both feasible and practical primarily because
the lESS system maintenance data can be sent to a remotely located
maintenance teletype. In addition, the functions of the MCC for the lESS
system can be performed remotely using specially designed interface
equipment that works with standard telemetry equipment. These capabilities allow the remote administration, monitoring, and control of lESS
switching equipment from a centralized maintenance center-a Switching
Control Center (SCC).
With teletype channels connected to a centralized location, it is
economically attractive to add a minicomputer system at the SCC to interface with the lESS system since the cost of the computer system can be
shared among several switching systems . This minicomputer system, the
Computer Subsystem (CSS), and the equipment units that remote the MCC
633

634

Operations

Part 4

capability make up the Switching Control Center System (SCCS). The SCCS
has been broadened to handle other SPCSs besides the lESS system.
The CSS can support a number of SCCs. In a typical Bell operating
company application, the computer is located in the minicomputer
maintenance center. A number of SPCSs within the same geographic
area can be supported from the center, so that the centralized work force
can collaborate on maintaining any that need immediate attention. The
use of a centrally located common pool of highly skilled technicians
offers clear economic and technical advantages. Further, when an on-site
visit is needed, for example, to replace a faulty circuit pack, a technician
can be dispatched from the center.
System Operation
Figure 14-8 is a schematic representation of two Switching Control
Centers (A and B) served by one Computer Subsystem and connected to
several Stored-Program Control Systems.

SPCS

-

SPCS

-

CRITICAL
INDICATOR
PANEL

CONTROL
EQUIPMENT

~ J;1

\
V
~ J)
WORKSTATION

•

CONTROL
CONSOLE

SCCA

CSS

•

SPCS

CRITICAL
INDICATOR
PANEL

CONTROL
EQUIPMENT

•

-

WORKSTATION

CONTROL
CONSOLE

SCC B

Figure 14-8. Two Switching Control Centers
served by a Computer Subsystem.

Chap. 14

Computer-Based
Systems for Operations

635

Maintenance and administrative data are sent to the SCCS from each
switching system connected to it. The control equipment provides the
interface for all forms of data received from the switching systems. The
MCC data on equipment status from all connected SPCSs are displayed
on indicator lamps on a wall-mounted critical indicator panel within
view of SCC personnel. In addition, a minicomputer-based control console also receives the MCC data. Using the console, a technician can
operate MCC keys remotely (that is, at the SCCS).
During any day, an SPCS may generate the equivalent of a hundred
or more pages of teletype messages. These data are of various types, for
example, equipment status data, administrative data, diagnostic results
data, data on abnormal conditions, and audit data. The SCCS computer
receives them, logs them on disk, performs a number of operations on
them, and takes various actions depending on the information contained
in the messages. All the received data can be viewed on a workstation
CRT either immediately or later, since a long-term history of the logged
data is kept. In addition to viewing and analyzing the data, a technician
using a workstation terminal can directly communicate with any connected SPCS and can remotely execute any command that is available
locally at the SPCS. For instance, a technician at the SCC can request an
SPCS to run a program to diagnose a piece of equipment, remove or
restore a trunk, etc.
The number of switching offices that may be connected to an SCCS
depends on the size of the offices and the amount of data transmitted.
Typically, about fifteen offices are connected, although it is possible to
handle thirty or more offices. Generally, this number would include a
mix of stored-program control switching system types located within a
geographic area surrounding the SCC. This proximity permits maintenance personnel to be dispatched to an office as needed. Generally, the
goal is to locate the SCC within a half-hour dispatch time from any office
it serves.
The number of offices and corresponding volume of maintenance
activity also determine the number of workstation terminals and control
consoles required for effective operation. Typically, an SCCS may include
fifteen to twenty workstations and two or three consoles. The SCCS
automatically gathers two basic types of real-time data from each monitored office-status data reflecting the state of the lamps on the MCC and
teletype message data (which also appear on the maintenance teletype in
the switching office). These two types of data are transmitted over
separate channels to the SCCS.
If a major alarm is triggered at a switching office, an alarm will sound
within seconds at the SCC. Further, the major alarm will cause an update
of the status of the office on the critical indicator panel and display a
specific description of the alarm condition on a CRT alarm monitor at a
workstation. Whenever an alarm occurs, the computer displays that condition only on the CRT monitor that is assigned to a particular set of

636

Operations

Part 4

switching offices. This assignment of responsibility allows technicians to
work in their areas of specialization.
In response to the alarm, a technician might use a eRT workstation
terminal to access ess software tools to analyze the situation and choose
an appropriate action, which might consist of sending a specific teletype
command from the workstation terminal (through the eSS) directly to the
spes involved. In a situation where an spes system restart is required
or the teletype channel is unavailable, a control console in the sees may
be used to operate Mee keys remotely. Thus, technicians can monitor
and control an office and have the same capabilities from the see that
they would have at the Mee in the switching office.
Added Software Capabilities
In addition to remoting the capabilities that are available locally, the
sees computer provides many additional software tools that simplify and
improve the maintenance and operation of the spess. These software
tools fall into four broad classes: enhanced alarming, interaction with
message history, mechanization of craft functions, and support for
SPCS administration.
Enhanced Alarming. The sees does much more than simply reproduce
the alarms generated locally at an spes. It may also facilitate the use of
the incoming data, where necessary, by generating a failure description
in a form that is easily interpreted. In addition, it provides many realtime analysis techniques. For example, the sees may generate alarms for
conditions that are not detectable by the spes itself (that is, when there
are more than a certain number of messages in a given interval).
Interaction with Message History. After an alarm occurs, additional
information may be needed to determine the appropriate action to be
taken. To ease information gathering, the sees computer provides many
tools to aid in sifting through the large amounts of historical information
related to the spes. For instance, one may use the ess to "browse"
through the history data; filter out information relating to the failure at
hand; sort that information by failure type, equipment unit type, unit
number, etc.; and analyze the results by looking for subtle patterns that
may pinpoint the cause of the problem.
Mechanization of Craft Functions. In some cases, the handling of certain conditions has become so routine that it can be reasonably mechanized. In these cases, the sees computer may intercept alarms and status
messages as they are received and perform functions that were formerly
done manually to analyze and respond to problems. For instance, the
system is capable of responding automatically when a trunk problem is

Chap. 14

Lomputer-HaSed.
Systems for Operations

637

detected. The sees computer can input messages to the spes to remove
the trunk from service and test it via the teletype channel. Depending
on the results of the test, the sees may either return the trunk to service
or print a trouble ticket, with which a technician may be dispatched to the
office to fix the trunk. The process of removing a trunk, restoring it, testing it, and writing the trouble ticket-previously a manual function-is
totally automated by the sees.
Support for spes Administration. Finally, the sees provides tools for
mechanizing and streamlining the basic administration of the spess. For
example, the ess can send spes data to operations centers and other
operations systems, as required, without user intervention. Furthermore,
time-consuming recordkeeping functions like the tracking of spes program changes have been simplified with the sees.
In summary, personnel at the sec, aided by direct access to powerful
software tools, have complete maintenance and administrative control of
the remote switching offices under their jurisdiction. Dispatching a technician to an office is required only when a problem is identified that
requires local maintenance action, such as replacing a circuit pack.
Present Status
The Switching Control Center System was first introduced into the Bell
System in 1974. It worked initially with the lESS system and provided
increased benefits by reducing the size of the maintenance force required;
by improving responsiveness through mechanized functions; and by providing sophisticated, more efficient support tools. The sees received
almost immediate acceptance by the BOes. In subsequent reissues of the
Computer Subsystem software, features were added to centralize the
maintenance and control of the other spes types as quickly as possible.
At the end of 1980, over 200 seess were in operation within the Bell
System supporting almost 300 sees. Using the latest available software
package, the sees can maintain not only the entire ESS system family,
but also the Traffic Service Position System (TSPS), the Automatic Intercept System (AIS), and several other auxiliary processor systems. 22 In
addition, sees now has the capability to support network transmission
terminal equipment.
The number of available sees software features has grown enormously since the first sees was introduced. Today, sophisticated programs are available for many diverse kinds of operations. For example,
within seconds, an sees program can analyze and pinpoint a component
failure in the complex switching network of an ESS system. Without

22 Section 10.4 discusses TSPS and AIS.

638

Operations

Part 4

these programs, the problem might go undetected for a much longer time
because of the sheer volume of messages that would have to be analyzed.

Future Plans
With the economic benefits resulting from the SCC concept and the rapid
deployment of SCCSs within· the BOCs, the future will most certainly
involve expansion of the role of the SCCS. In keeping with the philosophy of centralized maintenance for all switching systems for efficiency
and economy, SCCSs are expected to accommodate new SPCS types as
they are introduced into the Bell System. In addition, development of
additional software analysis programs will certainly continue.
Finally, with the growth of operations systems, there is an increasing
need to share data among them. For instance, data collected by an SCCS
from a particular switching office may also be needed by another operations system or another operations center. Today, the SCCS supports this
data distribution function to some extent. For example, it notifies the TCarrier Administration System (TCAS), an operations system responsible
for T-carrier alarms, of certain types of carrier problems. SCCS also interfaces with the Centralized Automatic Trouble Locating and Analysis System (CATLAS), an operations system that automates trouble location procedures that identify faulty circuit packs in an SPCS when trouble is
detected and diagnosed.
Interaction with other operations functions will grow. For example, a
secure, automatic dial-up capability is being added to the SCCS to provide
direct communication to BOC security bureaus and call annoyance
bureaus. In this way, data about a suspect call detected at a switching
office can be routed automatically to a security center via an SCCS.

14.4 PLANNING AND ENGINEERING
As described in Chapter 13, planning and engineering are extremely
important telephone company functions. They involve several thousand
people in the BOCs whose actions affect the entire construction budget.
Some of the areas for which planning and engineering functions are
required are the overall network; buildings, including power and common equipment; transmission facilities; and switching equipment. The
application of operations systems has provided significant benefits in all
these areas.
This section describes examples of operations systems that support
planning and engineering functions for local switching equipment and
for transmission facilities, respectively-the Central Office Equipment

Com puter-Based
Systems for Operations

Chap. 14

639

Engineering System (COEES) and the Metropolitan Area Transmission
Facility Analysis Program (MATFAP)-and two similar programs.
14.4.1 CENTRAL OFFICE EQUIPMENT ENGINEERING SYSTEM
(COEES)
COEES is the standard system for planning and engineering local switching equipment. It is a time-sharing system that runs on DEC PDP-I0 computers, centrally located and accessed via the computing system vendor's
communication network. Figure 14-9 is an information flow diagram of
the system.
COEES contains component systems for step-by-step switching systems, crossbar switching systems, the I/lAESS and the 2/2BESS switching
systems. 23 Each component system has a different capability. The one
that is the most complete-the component for the 1/ lAESS system-is
described here.
COEES uses a 4-year planning horizon. It is assumed that each year
has a busy season. For each of the four busy seasons, the following data,

TELEPHONE EQUIPMENT ORDER
1 ESS SYSTEM CALL STORE REPORT
NETWORK
DESIGN
ENGINEER

~

EQUIPMENT
ENGINEER

NETWORK
DESIGN
RECOMMENDATION

~

htl
~

DIAL.UP\"

DATALI~K ~

LJ· ·

•

,s

US MAIL

DEC PDP-10
LOCAL
WIRING

....ctJ
OFF-LINE
REPORTS

WESTERN ELECTRIC
REGIONAL
ENGINEERING CENTERS

Figure 14-9. COEES information flow diagram.

23 The necessary functions for the 5ESS system are performed by a similar system called the
Digital Ordering and Planning System.

640

Operations

Part 4

obtained from company forecasts, is stored in the COEES data base for
each local switching office:
• number of lines of all types
• number of trunks of all types
• average call rate per line and trunk
• average usage per line and trunk
• all special features, signaling types, etc., required.
COEES determines the quantity of each type of equipment in the
office needed to satisfy the forecasted load at objective service levels;
determines an estimated price for engineering, procuring, and installing
the equipment addition needed to reach the required level; and then
compares the present-worth costs (see Section 18.3) of each of the eight
different ways of satisfying the office needs (that is, four I-year jobs, a 1year job followed by a 3-year job, a single 4-year job, etc.). It displays all
eight alternatives for the network design engineer to review, including
the present-worth penalty or benefit of each alternative.
An important capability of the system is sensitivity analysis, where network design engineers vary different parameters to determine the sensitivity of the costs to that parameter. The following kinds of changes, for
instance, might be considered:
• The call rate (calls per line per busy hour) increases 10 percent.
• The proportion of lines with Speed Calling increases 20 percent.
• The number of centrex lines increases 15 percent.
The network design engineer may decide to change certain equipment
quantities (such as the number of incoming registers or the number of
intraoffice trunks) depending on the results of the sensitivity analysis.
The final result of the network design engineer's activity is a network
design recommendation.
After the network design recommendation is complete, a telephone
equipment order must be prepared. This task requires additional detail
because some simplifying assumptions that can be made for planning
purposes cannot be made for ordering purposes. The equipment
engineer provides the required detail, and COEES prints out an order,
ready to be mailed to the equipment supplier.
Another important output from COEES is the lESS system call store24
report. This report, prepared at the request of the network design
24 The call store is used to store information needed by the lESS system to process calls in
progress and certain other information that is likely to change with time. (The lESS
system is described in Section 10.3.3.)

Chap. 14

Computer-Based
Systems for Operations

641

engineer and based on inputs provided by that engineer, provides information to the equipment supplier (Western Electric) on the amount of
memory to allocate to the various functions of the call store. For
instance, there are many registers (areas of memory) that provide such
functions as storage of dialed digits or billing information. The appropriate sizes of these registers are a function of the traffic characteristics of
the office.
All BOCs and several non-Bell System companies use COEES. Studies
have shown that it can save about forty person-hours per telephone
equipment order for a typical switching office growth job, compared to
the amount of time that would be spent if the same functions were performed manually. There is also strong evidence that it results in more
accurately sized offices.
14.4.2 FACILITY NETWORK PLANNING PROGRAMS
The Metropolitan Area Transmission Facility Analysis Program (MATFAP) is a computer program that aids in facility planning. Using present
worth of future expenses and other measures, MATFAP analyzes the
alternatives available to an operating company for its future transmission
equipment and facilities. By combining trunk and special-services circuit
forecasts with switching plans, network configuration, cost data, and
engineering rules, MATFAP identifies what transmission plant will be
needed at various locations and when it will be needed. It also determines the economic consequences of particular facility and/ or equipment
selections and routing choices and provides the least-cost assignment of
circuits to each facility as a guide to the circuit-provisioning process. As
its name implies, MATFAP is oriented towards metropolitan networks
and towards equipment and facilities found there, such as voicefrequency plug-in units and cables, digital terminals and multiplex equipment, the Digital Access and Cross-Connect System, and digital carrier
systems such as TI, TIC, and FT3. 25
MATFAP provides two benefits. First, it helps automate the
transmission-planning process which, in terms of data handling alone,
could otherwise impose enormous engineering staff requirements.
Atlanta, Georgia, for example, is a moderate-size metropolitan area. Over
4000 circuit groups must be routed on more than 100 links between sixty
buildings. Then, facilities and equipment must be selected, the cable and
office equipment additions must be sized, and their costs must be calculated over multiple time periods. The hardware possibilities to be considered include several different types and gauges of wire pair cables, a
growing number of digital facilities, and a wide variety of voicefrequency equipment and digital terminals.
25 Section 9.4 describes these systems.

642

Operations

Part 4

Second, MATFAP deals with the entire network and, thus, takes into
account economies that cannot be identified by analyses restricted to
end-to-end circuit studies, link-by-link investigations, or office-by-office
evaluations. One example of this networking advantage is afforded by
the program's minimum-cost routing algorithm, which assigns forecasted
circuits to paths other than the shortest ones if long-term savings in facility expenditures can be obtained. Another example is MATFAP's circuit
segregation algorithm, which determines the cost advantage of digital carrier trunks (see Section 8.6.3) that can serve message trunks terminating
on the 1/ 1AESS system. Because any special-services circuits also provided from such an office cannot use digital carrier trunks, they require
different digital terminals and separate (that is, additional) digital facilities. The overall economics in an office depend on the numbers of each
type of circuit, their rate of growth, and the timing of digital carrier
trunk installations in other offices to which the circuits are connected.
MATFAP accounts for all these factors over the length of the study and
allows the planner to find the best time 26 to install digital carrier trunks in
each office and the best circuit growth policy. MATF AP also balances circuit loads on high-capacity digital lines with the additional multiplex
equipment that may be required for high line fill.
Because of the large amount of input to MATFAP, data are usually forwarded on tape to the computing center where the program is run. Editing can be done remotely, however, using Western Electric's Remote Data
Entry System (RDES).
The Outstate Facility Network Planning System (OFNPS) and the Intercity
Facility Relief Planning System (IFRPS) are similar in function to MATFAP
but are tailored to meet the needs of rural and toll networks, respectively.
For example, OFNPS contains a decision aid that identifies strategies for
introduction of digital facilities in a predominantly analog (N-carrier)
network; IFRPS deals with radio and coaxial cable but not with voicefrequency facilities.

14.5 DEVELOPMENT AND SUPPORT
A major software development effort was begun in 1967 with the formation of the Business Information Systems Programs (BISP) area, first as part of
AT&T and later transferred to Bell Laboratories. This area was given
responsibility for the design of a set of computer systems (then called
Business Information Systems) to support all nonfinancial aspects of the telephone business. Many of these have become operations systems.
From the beginning, Western Electric played a significant role in a set
of operations systems collectively called the EPLANS 27 Computer Program
26 The installation time that results in the least cost over the length of the study; that is, the
installation time that will be least expensive for meeting the need.
27 Service mark of Western Electric Co.

Chap. 14

Com puter-Based
Systems for Operations

643

Service. In recent years, Western Electric has become more involved in
operations system maintenance and user support and began, in 1981, to
assume funding responsibility for some BIS products.
As the number of operations systems grew, it became clear that the
various operations systems should not be considered separately; they
could, and should, have interactions and interfaces. Consequently, about
1975, in coordination with AT&T and the BOCs, Bell Laboratories began
to develop a series of plans for network and customer operations. The
goal of the plans is to establish a framework for the consistent development and application of operations systems for the maximum benefit of
the operating telephone companies. (Chapter 15 discusses the need for
planning, the planning process, and the impact of the plans in more
detail.)
The coordinated design and application approach permits the functions of several operations systems to be interconnected. For example,
when the Total Network Data System (TNDS) detects a need for an additional trunk in a trunk group, it might automatically send a message to
the Trunks Integrated Records Keeping System (TIRKS), which would
assign necessary equipment and facilities and send cross-connect instructions to frame technicians. It could also send an order to the Plug-In
Inventory Control System (PICS), which could print a shipping order for
plug-ins; to the Remote Memory Administration System (RMAS), which
changes translations in the switching systems; and to the Centralized
Automatic Reporting on Trunks (CAROT) system, which arranges for testing the new trunk. The only actions requiring human intervention
would be shipping and installing the plug-in equipment and making all
frame connections.
Similar direct interactions between other operations systems would
allow computers to do all straight-forward, clerical tasks. People would
concentrate on those tasks that require judgment, such as creating forecasts, approving expenditures, and isolating failures; and those that
require manual dexterity, such as inserting plug-in equipment, installing
new equipment, and making cross-connections.

AUTHORS

L. R. Bowyer
Fleischman
L. P. Hawkins
R. E. Machol
S. K. Stearns
H. R. Westerman
W. J. Zide

J. S.

15
Operations Planning

15.1 INTRODUCTION
Operations in the Bell System (see Chapter 13) are influenced by many
factors, as shown in Figure 15-1. These factors are continuously changing, and the operations they influence must respond to the changes. One
way operations are evolving is that, as a result of the introduction of
computer-based operations systems, they are becoming more sophisticated. These systems, some of which are described in Chapter 14, can
make operations more effective and improve customer service. But the
introduction of any operations system affects the related operations.
Specifically, an operations system takes over functions that were previously done manually and, in addition, often performs new or different
functions that were previously not feasible. Consequently, changes must
occur in the functions performed by people and the interactions among
people responsible for different functions.
Operations planning is the key to operating effectively in this changing
environment. Operations planning ensures that changes in the roles and
responsibilities of people are linked to changes in the nature of the telephone business. It further ensures that operations-related functions are
assigned to people and systems in ways that realize potentials for greater
efficiency and better customer service. As the number of operations systems deployed in the Bell System increases, the role of operations planning becomes more important.
This chapter provides a basic understanding of operations planning
and its effect on operations now and in the future. The rest of this section describes early work directed at developing an efficient operations
plan for a specific geographic area of one operating company and progress in developing operations plans for all Bell System operations. Subsequent sections describe the contents of operations plans, how the plans

645

NEW NETWORK
SERVICES

LEGISLATION - - -...."

OPERATING ECONOMICS

BELL SYSTEM
OPERATIONS

. . .- - NEW TECHNOLOGIES

NEW CUSTOMER PRODUCTS
REGULATORY POLICIES

Figure 15-1. Factors affecting Bell System operations.

aid operations in the operating companies, the effect of operations planning on the development of operations systems, and the planning for an
operations systems network to provide efficient transfer of information
between multiple operations systems and between operations systems and
a user's computer terminals.
This chapter describes operations planning as it was conducted in the
Bell System before divestiture. While operations planning is expected to
continue in each of the separated entities, its scope and approach are
necessarily changed to comply with the provisions of the 1982
Modification of Final Judgment.

15.1.1 INITIAL OPERATIONS-PLANNING EFFORTS
Planning for specific operations (for example, efficient collection
schedules for coin telephones) has gone on for years in the Bell System.
The need for planning on a broader scale arose when Bell Laboratories
began to apply computer technology to telephone company operations.
Around 1970, it was recognized that computer technology would profoundly affect operations even as it made those operations more effective.
And, indeed, as the Bell System developed and implemented more and
more operations systems, opportunities for new ways of operating
emerged. To understand and take advantage of the opportunities, AT&T,
Bell Laboratories, and Bell of Pennsylvania undertook a joint study in
646

Chap. 15

Operations Planning

647

1972. The study addressed the question of how to apply new operations
systems to the network most effectively in a specific operating areametropolitan Philadelphia. This was the first time all the functions
necessary for network operations were considered together in an overall
picture.
The formal result of this study was a plan for integrating and applying operations systems to the network operations expected in Philadelphia in 1980. Reorganization of the work force into operations centers
formed a major part of the plan. (An operations center consists of a group
of people, reporting to a common manager, who perform a set of related
functions for a specific geographic area, group of customers, or service.)
All the jobs in each kind of operations center were essential for effective
functioning of the network. The plan described how the people in the
different jobs would work together within and between operations
centers to make the best possible use of the capabilities afforded by operations systems.
A significant finding of the study was that integration of the new
operations systems and centers and their coordinated deployment and use
were essential if the full benefits of computer technology were to be realized for Bell System customers. AT&T informed the Bell operating companies (BOCs) of the study results so that they could perform similar
studies for selected parts of their operating areas.

15.1.2 PLANS FOR BELL SYSTEM OPERATIONS
The Philadelphia study had addressed the application of existing or
soon-to-be-available operations systems in an actual operations environment. AT&T and Bell Laboratories used the experience gained during the
study and the results of the BOC studies to construct a new kind of plan
for the Bell System. The goal was to formulate a fundamental, longrange network operations plan generally applicable to the entire Bell
System-a plan that would define the evolution of operations systems
and centers to meet changing nationwide network operations needs. The
new effort was a major step forward in that it aimed at a synthesis of new
operational patterns and new operations systems capabilities.
The result, in 1978, of three years of detailed interdisciplinary analysis
and planning by AT&T and Bell Laboratories was Issue 1 of the Total N etwork Operations Plan (TNOP). Since then, new issues have been released
periodically to reflect advances in telecommunications technology and
operations systems technology, new network services, and new operations concepts.
In the late 1970s, to complement the planning for network operations,
broad technical planning for customer-services operations was begun.
AT&T and Bell Laboratories produced four additional operations plans
that were interlocked with the plan for network operations presented in

648

Operations

Part 4

TNOP. These four plans presented an integrated picture of operations
systems and operations centers for customer services.
The plans for network and customer-services operations have become
the "road maps" for Bell System operations. AT&T and Bell Laboratories
use them to guide the planning and development of operations systems,
centers, and work methods and as the basis for detailed studies of future
operations system architecture and work force job design.
The BOCs assist AT&T and Bell Laboratories in preparing Bell System
operations plans by participating in field studies, periodic reviews,
resource-sharing programs, and Bell System-wide conferences. The
operating companies then use the Bell System plans as guides in formulating localized operations plans for the configuration of centers and systems in their specific geographic areas.
, Thus, operations planning has become a coordinated, disciplined procedure involving close cooperation between AT&T, Bell Laboratories, and
the BOCs.

15.2 ELEMENTS OF BELL SYSTEM OPERATIONS PLANS
Bell System operations plans include functional descriptions of operations
centers and operations systems and show, by means of operations
processes, how they interact to perform various operations. Operations
processes are detailed accounts of the sequential roles of centers and systems in producing a specific result. The Trunk Addition Process in
TNOP, for instance, describes how systems and centers interact in adding
a trunk to the telecommunications network.
Operations plans are generally presented as both short-term and
long-term views. The short-term view describes operations 1 to 2 years
into the future and includes only existing operations systems. It is useful
for guiding the development of detailed procedures for personnel in
operations centers and for providing a transition strategy toward operations in the long-term view. The long-term view, representing a "target"
plan, usually describes operations 4 to 7 years into the future. It includes
the directions in which operations centers are planned to evolve to meet
future needs, the changes planned for established operations systems
through new software program modules, and the proposed functions of
new operations systems currently being defined.
The approach used to formulate Bell System operations plans, which is
shown in Figure 15-2, consists of the following five major steps:
1)

Identify all essential processes within the scope of the plan and the
result of each process. For TNOP, this includes all processes neCessary to provision, administer, and maintain the telecommunications
network. This first step identifies what processes are necessary; the
remaining steps in the planning approach are concerned with
determining how processes achieve desired results.

STEP 1
OPERATIONS PROCESSES

...... .....
,,-<'" ....

., ..

....

I

.. ...

... ...

--- - ........ _---,.,......."

I
I

J.. ....

........

..

-

-

.... ........

,~,

<....., PROCESS B"., > <.... PROCESS C.,.. ')
... ,. ..
"
,,'

..

,.....

".....
.....
<,PROCESS 0">"<.,PROCESS E"'" ')
"
,,"
"
,
",

",,'

'......,"

STEP 2

I
I

I

r---L.---, ;:::::_---,
:

FUNCTION

:

I _______
B
L
JI

l

FUNCTION

:

I _______
C
L.
.JI

STEP 3

STEP 4

STEPS

PLAN B

PLAN A

PLAN B

Figure 15-2. Formulation of Bell System operations plans.

Operations

650

Part 4

2)

Break each process down into the major functions that must be performed in order to achieve the desired result. For the Trunk Addition Process mentioned above, major functions include: generating
the order for a trunk to be added, designing the trunk, installing
the trunk, testing the trunk and putting it into service, and updating all relevant network data bases.

3)

For each process, identify operations centers that are candidates to
perform the activities involved in each major fu.nction and identify
operations systems that will provide the mechanization needed to
support the centers. Appropriate systems may already exist, or it
may be possible to design software enhancements to increase the
capabilities of existing systems. Alternatively, the opportunity for
mechanization might lead to the development of entirely new
operations systems.

4)

Generate alternative plans for systems and centers to perform the
activities involved in each major function for each process.

5)

Evaluate alternatives based on such factors as operational viability,
commonality between processes, and economics and select a plan.

Bell System operations plans are usually reviewed annually to take
into account new technical concepts and current corporate policies.
15.2.1 OPERATIONS CENTERS
As stated earlier, an operations center is a group of people, reporting to a
common manager, who perform a set of related functions for a specific
geographic area, group of customers, or service. Centralizing related
functions in operations centers takes full advantage of the power of
operations systems and has the added benefit of presenting opportunities
to improve the quality of the work people do. An example is the centralization of trunk testing in the Switching Control Center, which is made
possible by the deployment of the Centralized Automatic Reporting on
Trunks (CAROT) system. CAROT is an operations system that tests
trunks at electromechanical and electronic switching systems and sends
its findings to a remote computer terminal. A technician at that terminal
is relieved of the tedious, time-consuming· task of manually testing trunks
and can concentrate on the challenge of problem analysis and solution.
Centralizing these terminals at a Switching Control Center (see Figure
15-3), to which data from several switching offices are routed, gives a
technician frequent opportunities to handle varied and interesting problems. The technician also has access to several computerized tools to help
in problem analysis and correction.
Bell System operations plans describe more than eighty types of
operations centers. The description of each type includes the functions

Figure 15-3. A typical Switching Control Center. One of New York
Telephone's Switching Control Centers is the hub of local switch ing office
surveillance, testing , and maintenance for over a dozen central offices.
Switching equipment technicians are assisted in their work by computer-based
operations systems.

performed in the center, the operations systems used, the data bases
needed (such as trunk circuit layout records in the Circuit Provision
Center), and the other operations centers with which the center
exchanges information.
15.2.2 OPERATIONS SYSTEMS
As defined in Chapter 14, operations systems are computer-based tools that
Bell System employees use in performing many operations activities. The
support functions provided by well over 100 operations systems may be
reflected in one Bell System operations plan such as TNOP. Some of the
systems are fully developed and available to the BOCs within the shortterm view presented in the plans. Others may be in initial development
but are expected to be available during the long-term view. The longterm view also describes enhancements to currently available operations
systems that enable them to support future network services, telecommunications technology, or operating methods.
The long-term view may include conceptual operations systems. A
conceptual operations system is a set of closely related functions that are
candidates for mechanization. The capabilities assigned to a conceptual
system may be implemented in a variety of ways: A new stand-alone
operations system may be developed; the functions may be incorporated
as enhancements to existing systems, or some of the functions may
remain manual if detailed economic analyses in the future do not justify
651

652

Operations

Part 4

the development of mechanization. The conceptual systems provide the
basis for further study by suggesting how functions might be mechanized
and integrated into overall operations processes.

15.2.3 OPERATIONS PROCESSES
Telephone company operations are described in the form of operations
processes. As mentioned earlier, identifying operations processes is the
first step in formulating a Bell System operations plan. As plan formulation continues, each process is further described until it is in the form of
a detailed account of the sequential roles of centers and systems in producing a specific result.
Bell System operations plans describe about eighty operations
processes. Each process is depicted in a diagram accompanied by a narrative explanation of the work flow, time sequence, and logical structure of
the process. In the diagram, the process is typically broken down into
twenty to forty activities assigned to systems and centers and the interfaces between the systems and centers.
Figure 15-4 illustrates a relatively simple process, the Residence Customer Billing Inquiry Process for 1985. 1 The figure shows the sequence of
centers and systems involved in responding to questions from residential
customers about their telephone bills. Other operations processes
describe the maintenance of an electronic switching office, the installation of complex equipment for a business customer, and the maintenance
of a specific network service. By associating required functions with
operations systems and centers, processes such as these establish functional requirements for operations systems (see Section 15.4) and provide
the BOCs with guidelines for their own operations (see Section 15.3).

15.2.4 MODEL AREAS AND MODEL COMPANIES
Because of their scope, Bell System operations plans cannot address the
special circumstances of each BOC. Fortunately, however, many telephone company operating areas throughout the Bell System have similar
characteristics (customer population density and number of telephone
lines, for example) and can be divided into groups on that basis. By
using a few model areas and companies to represent these groups, the
Bell System operations plans provide some practical examples for the
BOCs to use in formulating their own operations plans.
For example, one model area is a large, densely populated metropolitan region with about 2 million telephone lines (for example, Chicago,
Cleveland, Detroit). Another is a large, primarily rural state having no
metropolitan area with more than 200,000 telephone lines and including
large sections served by independent telephone companies (for example,
1 This process and others will change as the result of judicial and legislative actions.

o

OPERATIONS CENTER

o

OPERATIONS SYSTEM

REQUESTS FOR CUSTOMER
NAMES AND ADDRESSES

REQUESTS FOR
CUSTOMER NAMES
AND ADDRESSES

REQUESTS FOR CUSTOMER
NAMES AND ADDRESSES

BILL ADJUSTMENT
VOUCHERS

BILLING
ADJUSTMENTS

BILLING
ADJUSTMENTS
TO BE
INVESTIGATED

CRC
(10)

REQUESTS FOR
CUSTOMER RECORDS

REQUESTS FOR
MAINTENANCE INFORMATION

Figure 15-4. Residence Customer Billing Inquiry Process-1985. The billing
inquiry process begins when a customer (1) calls the Residence Account Service
Center (RASC) (2) with a question about a bill. The customer remains on the line
while the telephone company employee in the RASC accesses customer records
from the Billing and Order Support System (BOSS) (3), maintenance outages
from the Loop Maintenance Operations System (LMOS) (4), and customer name
and address from BOSS (5) or from the Customer Name and Address Bureau
(CNAB) (6). CNAB is contacted for name and address information for outlying
geographic areas not covered by the BOSS data base and accesses the
Premises Information System (PREMIS) (7) for this information.
If bill adjustment is required, the employee in the RASC accesses BOSS (8)
to generate an adjustment voucher. BOSS sends the voucher to the Customer
Records Information System (CRIS) (9), which updates the customer·billing data
base. If bill adjustments must be investigated, BOSS routes them to the
Customer Record Center (CRC) (10).
When the inquiry process is complete, the telephone company employee
answers the customer's question or indicates the need for further investigation
based on process results.

Wyoming, Montana, South Dakota). A third model area describes states
or portions of large states with moderately sized urban centers (Georgia
or Illinois excluding Chicago are examples). New York City, with over
4 million telephone lines, is treated separately. In addition, the basic
model areas can be grouped to form model companies.
653

654

Operations

Part 4

The model areas and companies serve as prototypes for ongoing
operations planning in each BOC by suggesting how operations centers
and systems can be deployed in different environments. Bell Laboratories
operations planners use the model areas and companies as the basis for
economic and operational studies analyzing alternative configurations of
operations systems. The models are also the basis for estimating data
traffic for the planning of the future operations systems network (see Section 15.5) that will enhance communication capabilities among the
centers and systems.

15.3 IMPACT OF OPERATIONS PLANS ON
BELL OPERATING COMPANIES
15.3.1 IMPROVING OPERATIONS
The capabilities and features of switching and transmission systems are
continuously evolving to provide new customer services and increased
operational efficiency. Operations plans guide Bell Laboratories in coordinating the operations system enhancements necessary to support this
evolution.
With the centralization of operations system enhancement at Bell
Laboratories, operations systems are kept compatible with rapidly changing telecommunications equipment technologies on a Bell System-wide
basis. In turn, the BOCs can keep up with technological change in a
timely and efficient manner because they do not have to spend time and
money individually arranging for enhancements. Most enhancements are
software program changes provided by Bell Laboratories. For instance,
when a switching system whose trunks are tested by the Centralized
Automatic Reporting on Trunks (CARaT) system is modified, Bell Laboratories only needs to develop a new software program for CARaT. The
software is thoroughly tested by Bell Laboratories and then made available to the BOCs. The program is then installed in existing trunk-testing
computers across the country, providing quick and economic deployment
of new capabilities.
By keeping the family of operations systems in step with network
innovations, Bell Laboratories also enables the BOCs to maintain stability
and consistency in their operations-a technician uses CARaT to conduct
a trunk test in the same way regardless of the vintage of equipment being
tested. Thus, it is possible to provide a continuous, consistent usercomputer interface in a changing environment.

15.3.2 OPERATING NEW NETWORK SERVICES
A major application of Bell System operations plans is in planning for the
implementation and operation of new network services. The set of
operations processes in the plans is the standard with which operations

Chap. 15

Operations Planning

655

for new services must be compatible, so that operation of a new service
will be consistent throughout the country.
Automated Calling Card Service (see Section 2.5.1), a nationwide service introduced in the early 1980s, is an example. Customers who subscribe to the service can dial calling card calls, collect calls, and calls
billed to a third number (that is, calls billed to a number other than the
calling or the called number) without operator assistance. The service is
offered by each BOC, yet a customer need not be within the geographic
area of the customer's "home" BOC to use the service. Each customer is
assigned a calling card number that is stored in one of several data bases
across the country. The data bases are maintained by AT&T Long Lines
as part of the stored-program control/common-channel interoffice signaling2 network. Regardless of the area from which a customer places a call,
there is an accessible data base that can be checked to validate the customer's calling card number.
The nationwide scope of this service poses special maintenance problems. What happens if a salesperson who lives in San Francisco is on a
business trip in Cleveland and has difficulty completing a call using
Automated Calling Card Service? Whom does the person call for help:
the operator, the local BOC business office in Cleveland, the local telephone repair service bureau found in the front of the telephone directory, or the person's "home" telephone company in California? How does
the BOC employee receiving the call respond? Operations planners must
consider these and many other questions in advance. The planners must
provide directions for BOC employees, plans for the operation of supporting systems and centers, and guidelines for customer education.
Before any new service is formally offered, a maintenance plan is
devised and tested. Its objective is to establish a uniform set of maintenance processes nationwide so that, when the service is implemented, it
operates smoothly and consistently throughout the country. For example,
a customer using the Automated Calling Card Service follows the same
simple procedure for reporting trouble in all areas of the country. In
addition, the BOC employee in Cleveland who receives the trouble call
follows the same procedures for resolving that trouble as BOC employees
in San Francisco, even though the source of the problem may be in a data
base owned by the Pacific Telephone and Telegraph Company.
Furthermore, these maintenance functions are incorporated into existing operations centers and systems within the framework set by the Bell
System operations plans. For Automated Calling Card Service, maintenance procedures for the already established Automatic Intercept System
(see Section 10.4.2) were used as the basis for the new service.
Planning before new services are introduced helps ensure that operations for new services will fit into the well-established, stable operations
picture and will be integrated with operations for other services with
2 Sections 8.4 and 8.5 discuss common-channel interoffice signaling.

Operations

656

Part 4

similar maintenance requirements. Thus, the effect of new services on
current operations is reduced, and services can be made available to customers more quickly than if each operating company had to plan individually for the operation of a new service.

15.3.3 OPERATIONS PLANNING IN THE BOCS
In addition to the operations-planning activity at AT&T and Bell Laboratories, each BOC is involved in its own operations planning. In contrast
to the planning at AT&T and Bell Laboratories, which is primarily concerned with the future structure and configuration of operations systems
and centers, planning in the BOCs focuses on adapting the concepts
presented in the Bell System operations plans to their own situations.
The major emphases of BOC operations planning are:
• determining the present status of operations
• evaluating the economics and the operational feasibility of alternative
ways of implementing operations systems and centers
• budgeting capital and expense commitments
• developing transition plans to achieve the objectives of the Bell System operations plans
• overseeing short-term planning and implementation of operations systems, centers, and processes
• providing input to the formulation of the Bell System plans based on
BOC experience.
In the BOCs, the operations-planning process for individual projects,
such as the deployment of a new center or system, generally consists of
four phases (see Figure 15-5):
1)

strategic, or long-term, planning

2)

tactical, or short-term, planning

3)

implementation

4)

ongoing support.

Strategic, or Long-Term, Planning
This phase answers the question: What should be implemented? During
long-term planning, the BOCs adapt corporate goals and objectives, such
as those presented in Bell System operations plans and the Bell System
Corporate Plan} to their local operations environment.
3 The Bell System Corporate Plan sets objectives for the BOCs in the areas of service,
investment, expense, work force, the deployment of operations systems and centers, and
the use of administrative programs.

OPERATIONS PROJECT

~,

ADAPT CORPORATE GOALS AND OBJECTIVES TO THE
LOCAL OPERATIONS ENVIRONMENT

LONG-TERM PLANNING

~,

TRANSLATE MAJOR MILESTONES OF
LONG-TERM PLANNING PHASE
INTO DETAILED IMPLEMENTATION SCHEMES

SHORT-TERM PLANNING

~,

FORM COORDINATION AND CUTOVER COMMITTEES,
PREPARE SITES, WRITE SPECIFIC ESTIMATES, ORDER
EQUIPMENT, RELOCATE PERSONNEL AS NECESSARY

IMPLEMENTATION

~,

MONITOR OPERATION TO ASSIST END-USERS THROUGH
TRANSITION PERIOD

ONGOING SUPPORT

Figure 15-5. Operations planning in the Bell operating companies.

Before conducting long-term planning studies, BOC operations
planners must prepare an inventory of the present status of operations in
their company. This is especially critical because most operations projects
will affect many facets of existing operations. The present status includes
the extent of deployment of operations systems and operations centers,
the company's current organizational structure for operations, the current
distribution of functions among operations centers, and basic data on the
telecommunications network within the company. This present-status
inventory provides information for all future operations planning.
Using the present-status inventory as a basis, various organizations
within a BOC initiate proposals for new projects or modifications to existing projects. These proposals may involve, for example, implementing a
new or enhanced operations center or system recommended in a Bell System operations plan; instituting a new way of maintaining new network
657

658

Operations

Part 4

equipment or service offerings; or adding data-switching capability to the
operations systems network (see Section 15.5).
Operations planners screen proposals and conduct evaluation studies 2
to 6 years before proposed deployment of a project. The studies generate
alternatives, and operations planners select a course of action based on
economic, operational, and technical considerations. Often, the planners
must coordinate their activities with corporate plans for the telecommunications network to ensure that proper operations systems and centers are
deployed to support growth or replacement of telecommunications equipment. In fact, operations planners may initiate plans for the deployment
of some telecommunications equipment that may have significant operational benefits, such as the Digital Access and Cross-Connect System (see
Section 9.4.3).
The course of action selected includes a transition strategy with major
milestones specified for the tactical-planning and implementation phases
of the project. Once this course of action is approved by BOC management and the financial requirements are met by BOC expense and construction budgets, the project proceeds to the tactical-planning phase.
Tactical, or Short-Term, Planning
This phase addresses the question: How should the project be implemented? It typically occurs up to 2 years before project deployment.
During tactical planning, the objectives and major milestones of the
long-term planning phase are translated into detailed implementation
schemes. These schemes may include number and location of proposed
operations systems and centers, personnel and operational transition
plans, and year-by-year deployment schedules.
A detailed economic study may be conducted to supplement the
evaluation from the long-term planning phase. This study includes final
size and cost relative to physical layout, hardware, software, and telecommunications equipment required. Necessary methods of operation, training, and performance measurement plans are also provided. During this
phase, BOC planners may actively consult with project managers from
AT&T and Bell Laboratories on technical matters.
Implementation
This phase proceeds according to the specifications resulting from tactical
planning. Typically, it includes forming coordination and cutover committees/ preparing sites, writing specific estimates, ordering equipment,

4 Cutover committees are responsible for planning and executing actions required to effect
the actual change (cutover) to use of a new system or system configuration.

Chap. 15

Operations Planning

659

and relocating personnel as necessary. The implementation team may
include representatives from engineering, information systems, and endusers organizations. The equipment supplier or purchasing agent may
become actively involved with the details of equipment installation.

Ongoing Support
Once the project is implemented and acceptance testing is complete, the
end-users assume responsibility for the ongoing operation. For a new
operations system or center, BOC operations planners-sometimes with
the assistance of Bell Laboratories and AT&T planners-may monitor the
operation to assist the end-users through the transition period.

15.4 IMPACT OF OPERATIONS PLANS ON THE
DEVELOPMENT OF OPERATIONS SYSTEMS
Bell System operations plans guide Bell Laboratories in evolving the
whole family of operations systems in concert. Proposals for new operations systems or enhancements to existing systems are viewed in terms of
their contribution to the overall plan. The objective is to maximize the
contribution of the whole set of standard operations systems by ordering
the assignment of functions among related systems.
The Engineering and Administrative Data Acquisition System
(EADAS)6 is an example of how this philosophy is applied. EADAS was
originally designed to supply traffic data on an hourly, daily, and
monthly basis for long-term traffic engineering for the network. Bell
Laboratories modified the system to supply traffic data at 30-second and
5-minute intervals for short-term network management purposes, and an
interface was built to the Network Management operations system (called
EADAS/NM). If EADAS had been viewed narrowly from the perspective
of its original purpose, its potential adaptability for short-term use might
not have been recognized. But, from the perspective of overall network
operations planning, modification of this traffic data collection system
was the most effective way to provide certain network management
capabilities.
Bell Laboratories has adopted a structured approach to planning the
integration of operations systems. To ensure that each operations system

5 The end-users are the personnel constituting the operations centers supported by the
operations system being implemented.
6 EADAS is part of the Total Network Data System (TNDS) described in Section 14.3.1.

660

Operations

Part 4

is developed and modified to fit the overall Bell System plan, planning is
required at several levels. These levels include:
• defining design requirements for individual operations systems and
allocating functions between operations systems and telecommunications systems
• grouping operations systems into specific operations centers, including
defining the roles of proposed or existing systems in order to provide
efficient support for the operation of a center
• specifying the interactions between groupings of related centers and
the operations systems with which they interact
• integrating the entire family of operations centers and systems into
the Bell System operations plans.

15.5 OPERATIONS SYSTEMS NETWORK PLANNING
The number of operations system computers (including minicomputers
and large mainframe computers) in use throughout the Bell System in
1981 was estimated at about 4500. These computers are accessed by about
114,000 computer terminals dispersed among the operations centers.
These numbers are expected to increase to about 6300 computers and
180,000 terminals by 1985 as implementation of operations plans matures.
To provide more efficient transfer of information among operations
systems and between operations systems and terminals in operations
centers, Bell Laboratories is planning the operations systems network
illustrated in Figure 15-6. When fully deployed in the mid-1980s, this
network will provide a flexible, cost-effective national network of operations systems, computer terminals, and data communications capabilities.
It will carry the over 2.5 trillion characters of information required each
month to operate the nationwide telecommunications network. In fact, it
can be viewed as a "second network," parallel to the telecommunications
network whose operation it supports.
The operations systems network will provide many new benefits to
nationwide network operations. For example, network operations tasks
often require Bell System technicians to access several computer systems.
To repair a circuit, a technician may first need to access a record of the
circuit from TIRKS, the Trunks Integrated Records Keeping System (see
Section 14.2.1), and then perform a test using the CAROT system mentioned earlier in this chapter. With the operations systems network, the
technician will be able to access these systems and others, as needed,
from a single computer terminal. Bell Laboratories is in the process of
deploying this capability using a data communications system called the
Bell Administrative Network Communications System, designed as part of the
operations systems network plan.

o
D

OPERATIONS CENTER
OPERATIONS SYSTEM

Figure 15-6. Operations systems network. The operations systems network
includes the collection of operations systems, communications terminals at
operations centers, and the switched and direct communications links
interconnecting them and connecting them to a variety of telecommunications
systems, such as an electronic switching system.

In the future, the flexibility afforded by the operations systems network in rapidly switching data between operations systems computers
and terminals will permit the Bell System to take advantage of advances
in distributed-processing technology. It will be possible for many
different operations systems to share data bases to a much greater extent
than is possible today. The result will be even more economical computer support, using more timely and accurate data.
Planning is also underway to incorporate "intelligent terminals" (computer terminals with self-contained processing capabilities) into the
operations systems network. With the integration of this technology, it
will be possible to shift much of the processing done today in centralized
com puter systems to a technician's terminal. This will make more
efficient use of computer resources and the computer communications
network and will provide even more rapid response to a technician's
inquiries. This new technology will also allow individual technicians to
customize computer displays into a form that they find personally convenient, both increasing the technicians' effectiveness and making the
human-machine interface more personal.
661

662

Operations

Part 4

15.6 SUMMARY
Operations in the Bell System are becoming more sophisticated through
the increasing use of advanced computer technology. Operations must
respond to a rapidly changing environment as the result of the evolution
of the telecommunications network as it embraces new technologies, the
development and marketing of new network services and customer products, and competitive and regulatory forces. All these factors change the
way operations are performed.
Operations planning is a relatively new and expanding cooperative
effort among Bell Laboratories, AT&T, and the Bell operating companies
to guide the evolution of operations in the Bell System. Centralized
planning for the entire Bell System ensures a consistent, high standard of
telecommunications services throughout the nation and the rapid introduction of new services and technologies into the network. The operations systems network will allow an even more effective use of operations
systems in the mid-1980s and beyond.

AUTHOR
B. A. Newman

16
Evaluation of Service
and Performance

16.1 THE SERVICE EVALUATION CONCEPT
When purchasing a new car, a customer considers not only price but the
quality of the car as a whole as indicated by mileage, warranties, safety
reports, and repair costs. Similarly, communications customers are concerned not only with their monthly bills, but increasingly, with the availability and suitability of their telephone service. To basic voice residence
customers, quality may simply mean that they are nearly always able to
place calls as desired, that they are able to hear and be heard well on a
given connection, and that they receive appropriate aid from the telephone company when they have a problem. To the sophisticated business customer, quality may mean the ability to transmit error-free data or
to transmit clear images over an electronic blackboard or other visual
telephone service. Acceptable quality, then, depends on the degree to
which the delivered service satisfies customer expectations. Service
evaluation, thus, becomes a matter of
1)

characterizing what customers expect of a service and what can be
provided, based on current network performance

2)

setting service and performance objectives based on these expectations

3)

assessing conformance to these objectives

4)

taking action to modify (usually improve) service or to re-examine
service and performance characteristics and objectives, if objectives
are not being met.

663

664

Operations

Part 4

The service evaluation process is shown in Figure 16-1 and described
briefly below.

Figure 16-1. The service evaluation process.

In characterizing what is expected of a service, the first step is to identify which service characteristics most affect the opinions and actions of
the people using the service and the sensitivity of customers to different
values of each characteristic. For example, the customer's satisfaction
with a given connection for basic voice service is largely determined by
the amounts of loss, noise, and echo! present on the connection. The
next step is to relate the relevant service characteristics to physical characteristics of the network, that is, the network performance parameters that
affect the service characteristics. Some relationships between service
characteristics and network performance parameters are direct, but others
are not. After the appropriate performance parameters have been determined, the levels of these parameters in the network are measured.
1 Section 6.6.1 discusses these transmission impairments.

Chap. 16

Evaluation of Service
and Performance

665

Network performance characterization studies have typically been
conducted jointly by Bell Laboratories with AT&T and the operating
companies. Section 16.2 discusses part of the characterization processthe measurement of performance parameters-in more detail.
The network characterization process feeds the process of setting
objectives by providing information on service and performance characteristics that are required to meet customer expectations. Marketing studies may also provide information on customer expectations and needs,
particularly in the case of a new service. It is not difficult to see, however, that as quality increases, costs also usually increase. This is acceptable only as long as the increased cost is offset by an increased net value
to the customer of the product for which the customer is willing to pay
an increased price. An important element in setting service objectives
and performance objectives, then, is striking a balance between the best
possible service and the minimum possible cost. Section 16.3 describes in
detail the process of setting performance objectives.
Section 16.4 discusses various means by which the Bell System has
routinely assessed conformance to its service and performance objectives.
Since people and equipment interact to deliver service to the customer,
service and performance measurements have ranged from customer opinion surveys to employee evaluations to direct measurement of network
and component operation. In addition to helping assess service and performance, the measurements themselves indicate strategies that might be
used to correct certain problems, thus providing for feedback in the service evaluation process.
Service evaluation is, thus, a dynamic, iterative process and must be
flexible enough to respond to new service definitions, the introduction of
new technology, and changing customer expectations.

16.2 NETWORK CHARACTERIZATION
As indicated in Section 16.1, network characterization involves several
steps that describe, analyze, and measure relevant customer-perceivable
service characteristics and related network performance parameters. This
section discusses the rationale for network characterization and focuses
on the process of measuring network performance parameters.

16.2.1 THE NEED FOR NETWORK CHARACTERIZATION STUDIES
Traditionally, Bell Laboratories has conducted field measurement studies
of performance on existing network services. The results have been used
to set performance objectives, to determine whether performance is adequate for existing and proposed new services, and to indicate where

666

Operations

Part 4

changes in network design or objectives could lead to improved performance or lower costs. While these traditional reasons remain, two new
forces make the need for characterization more critical:

• Rapid growth of new services and technologies - Rapid advances in
switching, transmission, and computing technology have spurred
development of a host of new voice, data, and video communications
services. During field trials and early introductions of new network
services and technologies, field characterization studies provide the
data needed to set realistic performance objectives and supply
developers and systems engineers with essential feedback. Examples
of new services are circuit-switched digital capability (see Section
2.5.1) and teleconferencing (see Sections 2.5.6 and 11.5.3). New technologies include low bit-rate voice and packet-switching networks (see
Section 5.8).
• New network structure - The 1982 Modification of Final JUdgment2
effectively mandates the restructuring of the Bell System network into
a set of separate components, or "piece-parts," connected together at
well-defined interfaces to provide end-to-end service. Examples of
components are the networks provided by interexchange carriers and
the access and intraexchange networks provided by Bell operating
companies. The components will be individually specified, administered, and sold. They will each require separate performance characterization so that those who file tariffs and maintain the components
and those who assemble services from the components can enter into
meaningful, performance-based contract negotiations.
For example, an interexchange carrier wishing to provide telephone service to end-users may pay the local operating company for
access to the local network in order to provide a path from the customers' premises to the long-distance access point. In return for access
charges, the interexchange carrier would require appropriate access
performance levels from the local telephone companies to ensure
high-quality end-to-end service. The telephone company would have
responsibility for maintaining the agreed-upon performance levels.
16.2.2 THE RESULTS OF MODERN CHARACTERIZATION STUDIES
As shown in Figure 16-2, network performance planners study performance
issues related to new services and technologies and network
restructuring, initiate characterization studies as appropriate, and identify
key parameters (for example, call-setup delay and transmission-related

2 See Sections 1.3.1 and 17.4.4.

Evaluation of Service
and Performance

Chap. 16

667

parameters) to be characterized. Characterization planners apply advanced
measurement technology, statistical sampling methods, and data analysis
techniques to the specified parameters in field characterization studies.
As indicated in Figure 16-2, the primary outputs from the studies are

• Performance-related documents - Statistical performance descriptions
obtained in traditional characterization studies have appeared in articles in technical journals (such as The Bell System Technical Journal),
public reference documents, and internal Bell System Practices (BSPs).
To meet the needs described in Section 16.2.1, current characterization

PERFORMANCE PLANNING
• NEW SERVICES
• NEW TECHNOLOGY
• NEW NETWORK STRUCTURE

~ KEY PARAMETERS
CHARACTERIZATION PLANNING
• ADVANCED MEASUREMENT TECHNOLOGY
• ANALYSIS/SAMPLING METHODS

~
CHARACTERIZATION STUDIES

I

I

I

G -MARKEn~G

PERFORMANCE-RELATED DOCUMENTS

I
JOURNALS
REFERENCE
DOCUMENTS
BSPs

I
NETWORK
SERVICE
PERFORMANCE
SPECS

I

I

INTERIM PERFORMANCE
SURVEILLANCE SYSTEMS

I
SURVEILLANCE SYSTEMS
PLANNING

I
I

SURVEILLANCE SYSTEMS

USER
INTERFACE

MENT
DEVICE

Figure 16-2. Overview of network characterization process.

DATA
BASE

I

668

Operations

Part 4

studies also supply information for internal network service performance specifications, tariff preparation, and marketing campaigns and
literature. These new applications for performance information are
especially important in the newly competitive arena for telecommunications services.

• Interim performance surveillance systems - After its design and initial
use in characterization studies, a new performance measurement system may serve as an interim performance surveillance system for a
service. In the first year or two of the new service offering, the
interim system provides the performance information needed to
ensure that published performance specifications are being met.
Meanwhile, experience with the interim system is used in the
development of the surveillance system that will ultimately replace
the interim system.

16.2.3 MECHANIZING PERFORMANCE CHARACTERIZATION
STUDIES
Traditionally, a separate measurement system has been designed for each
characterization study, used once, and then dismantled. Technological
advances and the rapid growth in the number of new services has made
it both possible and necessary to develop generic, reusable measurement
capabilities. Figure 16-3 shows one such system, the Automatic System
for Performance Evaluation of the Network (ASPEN), in its first
application-a comprehensive characterization of end office-to-end
office transmission performance in the Bell System network.
The key elements of the ASPEN system are remote test modules
(RTMs) and a central computer. The RTMs can be moved to different
locations from which they act as "robot customers." In the End-Office
Connection Study illustrated in the figure, RTMs were placed at selected
end offices throughout the country. Under the control of the central
computer, the test nodes placed calls to one another at all hours and
made over twenty-five measurements, mostly transmission-related, during
each connection. (Amplitude and phase response versus frequency,
noise, hits, and call-cutoff rate 3 are examples of the parameters measured.)
After each call, the test nodes sent the measurement data to the central computer to be screened, stored, and analyzed. For this study,
central-computer control programs were run using the UNIX4 operating
3 The probability that a call in progress will be interrupted or terminated other than by
actions of the calling or called party. Section 6.6 discusses amplitude and phase response
versus frequency, noise, and hits.
4 Trademark of Bell Laboratories.

TRANSMISSION
IMPAIRMENT
MEASUREMENT
SET
MICROCOMPUTER
DATA SETS

o

REMOTE TEST MODULE

TRANSMISSION PERFORMANCE
DATA GATHERING
CONNECTION AVAILABILITY
DATA GATHERING
TRANSMISSION OF DATA TO
CENTRAL COMPUTER

•

CENTRAL COMPUTER

MEASUREMENT SCHEDULING
NODE CONTROL
DATA COLLECTION/MANAGEMENT
DATA ANALYSIS

Figure 16-3. ASPEN End-Office Connection Study.

system, and measurements were stored using POLARIS, a data-base
management system originated at New York Telephone Company. The
principal data analysis tool used was the "5" Data Analysis System
developed at Bell Laboratories.
Variations of the original ASPEN design are being applied in newservice performance characterization studies for which the robot customer
structure is appropriate.

16.3 SETTING PERFORMANCE OBJECTIVES
16.3.1 OVERVIEW

Performance objectives must ensure satisfactory levels of service to customers with disparate needs, while being cost-effective for the service
provider. Moreover, customers' needs and perceptions with respect to
service change with time. These factors dictate that setting objectives for

669

670

Operations

Part 4

the performance of elements of the telephone network be an iterative
process. The major steps in this process are:
1)

dividing the network into components of manageable size and
describing and quantifying the operation or performance of each
component using mathematical models based on measurements
made during the characterization process

2)

determining user opinion and acceptance levels by subjective testing (or other laboratory testing if the "users" are equipment such as
data sets)

3)

determining grade-of-service distributions

4)

formulating performance objectives.

Iterations in this objective-setting process are generally stimulated by
the availability of new technology, which leads to proposals for new services or cost-saving changes to the network. A proposal is first evaluated
by sensitivity studies of models of the existing network. These studies
develop estimates of operation or performance effects of the proposal, the
impact on service, and the cost of achieving those effects. Given acceptable results, a trial consisting of a small-scale application of the proposal
may be conducted. Here, grade-of-service ratings are determined and
performance objectives are formulated. If the results of the trial application are satisfactory, the proposal may be implemented throughout the
system. Once the changes have been incorporated in the telephone network, performance or operational measures are again characterized,
mathematical models are refined, new grade-of-service ratings are determined, and objectives are reformulated.
16.3.2 CREATING PERFORMANCE MODELS

Performance models facilitate analysis of proposals for new services or
new network technology because they are more tractable and less expensive than field trials for evaluating alternatives. Generally, mathematical
models allow better understanding and more precise definitions of
processes and the characteristics of facilities than do word descriptions.
The models must be sufficiently detailed to account for all factors that
significantly affect or contribute to the performance of the telephone network. When general models do not exist (for example, for new services
under consideration as public offerings), they may be created as described
in the following paragraphs.
The first step is to specify the nature and features of the proposed service or technology, the facilities that will be involved, the performance
parameters that must be controlled in order to render satisfactory service,

Chap. 16

Evaluation of Service
and Performance

671

and the characteristics of the geographical area where it would be implemented. The next step is to collect all information pertinent to the
cause-and-effect relationships among these factors. This step may include
searching sources (such as Bell Laboratories new-system design information), referring to operating telephone company measurements of facilities, and if necessary, planning and conducting a new characterization
study. Mathematical expressions are then generated for each of the
parameters under consideration. These expressions are defined in statistical terms based on standard techniques of data analysis, for instance,
correlation analyses relating performance parameters and facility descriptors. The set of mathematical expressions derived constitutes the
mathematical model.
Preparing a performance model for transmission parameters affecting
voice telephone service (loss, noise, and echo) in the public switched telephone network (PSTN) provides an example of this process. A call using
the PSTN involves the telephone sets on the customers' premises, the
communication paths (the loops) to local class 5 offices, and a number of
interconnecting links (interoffice trunks) and switching systems. s The
actual route chosen for a particular call through the network could be one
of several possible combinations of links. Surveys indicate that transmission parameters are determined primarily by trunk length; therefore, a
performance model for transmission parameters requires a way of
expressing the expected route length for a call and combining that with
the relationship between trunk length and each parameter.
Computer simulation is used to generate routing data weighted by the
probability of occurrence of a given route. These routing data provide
information as to the number of trunks and the length in airline miles of
each trunk in a connection between two class 5 offices. Transmission
characteristics are then derived by combining this information with trunk
statistics based on system-wide measurements of transmission parameters.
16.3.3 ASSEMBLING CUSTOMER OPINION MODELS
Customer 0plnIOns and perceptions of the quality of a telecommunications service influence the demand for the service. Given the increasing
role of competition in telecommunications, customer perceptions may also
influence the customer's choice among alternative services. Therefore, it
is important to appraise customer satisfaction with offered services.
The models of customer opinions quantify subjective opinion of performance conditions by discrete ratings (for example, excellent, good, fair,
poor, unsatisfactory). These ratings, which are obtained primarily from
5 The toll network is currently designed in a hierarchical structure as described in Section
4.2.1.

672

Operations

Part 4

subjective tests, are first tabulated and then converted to mathematical
expressions of the form:
P(R IX).

In quantifying customer satisfaction with the transmission quality of a
telephone call, for example, P(R IX) is the conditional probability that a
customer will rate a call in opinion category R, given a value X of the
stimulus. This function can be estimated from measurement of the
responses of groups of subjects to various controlled levels of the given
stimulus (noise, loss, echo, etc.).
As Cavanaugh, Hatch, and Sullivan (1976) found, opinions depend on
various factors such as the subject group, the time, and the test environment (that is, whether the test was conducted during conversations in a
laboratory or during normal telephone calls). Furthermore, the effects of
various stimuli may be interdependent. In testing the effects of noise and
loss, for example, both the noise and loss stimuli must be varied so that
different combinations of noise and loss can be observed. 6 However, by
taking into account the type and size of the sample and controlling for
the influences of other pertinent factors and stimuli, customer opinions
that reflect the distribution of telephone users can be tabulated.
The techniques described above have been used to prepare customer
opinion models for transmission parameters (loss, noise, echo, etc.) and
call-setup parameters (dial-tone delay, postdialing delay, etc.f for voice
communications. Similar techniques are now being applied to data
transmission and graphics services.
The diagrams in Figure 16-4 show how variations in loss, noise, and
echo in transmission paths affect customer opinion about the quality of
transmission. In the upper diagram, the contours represent the percentage of laboratory subjects who rate simulated telephone service as "good
or better" for varying combinations of values for overall loss and noise.
Loss is measured acoustically in decibels (dB) from the input of the
talker's transmitter to the output of the listener's receiver. Noise is measured in dBrnC, that is, dB above a reference noise (rn) of 10-12 watts,
weighted to emphasize noise at frequencies where noise is relatively
more annoying (C-message weighting).8 For example, with a loss of 15 dB
and noise of 30 dBrnC (dashed lines in Figure 16-4), 75 percent of telephone
customers rate the transmission as good or better on a rating scale whose
categories are excellent, good, fair, poor, and unsatisfactory.

6 The text in Section 6.6.1 covering message circuit noise gives more information on the
effects of noise and loss combinations.
7 Section 8.3 discusses dial-tone delay and postdialing delay.
8 Section 6.6.1 discusses C-message weighting.

20 r-------------~~------------~------~~

ii
~

en
en
0

15

...I
(J

j::

en
:::)
0

(J

<

...I
...I

<
a:

10

w
>
0

5

~

________

15

~

________

20

~

________

25

~

___________
35

30

NOISE AT INPUT TO SET (dBrnC)

85
70

a:
w

l-

50

IW

III

a:
0

30
20

0

0
0

CJ
I-

10

ACOUSTIC LOSS - 15 dB
30 dBrnC

I NOISE AT SET -

I

5

Z

I

W

(J

a:
w
a.

0.01
0

10

20

30

40

50

60

ACOUSTIC-TO-ACOUSTIC ECHO PATH LOSS (dB)

Figure 16-4. Effect of transmission parameters on
customer opinion.

If loss and noise for the primary signal path are held constant at the
values shown by the dashed lines, opinion is then a function of loss and
delay in the echo path (shown in the lower diagram in Figure 16-4). For
example, for a 3D-dB echo path loss and a 100-millisecond delay in the
round-trip echo path, only 20 percent of telephone users find the quality
of transmission good or better.
673

674

Operations

Part 4

16.3.4 DETERMINING GRADE-OF-SERVICE RATINGS
Mathematical models of performance and models of customer opinion are
combined to obtain grade-of-service ratings for specific aspects of telephone service.
Grade of service is defined (for a single parameter, X) by the integral
r~:(R IX)f (X )dx,

where P(R IX) is the conditional probability that a customer will rate a
call in category R, given a value X of the stimulus, and f(X) is the probability density function of obtaining that stimulus. (Multiple impairments
are handled similarly with multiple integrals and joint density functions.)
The process used to determine grade of service is illustrated in Figure
16-5. Controlled subjective tests provide the needed opinion curves,
P(R IX), as discussed in Section 16.3.3, and characterization studies,
together with a mathematical model such as the one described in Section
16.3.2, provide the required performance distributions, f(X).

SUBJECTIVE
TESTS

PERFORMANCE
CHARACTERIZATIONS

,
DISTRIBUTION
OF
PERFORMANCE

OPINION
CURVES

GRADE
OF
SERVICE

Figure 16-5. Grade-of-service determination.

Grade-of-service ratings are essential criteria used in formulating
objectives for various performance parameters. Figure 16-6 shows that as
loss in a telephone connection increases, the grade of service for loss and
noise (in terms of the number of people who rate transmission as good or
better) decreases (dotted curve). On the other hand, this same increase in
loss increases the echo path loss, thus improving the grade of service for

ccw

100

W

90

ECH<}/

80

/

lICO
II:

,,"
/

0

Q

0
0

"zw
I-

/

•·• ••••1
70

CJ

I

II:
W

!!:.
w
CJ
:;
II:
w
II)

--

60

I

I •••••••

LOSS-NOISE

....

....... .
*~
••••••
....

I
I
I

50

II.

0

w

40

Q
c(

II:

"

30
5

10

15

20

25

END-TO-END ACOUSTIC LOSS (dB)

Figure 16-6_ Example of loss-noise-echo grade of service
(for connection length of 1270 miles)_

echo (dashed curve). The result is an optimum value of loss for which the
highest values of the combined loss-noise-echo grade of service (solid
curve) are achieved. The value of optimum loss is a function of connection length because the delay is longer on long connections, and thus,
more loss is required to control echo. For the connection length illustrated (1270 airline miles), an end-to-end acoustic loss of approximately
18 dB (the asterisk in the figure) results in the optimum loss-noise-echo
grade of service.

16.3.5 FORMULATING OBJECTIVES
An overall objective of the Bell System has been to provide high-quality
telephone service at low cost. As technology provides new methods and
facilities, the quality and ease of use of telephone services should
improve, and costs should be reduced. However, as services improve,
customers' expectations increase.
An important consideration in formulating objectives, then, has been
balancing desired service levels and incremental costs. This requires
identifying performance regions of diminishing returns (that is, the
points where the increase in customer satisfaction is not commensurate
with the increase in dollar investment needed to improve performance)
or where savings can be realized without significant reductions in customer satisfaction. Thus, formulating objectives takes into account: (1)
the grade of service being provided, including areas where performance
is lower or higher than objectives; (2) the costs of providing services,

675

676

Operations

Part 4

including costs of correcting poor performance or savings possible
without degrading performance; (3) trends in customer opinions; and (4)
economic trends.
In addition, the end-to-end objectives have been divided and allocated
to each component of the physical plant forming the connections (stations sets, loops, switching offices, and trunks) so as to ensure that the
overall objectives are met and to minimize differences between local and
long-distance calls. These objectives have been applied in preparing the
instructions used by operating company personnel in designing new
installations and making additions to existing ones. The performance
limits developed here are also useful for maintaining the systems in the
field.
Examples of parameters for which performance objectives are established include: signaling (dial-tone delay and postdialing delay), availability and reliability (blocking and cutoff), and transmission (loss, noise,
and echo for voice quality; bit error rate for data transmission; and graphics quality).

16.4 MEASUREMENT AND CONTROL OF SERVICE AND
PERFORMANCE
16.4.1 MEASUREMENTS IN THE BELL SYSTEM
The Bell System has traditionally monitored its own effectiveness through
a system of comprehensive measurements. These measurements were
designed to assess compliance with preset objectives and to detect unsatisfactory cost, service, and performance levels. Some of the measurements
have been applied across all Bell Operating Companies (BOCs) in an
attempt to achieve common Bell System goals.
As is the case with most corporations, the Bell System has monitored
its financial effectiveness by means of a number of financial measurements. Some of these measurements apply at the corporate, or macro,
level-profitability and rates of return are examples. Others, particularly
those relating to cost, also apply to the micro level, that is, to organizations in the lower levels of the corporate hierarchy. The Bell System,
however, has also used macro and micro measurements extensively to
reflect service and performance levels (see Militzer 1980), a rare practice
among large corporations.
Service measurements reflect aspects of operations perceivable by the
customer. Performance measurements reflect whether the intended operation of a Bell System operational unit (for example, an operations center
or an operations process) is meeting objectives. Service and performance

Chap. 16

Evaluation of Service
and Performance

677

measurements are combined in the overall process of measuring, assessing, and reporting both the quality of service provided to customers and
the efficiency of the system and its associated operations.
In the Bell System, service and performance micro measurements
specific to an operational unit have often been combined into one or
more measurement plans (sometimes called indices or index plans). Since
1980, approximately forty major service and performance measurement
plans have been in use in the Bell System network segment alone; a similar number serve the other segments. 9 (Section 16.4.5 describes typical
Bell System measurement plans.) In addition, suitable aggregations of the
micro measurements at the operational unit level are summarized into
results applicable to higher levels of Bell System management.
Those planning measurements of service and performance face many
challenges. Different classes of customers have different needs and
expectations. Business customers usually place a high value on fast reaction time and reliable workmanship in both installation and repair.
Residence customers tend to be more tolerant regarding reaction time and
more concerned with courteous, helpful service and neat workmanship.
Customers located in different geographic regions in the United States
can have different levels of expectations. Finally" customer attitudes and
expectations toward telecommunications tend to change with time. A
generation more acclimated to advanced technology is likely to expect
more from the telecommunications network.
In addition, measurements should foster rather than restrict efficiency
and productivity. Performance criteria should not be too rigidly determined and, moreover, should be in step with the changing needs and
objectives of the Bell System. In the late 1970s, for example, as a part of a
greater emphasis on customers' perceptions of service, many measurement plans included customers' subjective reactions. Planners must also
cope with the accelerating pace of new technology, which imposes new
requirements on the performance of equipment and operations systems.
New technology may also permit the measurement of performance
parameters that could not be measured previously or the more accurate or
convenient measurement of existing parameters. In addition, competition
creates the need to re-examine existing performance objectives
continually.
In summary, the changing nature of customer expectations and the
evolving nature of telecommunications dictate that service and performance measurement planning be dynamic. To meet these challenges, service specifications and performance measurements constantly undergo
re-evaluation and modification.
9 Much of the collection and compilation of measurement results is automatically carried
out by operations systems.

678

Operations

Part 4

16.4.2 THE MEASUREMENT OF SERVICE AND PERFORMANCE

Because the Bell System has traditionally stressed the quality of service
provided to customers, it has always carefully monitored reports from
customers concerning unsatisfactory service. These reports reflect customer actions and represent a type of service measurement. When a customer
reports a trouble to a repair center, for example, a trouble report is
created. Repair efforts aimed at correcting the trouble are then initiated
and tracked, and the final disposition recorded. The trouble report and
the ensuing effort to rectify the trouble provide the data for a number of
micro measurements, both on the nature of customer actions and on the
performance of the Bell System's maintenance effort.
Significant attention is also paid to customer attitudes, that is, expectations of the customer (see McDade 1979). An example of a plan that
measures customer attitudes is the Telephone Service Attitude Measurement Program described in Section 16.4.5.
Service can also be evaluated by measuring the Bell System operations
that provide service. From this point of view, service operations can be
divided as follows:

• Installation - providing what the customer wants. Measurements here
include whether the work is done properly and by the date agreed
upon.
• Availability - ensuring that the service or equipment for which the
customer has paid will be ready for use. Dial-tone delay and network
blockage are examples of measurements in this category.
• Suitability - ensuring that the contracted service meets objectives.
Transmission performance measurements are included in this
category.
• Billing Integrity - ensuring that the customer is not charged an erroneous amount. Measurements in this area include automatic message
accounting (AMA)10 errors.

16.4.3 THE CONTROL OF SERVICE AND PERFORMANCE

Measurements provide visibility into service and performance, but visibility is not enough. Follow-up activities designed to correct substandard
service and performance are necessary. Both the set of directed activities

10 Section 10.5 discusses AMA.

Chap. 16

Evaluation of Service
and Performance

679

(control actions) and the set of methods (control procedures) that direct
these activities are encompassed by the term control.
Ideally, control is based on several considerations. First, most measurements are observed effects. They are a quantification of symptoms
rather than causes. The first stage of control, then, is to identify symptoms and diagnose the cause of a problem. Once the cause is known,
which control parameter must be adjusted and to what extent to overcome the problem in a cost-effective way can be determined. Control,
therefore, requires information beyond that needed for evaluation: It
requires efficient analytical methods for diagnosis, and it further requires
procedures for best rectifying the problem.
To facilitate control, many measurement plans in the Bell System have
had a multitiered structure. The measurement components making up
the official measurement plans are at the top tier. They are used to assess
service and performance improvement or degradation. The next lower
tier consists of measurements supporting the official ones. These are typically more detailed measurements that are not components of the measurement plans but can furnish diagnostic aid. These measurements are
sometimes called performance indicators, and they frequently accompany
the official measurements. In many recent cases, the supporting measurements have been further supported by computer algorithms and documented procedures to analyze additional data.

16.4.4 THE ADMINISTRATION OF MEASUREMENT RESULTS
In the Bell System, official measurements results have been collected
monthly from all the BOCs. As mentioned previously, many of the
results extend down to the level of individual work centers. At the same
time, most results have been aggregated and apply to higher level
management units up to the company level. Measurement results have
facilitated comparison among BOCs, and they have been published and
widely distributed throughout the Bell System. This section examines the
measurement methodology and the mechanized system used by the Bell
System for handling measurement results.
The method typically used to present measurement results divides service and performance levels into four bands. The "H" (high) band is
applied to cases where service and performance levels are so high that
they may not be cost-effective. The "0" (objective) band is applied to levels of service and performance that meet objectives. The "L" (low) band is
applied to levels lower than objectives that may require future action, and
the "U" (unsatisfactory) band is applied to levels that require immediate
corrective action.

680

Operations

Part 4

Measurement Collection, Compilation, and Publication The Centralized Results System
The significant task of managing the large amount of measurement data
in the Bell System was made easier by the introduction of an operations
system known as the Centralized Results System (CRS). CRS is a management information system that automates the collection, analysis, and publication of many measurement results. In many cases, the analysis
includes banding.
CRS supports the BOCs by gathering data at the lowest organizational
level. It can receive the monthly raw measurement results from BOCs
either as terminal input or by a direct computer link. CRS checks the
received data for validity and summarizes and compiles them into a form
suitable for dissemination as official measurement plans. Based on the
organizational structure of each individual BOC, CRS also accumulates
and aggregates official measurement plan results up through the company
level. Each company can retrieve its own compiled results interactively
from CRS.
16.4.5 TYPICAL MEASUREMENT PLANS
This section describes two typical measurement plans. The Telephone
Service Attitude Measurement (TELSAM) Program measures customer
attitudes about service. The 1/1AESS Network Switching Performance
Measurement Plan (NSPMP), measures network component performance.
TELSAM
TELSAM is currently the primary method used to measure customer attitudes about Bell System service. It is based on short (typically, 3
minutes), direct interviews with customers by telephone. Trained personnel under contract to outside research firms carry out the interviews
using specially prepared questionnaires. Areas covered by TELSAM questionnaires include Service Centers, Installation Services, Repair Services,
and Operator Services.
The Residence Installation questionnaire is typical. The customer to
be interviewed is selected from a random sample of those who have had
a telephone or line installed recently. After ensuring that the person
receiving the call was the one who had the service experience, the interviewer asks whether:
• the customer got the desired type of equipment and telephone
• the customer got the desired type of line service (that is, private line,
measured service, or party line)

Chap. 16

Evaluation of Service
and Performance

681

• the work was done neatly and in the way that the customer wanted
• the installer was courteous
• the appointed installation date was set properly by the business office
• the appointment was kept
• it was necessary to have someone come out again for the work
• the installation is working satisfactorily
• overall, the customer is satisfied or dissatisfied.
The questionnaire is designed so that if in answering a specific question, a customer expresses dissatisfaction, the interviewer is guided to ask
a series of further questions to gain a deeper understanding of the cause
of the displeasure.
Because of the cost of interviewing customers and the resultant limit
on the number of samples taken; most TELSAM results are compiled at
the areal l level. A number of operating telephone companies, however,
have elected to sample at lower levels to gain a better picture of districtto-district variability.

l/IAESS NSPMP
This measurement plan is designed to monitor the performance of the
l/IAESS switching equipment. Since a switching system sets up calls, the
plan is also viewed as one that monitors the quality of customer service
provided by the l/IAESS switching equipment.
The plan has a 2-tiered structure. Each tier includes micro measurements judged to be important in evaluating switching service and performance. Ten measured components are at the top tier. These are the key
measurements that collectively reflect the service and performance of
switching equipment. The next tier consists of eighteen performance indicators. These more detailed measurements are designed to support the
,measured components in pinpointing specific potential trouble spots
when corrective action is necessary. (The document describing the
l/IAESS NSPMP instructs craftspersons to monitor a number of even
more detailed items not specifically reported by the plan.) Finally, for
control purposes, detailed diagnostic procedures or computer algorithms
are available for the more complex measured categories such as
Transmitter Timeouts (described below).

11 Typically, a significant part of a state.

Part 4

Operations

682

Table 16-1 shows the items in the measured-component portion of the
standard report for the plan. The components are grouped into four
categories:
measurements designed to reflect difficulties experienced by the customer in obtaining service from the switching equipment. For example, Receiver Overflow measures the number of times
that all of the digit receivers (the equipment in the 111AESS switch
used to collect the digits dialed by the customer) are busy. Restore
Verify Failure indicates the number of times that a customer's off-hook
signal is not properly detected by the switching equipment.

• Machine Access -

measurements of customers' call attempts (or
incoming call attempts from another switch) that failed during call
processing. For example, Transmitter Timeouts registers failures
between the measured switch and a far-end switch when the former
attempts to outpulse digits to the latter.

• Machine Switching -

TABLE 16-1
NSPMP MEASURED COMPONENTS
Category

Weight

MACHINE ACCESS
Dial-Tone Speed
Receiver Overflow
Restore Verify Failure

15
5
5

MACHINE SWITCHING
Transmitter Timeouts
Office Overflow
FCG* & Supervisory Failures
Receiver Timeouts
Equipment Irregularities

10
15
15
10
5

BILLING
Lost Billing

10

CUSTOMER REPORTS
Code 5 and 8 Equipment

10

"" False cross or ground; a test for a false ground on T or R
leads or a cross between the leads.

Evaluation of Service
and Performance

Chap. 16

• Billing -

683

AMA entries that cannot be billed because of AMA tape

problems.

• Customer Reports - counts of those customer-reported troubles that are
attributable to the switching equipment. Those trouble reports that
result in a "trouble found" (code 5) are counted separately from those
that result in "no trouble found" (code 8).
To report NSPMP results, the normalized,12 measured value for each
component is first' banded using the banding scheme described above.
Then, a weighted average of the measured components is obtained by
multiplying the measured value by the weights shown in Table 16-l.
The products are added to arrive at a band for the overall switching office
performance. In the results summarized for higher company levels, the
number and percent of offices in each band are given for each component
as well as for the overall office performance.
16.4.6 FUTURE TRENDS
Effective measurement becomes even more important in a postdivestiture
era of increasing competition and technological change. Each operating
company must individually choose the most cost-effective measurements
to determine its own internal efficiency. Furthermore, since no one company will be responsible for complete end-to-end telephone service, the
focus of the specific items to be measured will be shifted toward quality
assessment and control of the part of the network administered by the
company. The postdivestiture measurement activities will focus on the
following areas:

• Increased measurement mechanization - The measurement process itself
must be cost-effective. New technology needs to be explored and a
highly mechanized approach is increasingly necessary. Automated
systems capable of remote performance measurements, such as the
ASPEN system described in Section 16.2.3, are likely to become
extremely useful. To provide complete service and performance monitoring, this class of system may need to be supported by operations
systems capable of measuring end-to-end performance, such as the
No.2 Service Evaluation System (see Hester 1982),

• Added attention toward company and network interfaces - With each company responsible for a portion of the network, there will be many
complex interfaces occurring at the company boundaries. They will
12 Each component is normalized by dividing the number of occurrences by a factor related
to the appropriate equipment usage or activities.

Operations

684

Part 4

encompass operations such as installation and repair in addition to
interfaces between network components. Tools and methods will
have to be found to delineate the interfaces clearly and to measure the
service and performance related to such interfaces.

• More detailed performance specifications - With communications service
provided by a number of companies in tandem, the performance to be
expected from each company must now be clearly stipulated. This
means a more careful identification of those parameters that are crucial in determining whether service or performance falls within
acceptable bounds. Setting performance objectives for these parameters and determining how best to measure the parameters will take
on added significance.

AUTHORS
H. Aldermeshian
D. P. Duncan

J. c. Hsu

D. G. Leeper
P. Lopiparo
K. I. Park

E. Jeffers

K. C. Szelag

PARTFIVE
ENVIRONMENT
AND EVOLUTION

The first four parts of this book describe the Bell System, its resources,
and the way those resources are managed through operations to provide
service to customers. This part examines the evolution of products and
services in the Bell System and the environment in which it has occurred.
Chapter 17 discusses the external factors that have had a significant
effect on the Bell System operations. It assesses the constraints and challenges of regulation, tariffs, and competition, and examines the nature of
relationships between the Bell System and independent telephone companies and other common carriers. Chapter 18 addresses internal
processes and considerations that have affected the evolution of new and
modified products and services and discusses how the corporate units of
the Bell System have interacted as part of this process.

685

17
The Environment

17.1 INTRODUCTION
On March 7, 1876, Alexander Graham Bell was granted a patent for the
invention of the telephone. Two years later, nearly 11,000 telephones
were in service in the United States, and a 5-telephone central office had
been installed in Washington, D.C. Now, more than 100 years later, over
500 million telephones provide voice and data communications worldwide. The environment in which this progress has occurred is constantly
changing; actions taken by all three branches of government have been
instrumental in defining changes in response to such factors as new technology, marketplace demands, and changing social forces and customer
perceptions.
This chapter focuses on the regulatory and competitive aspects of the
evolving telecommunications environment and the development of the
Bell System 1 as it responded to opportunities and restraints of this
environment. To put the following discussion in perspective, Table 17-1
provides a brief, selective list of significant events related to regulation
and competition.

17.2 REGULATION
In general, regulation has been employed as. a substitute for competition
in markets where competition either does not exist or where its existence
was judged not to be in the best interests of the public. While this
chapter discusses regulation and competition separately, they are closely
linked, and the extent to which competition may exist depends on the
degree of regulation imposed.
1 On January 1, 1984, the Bell System in its present form will no longer exist as a result of
the 1982 Modification of Final Judgment.

687

688

Environment and Evolution

Part 5

Early government efforts to regulate business were not specifically
directed towards the telecommunications industry, but many were precursors of telecommunications regulations that followed. The following
sections discuss some of the pertinent events that have affected the regulation of the telecommunications industry.

17.2.1 THE INTERSTATE COMMERCE ACT
The Interstate Commerce Act of 1887 established the Interstate Commerce
Commission (ICC) to regulate interstate carriers, of which railroads were
the first. The ICC was empowered to inquire into the management of
interstate carriers, to examine records and documents, to summon
witnesses, and to use the federal courts for proceedings resulting from
investigations. Practices such as pooling,2 special rates, rebates, and
discrimination against persons, places, or commodities were outlawed by
the commission. The ICC also made it unlawful for a carrier to charge
more for short hauls than for long hauls if certain conditions were similar and when the longer haul included the shorter haul.
In 1910, the Mann-Elkins Act was added to the Interstate Commerce
Act enlarging ICC responsibilities to include regulation of the telecommunications industry. With this act, Congress recognized the natural
monopoly characteristics of the telephone business, that is, that the public
would not benefit from competition among telephone companies serving
the same area.

17.2.2 THE SHERMAN AND CLAYTON ANTITRUST ACTS
A few years after the legislation that produced the ICC was enacted,
major legislation of a different type was passed-the Sherman and Clayton Antitrust Acts.
In 1879, more than forty individual companies joined together to form
the Standard Oil Trust. 3 About the same time, other large companies combined to form monopolies in other commodities. Under pressure from
several states in which the trusts did business, the courts moved against
them using legislation that existed under common law, and by the 1890s,

2 An informal agreement by which a group of individuals pledged among themselves to
maintain prices or divide markets. Self-interest usually demanded that those pledges be
kept, but since the agreements were voluntary and therefore unenforceable under
common law, they were often broken.
3 A trust was an agreement between individual companies to place their collective stock in
the hands of a group of elected trustees. The trustees were empowered to vote the stocks
of all member companies, to evaluate the properties that made up the combination of
companies, and to issue trust certificates on the basis of which properties were divided.

TABLE 17-1
BELL SYSTEM MILESTONES
Corporate History

Regulation

Competition

1870
Western Electric formed-1872
Bell Patent granted-1876
Bell Telephone Co. formed-1877
New England Telephone Co. formed-1878

Bell Telephone Co. sues American Speaking
Telephone Co. for patent infringement-1878

Bell Telephone and New England Telephone merge
to form National Bell Telephone Co.-1879
American Bell Telephone Co. succeeds National Bell
Telephone Co.-1880
1880
Western Electric becomes sole supplier to American
Bell-1882
First long-distance telephone line, Boston to New
York-1884
AT&T incorporates as subsidiary of American
Bell-1885
Interstate Commerce Act establishes ICC-1887
Sherman Antitrust Act enacted-1890
1890
Bell patent expires-1893
AT&T acquires American Bell assets-1899
1900

AT&T acquires Western Union-1910

Mann-Elkins Act gives ICC jurisdiction over
telecommunications industry-191O

Independent telephone companies proliferate1900-1910

1910
Kingsbury Commitment averts Antitrust Suit-1913
AT&T relinquishes Western Union stock-1914
Nation's telegraph and telephone systems under
government control-1918-1919

DOJ prepares Antitrust Suit against Bell System1913
Clayton Antitrust Act enacted-1914

1920
Operating companies merge into larger regional
units-1920s
Bell Telephone Laboratories formed-I925

Graham-Willis Act affirms Bell System's natural
monopoly status-1921

1930
Federal Communications Act establishes FCC-1934
1940
DOJ considers Antitrust Suit against AT&T, then
postpones action because of WWII-early 1940s
DOJ files Antitrust Suit-1949
1950
Antitrust Suit settled by 1956 Consent Decree1956
Hush-A-Phone Decision-1956
"Above-890" Ruling-1959

Emergence of the manufacture and distribution of
interconnect equipment-1950s

Carterfone Decision -1968
Specialized Common Carrier Decision -1969

Demands for data processing in
telecommunications grow-1960s

DOJ files Antitrust Suit-1974
FCC adopts registration program -1975 (Docket
19528)

Computer Inquiry I Ruling-1971

1960

1970

Computer Inquiry II Ruling-1980
1980
American Bell formed in response to Computer
Inquiry 11-1982
DOJ
FCC
ICC

Department of Justice
Federal Communications Commission
Interstate Commerce Commission

1974 Antitrust Suit settled by Modification of Final
Judgment-1982

Chap. 17

The Environment

691

the trusts began to dissolve. Far from disappearing, however, they gave
way to holding companies4 and consolidation.
The Sherman Antitrust Act, enacted in 1890, was designed to curb the
activities of these trusts and holding companies. It forbade restraint and
monopolization of trade and placed the responsibility of enforcement
with the federal government. The Sherman Act, however, was not
actively enforced until Theodore Roosevelt's administration (1901-1909).
Then, the energies of the Department of Justice were directed at
vigorously enforcing it.
In October 1914, under Woodrow Wilson's administration, the Clayton
Antitrust Act was passed to supplement the Sherman Act. The Clayton
Act banned price discrimination, anticompetitive mergers, interlocking
directorates,S and exclusive-dealing arrangements. Most important, it permitted private individuals to file suits seeking damages as a result of violations of its provisions.

17.2.3 THE KINGSBURY COMMITMENT AND
THE GRAHAM-WILLIS ACT
Thus far, none of the attempts to regulate business had been devised
specifically to control the Bell System. In fact, as mentioned earlier,
Mann-Elkins actually recognized the natural monopolistic character of the
telephone business. But, in 1910, AT&T acquired control of the Western
Union Telegraph Company by a stock purchase. Three years later, when
Wilson assumed the presidency of the United States and started to fulfill
a campaign promise to destroy private monopolies and trusts, the Department of Justice considered an antitrust action against the Bell System.
This action was averted when AT&T Vice-President Nathan C. Kingsbury
committed the Bell System to a course of conduct that would satisfy
government complaints. The Kingsbury Commitment specified that the Bell
System would relinquish its Western Union stock (which it did in 1914),
stop further acquisitions of competing independent telephone companies
(except when in the public interest, and then only with ICC approval),
and interconnect the Bell operating companies and independent telephone companies with its long-distance lines.
Acting under the authority of a joint resolution of Congress during
World War I, President Wilson declared that as of July 31, 1918, the
nation's telephone and telegraph systems would be put under the control

4 A holding company either bought all the properties of its several subsidiaries or acquired
control of each company through purchase of a majority of the stock. In the latter case,
consolidation was effected through the election of common directors who sat on the
boards of all the participating companies.
5 Management of a number of separate corporations by the same or nearly the same group
of directors.

692

Environment and Evolution

Part 5

of the federal government, specifically, the Post Office Department. On
August I, 1918, Postmaster General A. S. Burleson assumed supervision,
control, and operation of those properties and continued in that role until
control was returned to private ownership on August I, 1919.
The Graham-Willis Act of 1921 affirmed that the Bell System had
become a national resource and allowed its existence as a natural monopoly because it provided a unique technology. During the 1920s, the Bell
System emerged in its modern form when the numerous associated
operating telephone companies merged into larger regional groups.

17.2.4 THE FEDERAL COMMUNICATIONS COMMISSION
With the Federal Communications Act of 1934, Congress established
universal telephone service6 as a national goal. This act created the Federal
Communications Commission (FCC) and initiated the modern regulatory
environment for the telecommunications industry.
The FCC was given the primary responsibility for regulating the rates
and conditions of interstate, international, and marine communications.
Immediately after its creation, the FCC ordered an intense investigation
of the Bell System and its operations. As a result of this investigation,
the Department of Justice began to develop a major antitrust suit against
the Bell System. With the imminence of World War II, however, the
Department of Justice postponed any action because it considered the Bell
System vital to national defense.

17.2.5 STATE REGULATORY COMMISSIONS
From 1935 through 1968, the philosophy of regulation matured. State
regulatory bodies generally referred to as public utilities commissions
(PUCs), but with different names in different states, regulated the rates
and conditions of intrastate communications for the common carriers in
their jurisdictions, just as the FCC regulated rates and conditions of interstate service. The basic task assigned to the PUCs was to establish rate
systems to promote the public good and the goal of universal service
while providing sufficient revenues so suppliers could meet the costs of
doing business in the state they served. Methods were established to distribute revenues among participating companies for calls that use the
facilities of more than one carrier. The goal of universal service was a
factor in formulating these distribution methods. (Section 17.3 provides a
more detailed discussion of rates and revenues.)

6 The goal of universal service was to make basic telephone service available at an
affordable price, anywhere in the nation.

Chap. 17

The Environment

693

17.2.6 THE "ABOVE-890" RULING
Technological developments that emerged from World War II-many
based on the work of Bell Laboratories-were applied to areas such as
customer terminal equipment, data services, and microwave communications after the war.
Before 1949, the FCC had assigned the microwave spectrum to the
telecommunications common carriers and had not allowed private systems to connect with the public network. After the war, microwave technology was in demand by businesses and industries located in remote
areas without telecommunications services, so the FCC began leasing
private microwave systems on a case-by-case basis to serve areas where
there were no common carriers.
The major common carriers, including the Bell System, objected that
these leased microwave systems would congest the airwaves and result in
poor service. Many private companies, however, saw the potential savings involved in using microwave systems for telephone service and were
eager to exploit a promising market. As the demand for this market
grew, the FCC relaxed its restrictions on the private companies.
By the end of 1951, 13,000 route miles of private microwave systems
were already in place or being started. In 1959, over common-carrier
objections, the FCC decided that there were ample frequencies higher
than 890 megahertz (MHz) to accommodate those who wished to construct their own microwave systems.
In that ruling, referred to as the "Above-890" Ruling, the FCC decided
in favor of licensing private intercity microwave systems for both voice
and data transmission. Then, in the Specialized Common Carrier Decision of 1969, the FCC decreed that the new microwave companies be
allowed to compete with existing (and regulated) telephone companies in
the sale of private network transmission services.
17.2.7 FCC INTERCONNECTION RULINGS
In 1947, the FCC permitted use of customer-provided recording devices
but ordered that direct electrical connection of such devices to the telephone network must be through protecting arrangements provided and
maintained by the telephone company.
A variety of interconnection devices, many of which were foreign
made, became available during the late 1950s. They were designed to
attach to existing telephone sets or to be used as terminal equipment
themselves. The major common carriers maintained that it would be
impossible to ensure efficient telephone service if devices supplied by
firms with no legal responsibility for the quality of service were attached
to the network by customers; interconnection of such devices could

694

Environment and Evolution

Part 5

increase the network's operating costs and disrupt its efficiency. This
could be particularly damaging in times of emergency.
The FCC supported this position and refused to allow the use of interconnection devices, but it was overruled by the court of appeals in the
case of the Hush-A-Phone device, a small cup-like nonelectrical handset
attachment that enhanced privacy when talking. Then, in 1968, the FCC
ruled in favor of Carter Electronics, a Texas firm that made a mobile radio
device that could be acoustically coupled to the common carrier voice
telephone network. This device, called the Carterfone, was primarily
being sold to oil exploration and drilling companies for use by field
engineers in remote areas.
This ruling by the FCC was a landmark: It set in motion the forces of
deregulation and led to intense competition because, unlike the Hush-APhone Decision, the Carterfone ruling permitted the direct electrical
attachment of devices to the telephone company's equipment provided
that the operation of the network was not adversely affected (see Section 17.4.2). The FCC recognized the concern for potential adverse effects
on the network and on the quality of service as a result of the attachment
of customer-provided equipment and contemplated the continued use of
network control signaling apparatus provided by a common carrier. Consequently, the FCC approved tariffs (see Section 17.3) requiring the use of
a protective coupler between customer-provided equipment and the network. In November 1975, the FCC issued a report and an order instituting a registration program. Under that program, carrier-provided protective coupling devices are no longer necessary if the customer-provided
equipment is registered with the FCC or uses a registered protective coupling device. (Section 8.7 discusses interfaces for interconnection.)

17.2.8 THE 1956 CONSENT DECREE
In 1949, the Department of Justice filed against the Bell System the antitrust suit that it had postponed because of World War II. In particular,
the suit attempted to force AT&T to divest itself of Western Electric. The
suit was settled by the 1956 Consent Decree, which allowed AT&T to retain
ownership of Western Electric as long as Western Electric manufactured
only products used by the Bell operating companies? The decree also
specified that Bell operating companies confine their activities, with some
exceptions, to providing telecommunications services under regulation.

7 Exceptions were made for government-sponsored work or cases where no competitive
products existed. Also, the Bell System was allowed to continue to provide the artificial
larynx (see Section 2.2.5) because it was felt that no competitive commercial supplier
would undertake the task.

Chap. 17

The Environment

695

As a result, the Bell System was effectively barred from providing commercial data-processing services. The Bell System was also required to
decline royalties on its then-existing patents and to license all future
patents to any applicant on a nondiscriminatory basis at reasonable royalty rates. s
The 1956 Consent Decree became a large part of the telecommunications environment in which the Bell System had to operate. Although
the decree had a major impact on the scope of its corporate activities, the
Bell System was able to maintain its corporate identity and vertical integration. 9 The decree remained in force until implementation of the 1982
Modification of Final Judgment (see Section 17.4.4) brought an end to
both the period during which regulation was the primary controller of
the telecommunications business and the Bell System itself.
In the years following the 1956 Consent Decree, advances in signaling
and data-processing capabilities brought telecommunications and data
processing closer together. In 1971, because of the potential impact of the
1956 Consent Decree on the ability of the Bell System to provide useful
new services, the FCC began the first of two computer inquiries to
explore the link between telecommunications services and dataprocessing services. Because of their significant impact on competition in
the modern telecommunications environment, the results of these computer inquiries and subsequent legislation are discussed in Section 17.4.

17.3 TARIFFS AND RATE SETTING
17.3.1 THE ELEMENTS OF A TARIFF
Tariffs are one manifestation of state and federal regulation of the telephone industry. A tariff describes a service, the rate that may be charged
for the service, and the regulations under which that service can be provided. It is a set of terms between the carrier and the customer that must
be submitted to a regulatory body before any new or changed services
can be provided. A tariff, for example, may state the nature of the service, the class of customer to which given rates apply, the availability of
the service in defined areas, the method of measuring the service (for
example, initial and overtime minutes on long-distance calls), and the
method of computing customer charges. Figure 17-1 shows part of a
simple tariff.

8 This had, in fact, been Bell System policy even before the decree.
9 A vertically integrated business is concerned with all the processes involved in a product or
service.

All. n1..' m.. eo. 1..
Ad...... t ••••• TerlU.
"'dIU.. ur. " •• Jer •• ,.

1..

TW", r.e.c:. 110.'"
hble."edtU1.,•••

OIJ'T.

Called. Sdl!leYS.,d ':'HI.,..,
01'1,' ..1 t.rlft tthceh•• Ka,.ll.lt16

01921

tU.ceIO.f

JIll,2,1"2

A"l,lfI, co DAtAlIONE Dl,iul "nteeo fvrA1.b.d b)' Ch. tll.phoN c:..~., for ,rh." 11..
·lflUntet.e_u"lc.elo... ·\UHI.rtll.e_nteecto... "ceof193., . . . . .IId •• ,'booIt ..c1u.taa
."eh.llli' "ntce, •••• t"oll'" • porc1011 of 'Ileb ,,"teo t. huuure ....btlnl.er '1,Ieb ,,"lCI It
.ubjlCtcor""LlaUotlb,'tat.orloe.a1.0•• ~"tol.ut"orlt"

d,•••

Twrr,.C.C:.1fO.267
Il.bl.Y1 ..
10
CaIlC.1I7ehltodlld,.,.10
Iffect!,,1 IkwoItber5. nil

AK.TEL.''nL.CO.L.L. Dut.

.w•• II' ......4T.1'1f,.
;:::!:~t'~e:::, J::'~~11 01911

~.n.:;.~ ~.; T:~t~;

L.

DIn.
"",,,_tor, "_J.r.... 01921
"'~I r.bnta",2',l'"

2.2

o..c,lpUon or

s....tc.

Otlital SentCIl 11 . . .rnc. fo, tb. craU&l.s1_ of dI,lcal .11.. b 0111,
IiIIlaaou,..t.lc.al
to.t.dUtI•••
~'tA'IONE

"r.".tal••

DltAP1l)IIEDi.ltalS.rdcepl'orlde.t... tho 11_11.nlOll.t_7t......t.llbe.bt41lltal

1.

!nll..ct. . .tl,.Mh
lx."" .. UII.fAbbhotul ...
l".nne. ".Ocll.r T.rtfh
",lIc.n' " ...it. ~...q ..... fratl_r
Coneun'illl;e.n'tor.
CO... ct1llcc:..rPl... .
Och.rflrcleb.tt .... COn'.......
"IOUuUollott.rHt

1.1.1

!:r:!:'

t!::h

t~:~~::!:.:t .:~~;.::..!:4,

,

,. l::::"E§:£~::~:~::'::'-:::::.'!.!.:;.:
~'::&l"Ms!

c.~!;to=:IC' c..,-IU
"""1':'l':T'II,.1

~].I~
2,].1

....... nIUt1.f dI.Cul_ ..

~:~:: ~::~!:I!~~.~~:II:"" ......

2.•

Ulltr.ik1LiJ~.!t:.!Il

..II.". ea......

2.5

ft:~::to~:~:~:·=!

c..

g~l §:::~~:1¥;E=~ ;~;';~1:::
~::::

""•• 2".1

_ " .... h.tor.ald_perlodofo. . . . . th.

boIil' pel' . .,., " _ " " , ...

lnel84I.lq1,ll,..lIt proYl'.d.,. cho tel_phil" coepa..,. . . . ..crU•• I. 2.2.4(1)(2) 'ollewt.. .
1 . '.... r.ou.d 1.0 prod . . . . . . . , ... portor-tlC••ac• • • ".51 .. 1'o'-t .... _ ... foropo ... U ••
• t.ll
Vbe • • DAtAlIlOMl: 1)1,ltal ......lc." .,.ractq . t ........... ,.rt......c.
11.," Which 1. _thfacto" co tho C:Un_' or VII' .... Ic 1•••t..N1_d " ebe hie.....
eo.,. ..,.tbatthe.rr.rp.rfo~.1....11."'1_thlt ...c1tl.".,",-.tbeo ..rl"ol
lub-et.nd.r4 p.'fam.IIC.w1U ... collltd.n' •••• 1I1t ....... .,10. . . . . c ....it.U_IIC•
.,111 . . . .do. 111 accord.IIC' .,1tll tbe proo"lo... of 2.6.1 f.11_ ... All _b cndlt
.ll_IIC" oball bed. f .... clll cl_ of ..,UlI lIJ the Oaet._, or 01., to tho t.l.pbOOI
Co.~II" that .lIuIINt1.fec:tor.,.,..rfora&llC.le".lba.occ... rd, ,ro""dM I.batthlc.a.tba "r'de. II ....... t.d bJ tbe t.lIpbo11.
to. porton

'..-.d"

:::fa!':::C!!,!:!:::::.

c-,..,.

'nI• • •1,0 . . .laUIIIIIC. all4 oPltltloa of O4tAl'IOIII Dl.le.l St,'d(O eoal• .,latll e~c.tlou od.l .... l1QtOf coftll.... UQt.t .tatlO11.01ltlHlD4t..,8CltZDi.ltal Senlt1. ilMl.
c.lI'IIIIeUo•• to DATAPIOtfE Di.le.1 $&"lc. of c-.tullo. . .,.u.. prowlilu lI,- ot.......,.
. . . . d. . . . ,.r-t .. t.,. ba.l. II p....1II•• for 11 Seel.io. 2.'. tlMi 'fllaphooa c:o.paay dOli
not"p" •• ntlt.DATAPHONE01.1u1S.rrl.c. . . . . .pcodfo...ucbco_l.loaa • • • •hall
:!.::,"::,::;b!:~:~t~::.:hroqh cr....UIl. . of .11".11. or thl ._111., of such t"..-

Ttl" oU.,i", _, eouht of (1) OM c-potat c _ _ 1 flll'al.had ... t .... M,ttal
C1U•• fo, ••"ie. whI .... 01111 ."aUo. 1. lOCUM til the ..r'fl. . . . . . of .. cb _II
c1t'.D 0).
Regulatory Measures
Cash flow measures are calculated using real cash flows: revenues,
expenses, current taxes, and investments. These items represent funds
actually received or paid out in a given year. However, another class of
measures depends on rules and definitions prescribed by regulatory and
accounting practices. The rest of this section discusses revenue requirements and contribution, both regulatory measures, and net income, an
accounting measure.
Revenue Requirements. Revenue requirements is a measure of the revenues needed by a company to sustain its operations. In a regulatory

724

Environment and Evolution

Part 5

environment, the rates a company is allowed to charge for its products
and services are set by regulatory agencies so that the company's revenues are equal to its revenue requirements.
The impact of a project on the revenue requirements of a regulated
company is important because of the fundamental role of this measure in
determining the allowed revenues of the company. Projects associated
with existing products or services should reduce revenue requirements
relative to a baseline without the project. This will reduce the overall
need to increase rates resulting from the pressures of inflation and other
sources. For a new customer product or service, the new investments and
expenses normally required would increase the revenue requirements but
would be offset by the new revenues.
Revenue requirements are determined as follows: Revenue requirements

equal expenses plus required earnings plus book depreciation plus operating taxes
based on revenues equal to revenue requirements (revenue requirements taxes).
The expenses used to calculate the impact of a project on revenue
requirements are the same (the operating costs) as those used to calculate
the impact of a project on cash flows. But the other components of revenue requirements are fundamentally different from the components of
cash flow measures:
• Required earnings. Revenue requirements include a charge, or
return, required by investors for the use of their funds (equity or
debt). The overall required earnings used to determine revenue
requirements are calculated by multiplying a rate of return times a rate
base. The rate base is the current value (defined as the first cost minus
accumulated depreciation) of existing capital investments. The allowed
rate of return is set by regulatory agencies, which also decide exactly
what can be included in the rate base. It is principally through this
mechanism that regulatory agencies influence revenue requirements
and, thus, control the overall rates charged to subscribers for products
and services.
• Book depreciation. Book depreciation is sometimes classified as an
operating expense in the calculation of revenue requirements or net
income (see below). However, it is important to realize that book
depreciation is not a cash flow but a regulatory or accounting artifact
that distributes or allocates the impact of an investment (a real cash
flow) over the life of the investment. Regulatory policy requires that
book depreciation be calculated on a straight-line basis with an equal
amount allocated to each year of the life of the project.
One consequence of this procedure is to "smooth" the impact of an
investment on revenue requirements and eliminate abrupt year-toyear variations due to investments with useful lives of several years.
• Revenue requirements taxes. Like the taxes used to calculate cash
flows (current taxes), operating taxes for revenue requirements depend

Chap. 18

Evolution of Products
and Services

725

on revenues, tax deductions, and tax credits. However, these components -differ from those used to calculate current taxes in important
ways:
Taxes on revenues in excess of revenue requirements are not
included.
The depreciation used as a deduction for revenue requirements
taxes is not the same as the tax depreciation used to calculate
current taxes. The calculation of book depreciation as a deduction
in revenue requirements taxes is on a straight-line basis. Depreciation as a deduction for current taxes often uses an accelerated
schedule. Moreover, the assumed life of an investment can be
different for these two calculations. 4 The depreciation rate for revenue requirements taxes is determined by the average service life of
the equipment in its capital account. For current taxes, the tax rate
is specified by tax regulation.
Finally, there is a difference in the method of treating investment
tax credits in current taxes and in revenue requirement taxes. For
current taxes, the entire tax credit associated with an investment is
taken in the year of the investment. For revenue requirement
taxes, the same tax credit is allocated equally to each year of the
life of the investment.

Contribution. Contribution is a regulatory measure defined as the cumulative discounted value of revenues minus revenue requirements over the life of a
project. This measure is analogous to the net present value (NPV) from
the cash flow point of view. When the NPV and the contribution associated with a project are calculated in a consistent manner and with a
discount rate equal to the composite cost of money, they are proportional
by a constant that depends on the tax rates:
NPV == (I-n (I-g) (Contribution),
where T is the income tax rate, and g is the gross receipts tax rate (a percentage of the gross revenues).
Both measures describe a project over its entire lifetime and should be
expected to give consistent answers. (That is, a collection of projects
should always be ranked in the same order whether the ranking criterion
uses a cash flow or a regulatory measure.)
The relationship between the two cumulative discounted measures,
NPV and contribution, cannot be extended to their annual counterparts.

4 The original investment value may also differ for the two calculations because certain
costs may be included in one calculation and not the other.

726

Environment and Evolution

Part 5

On a year-by-year basis, there is no correspondence between annual posttax cash flows and revenue requirements.

Accounting Measures - Net Income
The income statement of a company presents the course of business over
a period of time from an accountant's point of view. Quarterly income
statements for AT&T and its consolidated subsidiaries can be found, for
example, in the AT&T Share Owners Newsletter; yearly income statements
are normally an essential part of any company's annual report. Values in
the income statement describe the company's profit in terms of revenues
and costs as defined by accounting rules and regulations. Analysts and
investors use the income statement, together with a statement of a
company's assets and liabilities at a point in time and a description of
cash flows during the period of the income statement, to assess the financial and economic well-being of a company. Therefore, the year-by-year
impact of a project on the income statement is an important aspect of the
economic evaluation of a project.
Net income is the final figure (the "bottom line ") in an income statement: Net income equals revenues minus expenses minus book depreciation minus
operating taxes minus interest. Net income shares many features with the
annual revenue minus revenue requirements for a project. For example,
both depend on revenues, expenses, operating taxes, and book depreciation. In both cases, taxes and book depreciation are bookkeeping artifacts
defined by regulatory or accounting practices. However, there are also
important differences in the calculation of net income and contribution:
• The taxes used to calculate net income include not only taxes on revenues equal to revenue requirements, but also operating taxes incurred
by the total revenues. Thus, net income is fully a posttax measure.
• Net income accounts only for the interest paid to the debt holder
rather than the return required by the composite investor (a weighted
average of debt and equity holders). It is the earnings available to the
equity holder after debt obligations have been satisfied. In fact, the
annual earnings per common share (commonly used as a measure of
corporate performance by financial analysts) is calculated by subtracting
the preferred dividend requirements from the net income and dividing by the
number of common shares outstanding. Part of net income is paid directly
to the owners of a firm (the shareowners) in the form of dividend
payments. The remainder, by definition, is reinvested (retained earnings). Since retained earnings increase the total assets of a company,
the value of each share of outstanding stock is also increased. (This
book value of a share of stock should not be confused with the market
value of a share, which depends on many other factors.) Thus, the
shareowners receive net income either directly as dividend payments
or indirectly as retained earnings that increase the value of their stock.

Chap. 18

Evolution of Products
and Services

727

18.3.4 PRESENTING AND INTERPRETING RESULTS
The final stage of project evaluation is presenting and interpreting the
results of the calculation of economic measures. Standardization and uniformity of format and content are essential characteristics of this stage
because they enhance the efficiency of incorporating the results of an
economic analysis into management decisions. Preparing a standard
report including tables, graphs, and supporting text ensures that three
criteria are met:
• The report will be complete.
• Reports prepared for different projects will be directly comparable.
• Reports prepared at different times will be directly comparable.
As with the preparation of input data and the calculation of economic
measures, the trend in the Bell System has been toward standardization
in reporting. For example, at Bell Laboratories, a Bell Laboratories Executive Summary, describing the economic impact of a proposed project on
the Bell System, is currently required for all funds authorized or reauthorized by Western Electric for specific development projects.
The Bell Laboratories Executive Summary is a concise document that
meets the three criteria stated above. It consists of five principal parts:
1)

Project description. The project description presents a summary of
the objectives of the project and a brief functional description of
the hardware and software products expected to result from its
development and implementation. The project is characterized in
terms of its principal economic nature (capital saving, expense saving, or revenue producing), and any existing or potential future
projects that would be significantly affected by implementation of
the project are identified.

2)

Project implementation assumptions. Project implementation
assumptions identify and describe the major assumptions included
in the economic evaluation of the project. For most projects, the
types of information that need to be discussed include demand,
prices and tariffs, time periods (lifetimes and study periods), and
personnel levels. Besides these project-dependent data, any
changes to standard factors such as inflation rates, wages, or financial factors must also be explained.

3)

Display of economic measures. Figure 18-6 shows the display format implemented in the Bell Laboratories Executive Summary for a
typical project. These are specified measures displayed in a
required format. Charts A and B describe the impact of a project
on the Bell operating companies in terms of cash flow and regulatory measures. Chart C describes the project from the point of

EleMENTS OF OTC ANNUAL FUNDS

OTC ANNUAL MEASURES

C!

2

;
;
Ii

!

i

I

~
~

j

/

"

/

/"

\"

/

~+---~r---~~--'-----r-~~----~

~rv~~~~r-~~----'-----r-~~----'

0]'78

..

~".J

1_

11184

11182

11180

.,

11182

11184

- - - OTC POSTTAX FUNDS
-- " -

REYENUE MINUS REYENUE REQUIREMENTS

CUMULATIYE MEASURES:

•

- - - REY!NUE

- - - - - INY!STMENT

-

- - " - - PRETAX FUNOS

- - EXPENSE

CONTRIBUTION - S43.24M
IROR - 32.24%
CHART B

CHARTA

;
Ii

~

i

~

{"../",\

\
\

"

/
/

!

"

"

"

\

~'"

I,
~
11178

~

SUMMARY OF CUMULAnY! DISCOUNTED FUNDS

WE MEASURES

g

.... ,

~":~ - - - ~~........ - ;"~

---- --1_-----1_
11184

J
- - - - - NEW PRODUCT SAleS
-

- -

- - - -

NEW PRODUCT CDF

SALES (SMI

HAROWARE

SOFTWARE

TOTAL

NEW PRODUCT
TOTAL EFFECT

23.61
74.711

0.00
0.00

23.61
74.79

-'-OTC

- - - - - WE TOTAL EFFECT
- - - - - BELL SYSTEM

TOTAL EFFECT SALES

NET PRESENT YALUE (SMI
WE
1.08

OTC
20.26

BELL SYSTEM
22.19
CHART D

CHARTC

BTL DEVELOPMENT SUMMARY TABLE
FUNDS

.seENI.

.1.IlZL

19H

TOTAL
BEQI.IUI.

HARDWARE (SM)

0.20

1.10

0.10

1.20

SOEIWABE ISMI

..Q..1IL

...Q..QO.

.o.JXl..

--D.JKL

TOTAL EXP (SM)
CAPITAL (SM)
EMTS-YEARS

0.30
0.00
2.00

1.10
0.00
19.00

0.10
0.00
2.00

1.20
0.00
21.00

Figure 18-6. Example of a standard display in the Bell Laboratories Executive Summary.

Chap. 18

Evolution of Products
and Services

729

view of Western Electric, and the table describes the impact of the
project on Bell Laboratories. Finally, chart D plots the impact of
the project on the year-by-year value of the cumulative discounted
cash flow for the operating companies, Western Electric, and the
consolidated Bell System. The net present values for these three
curves are explicitly displayed on this chart. The information in
Figure 18-6 was chosen to satisfy the needs of the Bell Laboratories
Executive Summary; for other standard reports, different measures
or different formats might be more appropriate.
4)

Interpretation of economic measures. The analyst should interpret and explain the significance of the results of the calculation of
economic measures. The salient features of the analysis should be
addressed and overall conclusions summarized.

5)

Sensitivity to alternative assumptions. An essential part of any
economic evaluation is an analysis of the sensitivity of the results
to the input data. This includes an analysis of the risks associated
with project implementation. Both pessimistic and optimistic versions of the input data should be used to complement the results
found in the nominal study based on the most likely values of the
input data.

18.4 APPLICATION OF NEW TECHNOLOGY
The success of the Bell System has always depended on products and services fed by a stream of technological advances. In some cases, the
advances have reduced the first cost or the operating expenses. Sometimes technology has led to the introduction of an entirely new product
or service or to the addition of new features at negligible additional costs.
This dependence on technology requires a close cooperation between
those who plan and develop the products and services and the scientists
and engineers who develop the underlying technologies. In the Bell System, the overall process has been described by Ian M. Ross, current
president of Bell Laboratories, as "organized creation and application of
technology. "
The following sections discuss the coupling of technology and market
needs and describe two examples of the effect of new technology on
products and services.
18.4.1 MATCHING TECHNOLOGY TO BELL SYSTEM NEEDS
Creating a new product or service is virtually synonymous with the
application of new or improved technology. But, while technology
shapes new markets, it must itself be shaped by the marketplace.

730

Environment and Evolution

Part 5

The process of developing new technology is somewhat analogous to
using two teams, one on each side of a mountain, to drill a tunnel
through the mountain. Both teams are ready to burrow ahead, but
without communication and coordination, their efforts to construct a useful product (in this case, a tunnel) would prove futile. Specifically, the
optimal point from which to dig may appear quite different if judged
with the limited vision available on only one side of the mountain. So it
is with new telecommunications technology. The technology team must
be aware of product or service needs where the fruits of its work might
be used. Likewise, the team responsible for the development of a new
product or service must be aware of the potential of new technologies
that it could successfully apply. The most useful advances and the most
successful products arise from the combined and closely coordinated
efforts of technologists and system designers.
Matching new technology to market needs in the Bell System has
been an important role of Bell Laboratories systems engineering and
development groups. Interactions with AT&T and operating companies
provide information on market needs. Interactions with technology areas
in Bell Laboratories and knowledge of new technology developed elsewhere provide the means for meeting those needs. The latter interactions also may result in ideas for new, marketable products and services. Because many organizations in the Bell System are involved in the
overall process, committees and forums have been established to help
bridge organizational and geographic boundaries.

18.4.2 EXAMPLES OF APPLYING NEW TECHNOLOGIES
No other recent technology can match the impact that integrated circuits
have had on Bell System products and services. By packing more and
more functions onto a single piece of silicon, microelectronics5 technology has led directly to equipment that is smaller, less costly, cheaper to
operate, and more reliable.
While successive generations of integrated circuits, such as memories,
can provide the same electrical function at progressively lower costs,
these cost reductions can be overshadowed by the savings that they
trigger in other parts of the product. Packing more onto silicon reduces
the number of circuit boards (and the associated costs of assembly and
testing), connectors, backplane wiring, and frames. In addition, operating
expenses become lower as a result of higher system reliability and lower
requirements for operating power and cooling. Thus, even if there were

5 Scientific American 1977 is an excellent introduction to the technology and applications of

microelectronics.

Chap. 18

Evolution of Products
and Services

731

no reduction in the cost per function of the integrated circuits themselves, there would still be an incentive to use devices of smaller size and
increased capability to achieve lower equipment costs.
More frequently, modification of an existing product represents an
opportunity to use "cheap" silicon to add new features. Often this adds
intelligence to the network and provides more capabilities for the customers. Some products and services become economically feasible only as
the cost per electronic function decreases sufficiently.
Thus, integrated circuits can reduce product costs and operating
expenses, can provide new features at negligible additional cost, or can
make a new product or service feasible. The following examples illustrate most of these characteristics.
Cost Reduction Example - Memory for Electronic Switching Systems
A dramatic example of the results brought about by advances in
microelectronics is the evolution of electronic switching system memory
portrayed in Figure 18-7.
By the mid-1970s, semiconductor memories, which had been deemed
too expensive at the I-kilobit (IK) level, finally displaced magnetiC
memories. First came the 4K random-access memory (RAM), introduced
into the main store for the lAESS switching equipment in 1977. As a
result of advances in semiconductor technology, it was supplanted by the
16K RAM in 1978. In 1981, the 64K RAM took over; the 256K memory
will be introduced in 1984. While the device costs (measured in cents per
bit) were dropping significantly, the cost of memory (measured at the system level) was dropping even more rapidly. By eliminating frames,
boards, and backplane wiring, the system costs for memory dropped by a
factor of more than 30 from 1977 to 1981. As memory costs decline, more
memory tends to be used. This is partly due to the shift to higher-level
languages for the stored-program control-in effect, trading relatively
expensive software development for cheaper hardware.
But cost savings are not the whole story. Figure 18-7 shows an important reduction in power dissipation accompanied by an increase in speed.
Both characteristics reflect the improvements that result from shrinking
the dimension of the circuits on the devices. In addition, it is well
known that the reliability of connections is higher on silicon than in any
other part of the system. Thus, moving a connection from a printed wiring board to a silicon chip improves the system reliability.
It 'lffiight be tempting to conclude that in semiconductor memory
design, the technologist could simply press ahead to higher levels of
integration without consulting system designers. After all, the benefits of
previous designs are obvious. But there are important tradeoffs to be considered in designing these memories. One of them is the balance

3B20
SEMICONDUCTOR

I
NO.1
FERRITE

NO. 1A
SEMICONDUCTOR

I
SHEET

CORE

I

I
14K RAM

16K RAM

256K RAM

I

RELATIVE VOLUME

3 ,840

320

80

20

2

POWER
(microwatts per bit)

2,800

175

70

20

4

SPEED
(microseconds)

5 .50

5 .50

1.40

.70

.55

.55

MEMORY SHOWN
(me9abytes)

1. 18

1. 18

1.18

.79

1.05

1.05

YEAR OF INTRODUCTION

1965

1971

1977

1978

1981

1984

Figure 18-7. Electronic switching system memory evolution.

Chap. 18

Evolution of Products
and Services

733

between speed and power. If the device designer knows the speed
requirements of the system, the memory can be tailored appropriately. A
device operating at a speed that is extremely high would lead to a wasteful dissipation of power. Should the memory be organized to output a
single bit or a full byte? Again, the answer lies with the system designer.
It is the ongoing dialogue at Bell Laboratories between the system
designer and the technologist that has been a key element in bringing
the appropriate innovative changes to the Bell System.

New System Example - Digital Access and Cross-Connect System
Assessing precisely the impact of new technology in shaping new systems is more difficult than evaluating cost reductions. It is seldom possible to identify dear-cut cases where a new development leads immediately to a new product or service. More frequently, someone realizes that
both technology and market potential exist. The pocket calculator and
video games are examples from the consumer market. Each appeared
when the evolution of microelectronics had reached the point where
most of the functions could be performed by a few devices, bringing the
cost of the product into a marketable range. Analogously, the Digital
Access and Cross-Connect System (DACS)6 was developed when largescale integration became the state of the art, and it became feasible to
place a very important function, the time-slot interchange (TSI), on a
single chip of silicon. This provided connecting and test access functions
that were much improved and less costly at a time when there was a
growing need to handle the large volume of circuit rearrangement.
In DACS, cross-connection from incoming to outgoing circuits is
accomplished by a combination of time-division and space-division
switching. The former is performed by the TSI chip (see Figure 1S-8),
which takes a stream of twenty-four 12-bit words and rearranges them
into 256 time slots of a data bus. (Section 7.3.3 discusses the operation of
a TSI.)
An examination of Figure 1S-S suggests why the device only became
feasible in the very early 19S0s: The chip accommodates almost 4000 bits
of "static" metal-oxide semiconductor memory. In addition, even more
area is taken up by the control logic. This level of complexity
represented the state of the art at that time. In fact, when first manufactured, the TSI chip was Western Electric's largest integrated circuit.
The TSI chip is not the only complex integrated circuit in DACS. In
combination with other large-scale integrated circuits, it makes possible a
compact and flexible hardware design that is expected to create an attractive market for the system.

6 See Section 9.4.3.

256·WORD
CONTROL RAM

CONTROL
LOGIC

24·WORD
DATA RAMS

24·WORD
ALTERNATE
MESSAGE STORE

Figure 18-8. The Bell System's largest very large-scale integrated chip in 1980
(429 by 328 mils)-a time-slot interchange (TSI).

18.5 INTEGRATION OF NEW AND OLD SYSTEMS
18_5.1 THE NATURE OF THE PROBLEM
The tradition of designing telephone equipment for long life has
influenced depreciation rates and the tariffs approved by public utility
commISSIOns. However, tariffs have recently given consideration to
obsolescence as well as equipment life, because equipment offering new
technology, economic savings, and new features usually phases out old
equipment. Since the rate of replacement is usually much slower than
the evolution of new designs, though, several generations of equipment
will usually be in operation simultaneously. Local switching systems are
a good example of this phenomenon. As seen in Figure 18-9, from the
1930s to the 1980s, telephone lines have b een served by various switching

734

100r------------------------------------------,
80

en
z

0
::::i
oJ

!

ELECTRONIC
SWITCHING
SYSTEMS

60

U)

III

Z

NO. 1 CROSSBAR
SYSTEM

::::i
III

>

~

:5

.........

40

;:)

"

:Ii

;:)

(J

20

0~

1930

__4 -_ _

~

__

1940

~

__

~

__

~~

1950

__

~

1960

__

~

_ _ _ L_ _

1970

~

_ _~

1980

YEAR

Figure 18-9. Automatic switching systems
serving Bell System lines, 1930-1980.

systems. The birth, growth, and decline of switching technologies is evident, along with the fact that a mixture of switching technologies will
usually exist at any point in time. Eventually, old technologies are
entirely replaced by newer systems?
The oldest type of Western Electric switching system currently in service in the Bell System is the step-by-step, which was first used in the
Bell System in 1919. The continued serviceability of this older equipment
along with limitations on capital for its replacement are the primary reasons for its continued presence in the system. However, step-by-step systems are expected to be phased out completely by about 1990. New electronic switching systems (both analog and digital) are currently being
used for the expansion of service as well as for replacement. Since
different electronic switching designs are being developed with time,
various electronic switching systems coexist in service just as various
electromechanical switching systems still do.
7 The last of the panel systems was retired in 1982.

735

736

Environment and Evolution

Part 5

Each new design of telecommunications equipment must consider
several factors: (1) new switching and transmission systems must be
designed so that interfaces with the variety of systems in service will
function properly; (2) new designs may be restricted by building characteristics such as floor loading, frame height, temperature variations, or
electrical induction; and (3) customers will react to changes in operations.
Standardization of interfaces, environments, and operations allows for a
smooth introduction of new equipment but, at the same time, may restrict
the economies and service features that new technologies might realize.
Most new services must be made available in areas served by old and
new equipment and at the same price to the customer. The biggest challenge of a new service feature is to retrofit it into the old equipment
without losing money. As a specific example of integration of a new system, the following section discusses some of the interface problems that
had to be resolved when TOUCH-TONE calling was introduced.
18.5.2 SOME PROBLEMS IN INTERFACING TOUCH-TONE SERVICE
WITH OLDER EQUIPMENT
The rotary dial represents a highly successful coupling of human capability and telephone technology. For reliability and economy, the rotary
dial has been unsurpassed since the earliest days of automatic switching,
and it is more reliable than a pushbutton dial. The convenience of pushbuttons, however, was widely recognized. The advent of the transistor
offered an economically feasible way to implement pushbutton dialing.
TOUCH-TONE dialing was introduced in 1963. It can be tariffed as a
premium service because it is a convenience to the customer. Since it is
faster than rotary dialing, it ties up registers for shorter intervals in
common-control offices. It also provides two additional characters (# and
*), which are useful in providing new capabilities.
However, as a premium service, TOUCH-TONE dialing had to be made
available on individual lines in areas that were predominantly rotary dial.
It had to be possible to get service from any type of central office to avoid
number changes for customers. Common-control offices that receive customer-dialed digits in register circuits had to be designed to receive signals from TOUCH-TONE telephones as well as rotary-dial signals.
Conversion from rotary-dial service to TOUCH-TONE service required the
installation of a TOUCH-TONE telephone and a central office change to
access an appropriate register. To avoid problems that would occur if
these two events were not synchronized, registers that accept signals
from either rotary or TOUCH-TONE dialing were designed. As an added
benefit, the registers accommodate party lines with a mix of TOUCHTONE telephones and rotary-dial station sets or customers who have both
types of telephones on their premises.
Because step-by-step switching systems use direct control of the
switches from the station, conversion to TOUCH-TONE dialing is difficult

Chap. 18

Evolution of Products
and Services

737

and relatively expensive. All TOUCH-TONE telephone (or mixed) lines
must be grouped together and new switching stages introduced so that
these lines may access appropriate registers. The registers must then generate dial pulses to drive the succeeding switches.
Another problem in introducing TOUCH-TONE service is related to
dial tone. When dial tone is sent to the customer, some of its energy is
reflected back to the central office by the station set, and in early installations, this reduced the sensitivity of the receiver at the central office to
the first digit dialed. It was necessary to use a new dial-tone signal composed of frequencies that did not interfere with TOUCH-TONE telephone
operation in all central offices accommodating TOUCH-TONE service.
The new signal also had to be similar enough to the old one to be
accepted as a dial tone by customers.

18.6 HUMAN FACTORS IN THE BELL SYSTEM
The Bell System, historically, has recognized the importance of factoring
people's capabilities and behavior into the design of telephone networks,
customer equipment, and services. As early as the 1920s, human hearing
and speech characteristics were studied in order to set standards for
transmission quality. Subsequently, Bell Laboratories engineers and
behavioral scientists also became concerned with considerations such as
the shape, size, and weight of telephone sets; the design of telephone
dials; and, later, the development of direct distance dialing codes:
As the value of human factors became more apparent, activities
expanded to include the consideration of Bell System employeestechnicians, business office employees, telephone operators, and
managers-as well as customers. Today, human factors specialists are
helping to improve the productivity and effectiveness of Bell System
employees by contributing to the design of computer-based operations
systems, automated operator systems, stored-program control systems and
services, and customer-service procedures.
Bell Laboratories does more human factors work than any other organization in the world outside of the United States government. 8 The following sections provide a brief description of the human factors discipline at Bell Laboratories and some illustrative applications.

18.6.1 THE HUMAN FACTORS DISCIPLINE
The goal of human factors at Bell Laboratories is to ensure that Bell System products and services are designed for the people who use them. For
customers, this means ensuring that a product or service satisfies customer needs and preferences, operates easily, and is fully supported by good
8

Bell Labs News, March I, 1982, p. 4.

Part 5

Environment and Evolution

738

instructions and service delivery systems. For employees, it means allocating system tasks among people and system components and ensuring
that employees can perform their duties efficiently and effectively in
rewarding and meaningful jobs. To deal with these issues, human factors
specialists must have two types of knowledge. They must know about
people-their perceptual skills, their learning abilities, their physical
characteristics, how they make decisions-and they must know methods
for analyzing human performance in systems. Because of the knowledge
required, many human factors specialists have been trained in experimental psychology, which emphasizes methods for measuring and analyzing
human performance.
Some of the information used in human factors work is available in
handbooks and other references, particularly facts that are useful in physical design (see Figure 18-10). But, because of the absence of a large body

KEEP LIGHTS ABOVE
TO AVOID GLARE

OPTIMUM MANUAL
WORK ELEMENT
ZONE BETWEEN
ELBOW AND
SHOULDER
HEIGHT

-L

+ - NORMAL SIGHT LINE

-

15 IN.

AREAS
A - TYPING. TYPEWRITER
KEYBOARD
(IF ADJUSTABLE, CENTER
OF KEYBOARD AT ELBOW
HEIGHT) LOWEST LEVEL
SHOULD BE 26 IN. UPPER
ROW NO HIGHER THAN 31 IN.
B - PRIMARY DISPLAY
AND WORK ELEMENTS
C - PRIMARY DISPLAY,
SECONDARY WORK
ELEMENTS
D - EMERGENCY DISPLAY,
SECONDARY WORK
ELEMENTS

NOTE: 5th-95th % OPERATORS MALE AND FEMALE

E - SECONDARY DISPLAY
AND WORK ELEMENTS

Figure 18-10. Recommended dimensions for seated work position.
(Example of information available from human factors reference sources.)
Redrawn from Van Cott and Kinkade 1972, which is based on data from
Dreyfuss 1959, Kennedy and Bates 1965, and Woodson and Conover 1964.

Chap. 18

Evolution of Products
and Services

739

of useful theory, much design information must be obtained on a caseby-case basis. Data collection techniques used in human factors analysis
include laboratory experiments, field observation studies, evaluations of
alternative designs, and user interviews.

18.6.2 DESIGNING FOR CUSTOMER SERVICES
Until recently, customer interactions with the telephone system have
been relatively simple, involving only placing and receiving voice calls.
But advancements in technology have provided the opportunity to create
products and services of great complexity. And, as services become more
complex, the interactions between the services and their users become
more complicated. Human factors specialists have the responsibility of
minimizing the problems this can create.
For example, Advance Calling is a potential new service that has been
installed for field trials. This service enables a customer to record a voice
message, address it to any telephone that can be dialed directly, and
specify the time at which the message is to be delivered-all from a standard telephone set. A list of the human factors questions that arise in
designing such a service would include:
• In what sequence should the necessary information (voice message,
time, destination telephone number) be entered, or is it unimportant?
• Should the time and destination telephone number be repeated back
to the customer for verification? (This complicates the interaction.)
• Should the recorded message be replayed to the customer, and if
found unacceptable, should the system permit recording of a new
message without requiring the customer to hang up and start over?
• Should messages be accepted for delivery during late night hours?
Should attempts to deliver messages scheduled earlier be continued
into late night hours?
• What sorts of instructions should be provided for message delivery
time entry? Should a 12- or 24-hour clock be used? If a 12-hour clock
is used, how should morning and afternoon be specified? If a message
is to cross time zones, which time should the customer specify?
• Should written instructions be available to the customer?
should they emphasize?

What

• How many times should the message be played to the recipient?
• How good must the voice quality of the recorded message be?
• How can the instructions be designed to satisfy the needs of both
inexperienced and regular users?

740

Environment and Evolution

Part 5

The human factors specialists who actually addressed these questions
used a variety of methods. First, they used logic and experience to identify the most important questions and some possible solutions. A simulation was then constructed that enabled potential customers to attempt to
place advance calls. Errors, other difficulties, and user reactions were
measured and led to changes and additional testing. Laboratory tests
were performed to assess the effect of various digital encoding rates 9 on
users' abilities to recognize callers and understand their messages. Surveys were conducted of customer knowledge of 24-hour "military" time
and the differences between time zones. Studies previously conducted
for a related service provided information about recipients' problems and
the number of times a message should be replayed to the recipient.
Because of the close interaction between the procedures the customer
must follow and the hardware and software, design decisions resulting
from these studies were made in close cooperation with both the systems
engineering and development areas. And, because laboratory analysis
does not perfectly reflect real life, the initial introduction of Advance
Calling included a built-in feature for tracking customer performance.
Many other recently developed customer systems have undergone
intensive human factors study, and each had its own unique problems.
In Advanced Mobile Phone Service (see Section 11.4.1), questions of the
impact of mobile telephone usage on driver performance required in-car
studies (see Figure 18-11). Design of the Automated Coin Toll System
(ACTS)10 required studies to determine the length of time needed to
insert coins and the sorts of situations in which the customer should be
connected with a live operator. Automated Calling Card Service (see Section 11.3.1), PICTUREPHONE meeting service (see Section 11.5.3), the
HORIZON communications system, and other recently developed products and services have received similar scrutiny to ensure their "friendliness" to the user.
18.6.3 DESIGNING FOR EMPLOYEES
Human factors has been especially relevant to the design of large operating company data-base systems, such as the Trunks Integrated Records
Keeping System (TIRKS),l1 which replace tedious and labor-intensive
manual operations. Such systems depend critically upon the accuracy
and timeliness of employee-entered data. They also depend on a reasonable allocation of tasks between people and computers. For example, routine data handling and error checking are best done by a computer, while
9 The caller's message is encoded and stored in digital form.
10 See Section 10.4.1.
11 See Section 14.2.1.

Figure 18-11. Human factors simulation. Laboratory and field simulations
playa significant role in the design of user interfaces. Here. human factors
engineers use a television camera to study the ease of placing and receiving
calls while driving. Tests showed that placing a call with Advanced Mobile
Phone Service telephones is as easy as tuning a car radio .

people are better at handling the more complicated and less predictable
parts of the job (and find them more interesting).
The human factors role in designing such a system includes:
• analyzing the organizational environment in which the system must
operate
• analyzing the tasks to be performed
• dividing tasks between people and machines
• defining required employee skill levels
• specifying training and on-the-job documentation.
The success of human factors work in the design of operations systems
has led to involvement with more traditional employee groups as well.
Below are just a few examples:
• designing and evaluating new hardware for the outside plant, including cable cross-connection terminals and interfaces, splicing equipment, and tools
741

742

Environment and Evolution

Part 5

• designing and modifying the Traffic Service Position System (TSPS)12
operator console and working on computerized directory assistance
systems
• developing improved computerized tools for outside plant planning
and engineering (see Section 14.4) and systems for detecting and locating faults in local loops.

Future challenges in employee human fact0rs rival those on the customer side. Telecommunications is becoming more computerized and less
manual, and Bell System employees must make the transition. At the
same time, there is a growing emphasis on the quality of work life, which
is substantially determined by the tasks one performs. The challenge of
human factors work on employee issues will be to improve employee
productivity and the quality of work life as telecommunications technology becomes more mechanized and complex.

18.7 QUALITY ASSURANCE
lS.7.1 THE ROLE OF QUALITY ASSURANCE
Primary responsibility for the quality of products used throughout the
Bell System resides with the organizations directly involved in their
design, development, manufacture, or supply. Bell Laboratories development organizations are responsible for the quality of design, including
specifying requirements and evaluating performance. Western Electric,
in carrying out its supply functions, is responsible for the quality of
manufacture and repair, including engineering and installation. In addition, Western Electric and other suppliers are directly responsible for the
operation of effective, on-line quality control programs.
The distinction between quality control and quality assurance is
important. Quality control is the set of procedures used by the supplier to
provide sufficient process control over the machines, personnel, and
material necessary to meet acceptable quality criteria consistently and
economically. Quality assurance, on the other hand, provides continuing
independent verification of satisfactory results from the standpoint of the
customer-user. The quality assurance system is structured to determine
the effectiveness of quality controls used by the designer and producer

12 See Section 10.4.1.

Chap. 18

Evolution of Products
and Services

743

and is effected through feedback directed to the responsible management
of Bell Laboratories, Western Electric, and outside suppliers. 13
This discussion focuses on the quality assurance system. The basic
system consists of two major elements (as shown in Figure 18-12). The
first is a highly structured quality audit-a system of inspections carried
out within Western Electric and at the interfaces between Western Electric and the Bell operating companies. The second element, quality
assurance monitoring, involves a variety of activities that confirm the
effectiveness of prior controls or identify problems requiring corrective
action. Section 18.7.2 discusses rating of quality through formal, ongoing
quality assurance audits, and Section 18.7.3 discusses quality assurance
monitoring. (More detailed information and numerous examples of the
wide variety of recent quality assurance activities can be found in a series
of articles in the Bell Laboratories Record October through December 1978.
Peters and Karraker 1975 contains a description of the Western Electric
role in assuring quality.)
Periodic reports highlighting quality problems identified by the quality assurance activities are distributed within Bell Laboratories and, more
importantly, to the upper management of Western Electric. Prompt
response by Bell Laboratories and Western Electric management resolves
the majority of quality problems. Occasionally, corrective actions require
additional technical effort involving the Bell Laboratories Quality
Assurance Center.
Many products are purchased outside the Bell System for use by the
Bell operating companies or for assembly into products manufactured by
Western Electric.
Usually these products are procured under
specifications provided by Bell Laboratories or Western Electric, and quality inspections are performed at the source. In addition, a Quality Surveillance System (QSS), under the administration and guidance of AT&T,
has been developed for non-Western Electric (general-trade) communications products purchased directly by the Bell operating companies. The
basic responsibilities of the QSS include: (I) technical evaluation and
determination of risk involved in product use (for example, safety, service, economy), (2) evaluation of the supplier, (3) evaluation of product
quality and reliability, (4) an inspection program (including criteria for

13 Many of the basics of both quality control and quality assurance originated within the
Bell System. w. A. Shewhart of Bell Laboratories first published the concept of a control
chart, widely used in quality control throughout the world (see Shewhart 1926).
Shewhart, Dodge, and others at Bell Laboratories continued to develop the statistical
foundations of quality assurance auditing and refinements in the implementation of
quality assurance through the 1950s. Recent contributions have been towards more
economical and statistically powerful tools. Additional Reading for Chapter 18 at the
end of the book lists documents that contain more information on the development of
quality control and quality assurance.

DEVELOPMENT

Figure 18-12. Major elements of the quality assurance system.

Chap. 18

Evolution of Products
and Services

745

setting standards, scope of inspection, and acceptance sampling procedures), and (5) field performance monitoring. Although the basic elements of the general-trade QSS are similar to the major functions of the
quality assurance program for Western Electric products, significant
differences in implementation details may exist because of the proprietary
nature of general-trade designs and manufacturing processes. (AT&T
1983 contains additional information on the Quality Surveillance System.)

18.7.2 QUALITY ASSURANCE AUDITING
Quality assurance audits are conducted on products manufactured,
installed, and repaired by Western Electric. For example, Western Electric
quality assurance personnel take samples of units from the ends of production lines and subject them to extensive tests and inspections. Sampling and inspection are repeated, and the results accumulated over a rating period (currently about six weeks). The Bell Laboratories Quality
Assurance Center compares the cumulative statistics to quality standards
specified for each product. These quality standards are intended to
minimize total cost to the Bell System, that is, to ensure an economic
level of quality that is consistent with the costs of manufacture and the
subsequent costs of use.
Data on products are gathered from the numerous Western Electric
locations at audit points (shown in Figure 18-12) and are transmitted to a
computerized data base maintained by the Quality Assurance Center. If
the data confirm that quality is at a satisfactory level, the results are
recorded and filed for future reference. If the data indicate with a reasonably high degree of certainty that the product is not meeting expectations, the results are reported to management and included in an exception report. The audit data are also used to conduct special studies and to
analyze quality problems in detail when the need arises. The following
paragraphs describe the auditing process.
Audit Structures
In general, quality assurance auditing for a product addresses these basic
questions:
1)

Was it made correctly?

2)

Will it operate properly?

3)

Will it continue to operate for a reasonable period of time?

Since it is neither feasible nor desirable to formally rate the quality of
each unique product in the many product lines manufactured by Western
Electric, products are grouped into rating classes according to major system, similarity of function, Western Electric location, etc. For example,

746

Environment and Evolution

Part 5

exchange area cable is grouped into three major categories-pulp, PIC 14
air-core, and waterproof-and rated accordingly at each of five manufacturing locations. Within each of these categories, the audit results are
often subdivided according to the quality characteristics examined. For
example, each exchange area cable category is rated by (1) electrical
transmission parameters and (2) visual and mechanical construction
characteristics. In total, there are about 2000 subdivisions called scoring
classes. The audit for each of these classes is a highly structured system of
checks and inspections consisting of these basic procedures:
• Checking - defining the scope of an audit: tests to be performed and
attributes to be checked on each unit of the sample
• Appraising - classifying defects found during the checking, and
assessing their seriousness: procedures to be used for declaring a product to be nonconforming and subject to corrective action
• Establishing standards - determining the expected level of quality
• Sampling -

selecting units to be inspected

• Rating - comparing audit results with established standards and
assessing the statistical significance of departures
• Reporting - portraying rating results in regularly issued reports and
highlighting quality problems for corrective action.
The following paragraphs discuss these procedures in more detail.
Checking. The tests, checks, and inspections performed on each unit of a
sample are intended to verify design specifications, engineering requirements (specified or implied in applicable product drawings), generally
accepted good workmanship and manufacturing practices, and basic
design or application considerations (such as system performance and
function, reliability, appearance, life, interchangeability, maintainability,
and cost of use or repair). Specific product requirements and testing procedures are derived from a variety of other sources and include visual,
mechanical, and electrical tests.
Appraising. Bell System products are currently rated in terms of three
alternative quantities: defects per unit, percentage defective, or demerits
per unit. A defect is defined as a failure to meet a requirement, for example, a measurement exceeding a specified limit. In a defects-per-unit
audit, the average number of defects per unit is determined over the rating period. In a percentage-defective audit, a unit of the sample is considered defective if it contains one or more defects, and the percentage of
14 Plastic-insulated cable.

Evolution of Products
and Services

Chap. 18

747

defective units in the total sample is the measure of quality. In a
demerits-per-unit audit, defect seriousness is quantized, as shown in Figure 18-13, according to the classification guidelines summarized in
Table 18-1. The numerical sum of the demerits observed during the
rating period normalized by the number of units sampled is the measure
of quality.
When many different defect items are included in the scope of an
audit, there is some risk that the existence of major defects will be
masked in an overall rating. To mitigate this risk, the Bell Laboratories
Quality Assurance Center frequently specifies a dual rating structure,
wherein the major defect types are rated separately, in addition to the
overall composite rating of the product. Such dual rating is appropriate
when, for instance, Bell operating company service needs specify relatively few incidences of major (class A and B) defects (discussed in more
detail in Establishing Standards below).
In another type of dual rating structure (for example, in the auditing
of electronic switching systems and circuit packs), test results and workmanship are rated separately. The quality of workmanship during
manufacture of a product is generally considered to be an indicator of
long-term reliability. Serious defects of workmanship (such as poor
solder connections) can affect operability in service and may not be
readily detected by electrical tests at the time of inspection. This was

100 -

D(X)--...

...a:

en
~

50-

1&1
Q

10 -

____

J

A
D
C
B
~~======L-------L-------~------- X
DEFECT SERIOUSNESS CLASS

SPECIFICATION
LIMIT

Figure 18-13. Assessing seriousness of defects.

TABLE 18-1
DEFECT SERIOUSNESS CLASSIFICATIONS
Class
Type
B
50 Demerits

A

100 Demerits

Major

1.

Will surely cause an
operating failure of the
unit in service.

1.

Will probably cause an
operating failure of the
unit in service.

2.

Will surely cause
intermittent operating
trouble.

2.

3.

Will render unit totally
unfit for service.

Will surely cause trouble
less serious than an
operating failure, such as
substandard performance.

3.

Will surely involve
increased maintenance or
decreased life.

4. Is apt to cause personal
injury or property
damage under normal
conditions of use.

4. Will cause a major
increase in installation
effort.

5. Has extreme defects of
appearance or finish.
C
10 Demerits

Minor

D
1 Demerit

1.

May possibly cause an
operating failure of the
unit in service.

1.

Will not affect operation,
maintenance, or life of
the unit in service.

2.

Is likely to cause trouble
of a nature less serious
than an operating failure,
such as substandard
performance.

2.

Has minor defects of
appearance, finish, or
workmanship.

3.

Is likely to involve
increased maintenance or
decreased life.

4. Has significant defects of
appearance or finish.

Chap. 18

Evolution of Products
and Services

749

particularly true for the older, wired electromechanical equipment. However, with modern, miniaturized electronic equipment, test-type auditing
has gained increased importance and become efficient to implement, particularly with systems combining hardware and software.
Another type of audit, involving a check of early-life failures and
longer term reliability performance, has also gained prominence. In this
type of audit, devices (such as integrated circuits), circuit packs, or systems (such as complete private branch exchanges) are tested for a
specified period of time, and the cumulative test failures, indicative of
in-service trouble rates, are used as the measure of the level of quality.

Establishing Standards. Usually it is neither feasible nor economical to
expect zero defects during continuing production (except for safetyrelated items). A quality standard is, therefore, an estimate of the
economically optimum value of the expected level of the measure of quality. Responsibility for specifying these standards resides with the Bell
Laboratories Quality Assurance Center. Setting standards should be an
independent and unbiased process, balancing the possibly conflicting
pressures of production, the parochial views of design and development,
the difficulties encountered during installation, the strong desire to hold
down the costs of field operation, and of course, the expectations of the
customers or end-users of the products and services. The process of setting a standard also should recognize the scope and structure of the audit
to which it applies.
For example, to set standards for a demerits-type audit (where the seriousness of the defects is quantized), in which there are X A , X B , Xc, XD of
type A, B, C, and 0 defects, respectively, the total number of demerits
(TD) will be:
TD == 100XA

+ 50XB + 10Xc + X D ,

where the Xs are Poisson 15 random variables. If AA' AB, AC, and AD are
used to denote the means of the respective distributions when the production process is at standard, then, setting standards in demerits-type
audits usually consists of specifying AA' AB' AC, and AD. This is accomplished by taking into account the impact of the different levels of defect
seriousness as specified in Table 18-1 on the Bell operating companies. In
addition, manufacturing process capabilities must be considered. These
studies are frequently carried out empirically, using historical audit data
on similar or closely related products. Such an approach is called the base
period study method, and it presumes that over a period of time the
conflicting forces of operating company desires for improved quality and

15 Chapter 5 discusses Poisson distribution.

Environment and Evolution

750

Part 5

the producer's attempts to hold down costs have reached, or at least
closely approached, an equilibrium value of the level of quality.
This equilibrium model for the optimum cost standard is depicted in
Figure 18-14. The standard is defined to be the quality level yielding
minimum total cost, that is, the minimum of the sum: field cost plus

manufacturing cost.
A viable static model is obtained by recognizing that field cost
increases as quality degrades. Over the range of interest, the field cost is
assumed to increase directly as a power function of the quality level.
Manufacturing cost is assumed to increase as quality improves, and this
relationship can reasonably be modeled as an inverse power function of
the quality level.
Assigning actual field cost values can be very difficult in practice.
These costs should take into account both the cost of repair or replacement and the more intangible cost of adverse customer reaction or product reputation.
For major defects, as defined in Table 18-1, it is highly likely that
repair activities will be required. The cost of these activities can vary
widely, depending on the type of product, location, etc. The extent of
adverse customer reaction to, for example, failure of a packaged electronic
product will depend on the importance of the system in which it is used
(how many customers are affected when the system experiences a service

l-

f/)

o(J

QUALITY LEVEL

Figure 18-14. Equilibrium model for the optimum cost standard quality,

Us.

Chap. 18

Evolution of Products
and Services

751

problem) and the criticality of the function of the unit (how seriously a
customer's service is affected).
Since precision in these cost estimates usually is not feasible, the
parameter values are quantized. For example, the constant of proportionality for the field cost, Ct , may take on three relative levels: high,
normal, and low, depending on the relative weighting of repair costs and
the estimated degree of adverse customer reaction as determined by system importance and criticality of the product's function.

Sampling. Thorough inspection of a single telephone set may cost, at
least, several dollars; other more complex equipment may cost hundreds
of dollars to inspect. Therefore, it is not feasible to audit each unit when
large numbers of units may be manufactured or installed under similar,
usually well-controlled conditions. Hence, the optimum quality
assurance strategy is to perform thorough tests on a few units and use the
results to indicate the quality of the total production. The audits, in general, are designed to monitor the long-term quality of a continuous
stream of products, rather than the acceptability of any given lot.
Since a sample is used to estimate the overall level of quality for a
scoring class, it should be representative of the ongoing production. This
is achieved by randomly selecting samples of units from randomly
selected lots suitably distributed over the full rating period. The
optimum sample size is determined by an economic model that forms a
basis for the Universal Sampling Plan (USP), described in detail in Hoadley
1981b.
The fundamental concept underlying the USP is that more effective
audits enhance the information feedback loops and, therefore, result in
better quality due to management actions. Improved quality reduces field
maintenance cost, but this must be balanced against increased audit cost if
larger samples are selected. The economic model underlying USP portrays this tradeoff. An optimum sampling plan will minimize the total
cost: Kb(audit cost) plus field cost, where the constant Kb is determined by
a constraint on the overall auditing budget.16 An audit cost can be
modeled 17 as a fixed or overhead cost plus a cost that depends on the
sample size, n, and the cost per sample, Cj •
The field maintenance cost affected by the audit can be represented as:
field cost equals defects per unit produced times NC t , where N is the number
of units produced during the quality rating period and Cf is the field cost
per defect sent to the field.
16 The complete problem is to minimize the sum of audit costs and field costs over all
scoring classes, subject to a budget constraint on the total auditing activity.
17 The economic models used for sampling and setting standards are separable; audit costs
are assumed to be independent of standards and manufacturing costs (mentioned in
Establishing Standards) independent of sampling.

752

Environment and Evolution

Part 5

The probability of detecting poor quality will depend on the statistical
procedures used for estimating the quality level and on the threshold
selected for deciding that the quality is unsatisfactory. To a good approximation,18 this detection probability will be proportional to the extent of
the poor quality (measured by 9 p ' the multiple of the defects per unit
produced at standard quality), the sample size (n), and the expected defect
level per unit as inspected in the audit (Us), or
Pr[detecting poor quality Itrue quality is poor] = Kdn9pUs'
where Kd is the constant of proportionality.
Starting with these considerations, a model for total cost can be
derived, which is then minimized with respect to n to yield the optimum
sample size formula: 19

expressed in terms of the sample expectancy, e. The sample expectancy e,
or nUs , is the total number of defects expected in the sample if production is operating at standard. The risk factor or cost ratio, r = CfiCa'
where Ca = CJU s can be interpreted as the (incremental) cost to sample
and inspect enough units to get one expected defect if production were
operating at standard. P is the probability that the true quality is poor,
and the poor quality is detected.
Current values of the constants relating to budget restrictions and
detection threshold, Kb and Kdl have been determined empirically to be 5
and 1/10, respectively. Figure 18-15 shows the universal sampling curve
using these values.

Rating and Reporting. The rating of quality consists of estimating the
true level of quality, ~ based on results of the quality assurance audit
samples, and then determining whether the quality level is satisfactory,
based on a statistical comparison of the estimate with the standard or
expected level of quality for each scoring class. The Quality Measurement Plan (QMP)2o provides a uniform and consistent set of procedures
for the rating and reporting of quality.
18 More complicated models of audit detection power do not significantly affect the optimum
sample size result. With the approximations used, the sample size result is within a few
percent of the more accurate value, over the range of interest.
19 The square root dependence on production (N) has been used in quality assurance
sampling since at least 1930.
20 See Hoadley 1981a.

10
7.0
5.0
Q)

>

()

Z

C

~

2.0

()

w
Q.

e -

)(

w

J

2PrNUS*

1.0
0.7
0.5

100
0.7

7.0

70

Figure 18-15. Universal sampling curve.

It is convenient to describe the (relative) observed level of quality in a
particular rating period, t, by the sample index, It, defined as the ratio of
the observed quality level in period t to the sample expectancy (e, as in
Sampling above). An index of unity means that the observed quality in
that period is at standard, and an index greater than one means that the
observed level is worse than standard. If T denotes the current rating
period, then IT could be interpreted as a simple estimate 21 of the true
quality E>T with a sampling variance (J2.
Each production process is generating a random time series of process
indices let}. If the random process has a fixed (but unknown) process
average index E> and process variance -y2, at least over a reasonably short
interval of time, then one possible estimate of the process average index,
taking into account the information contained in the audit results of the
previous rating periods, is a weighted long-run average

where the weights are inversely proportional to the individual period
variances.
21 This is, in fact, the statistic used in the older t-rate method of rating quality. (See Dodge
1928, pp. 350-368 and Dodge and Torrey 1956, pp. 5-12.)

753

Environment and Evolution

754

Part 5

For rating purposes, information on the long-run average and the
current sample index are combined to form an estimate of current
quality:22
~

0T

==

-

we + (l-W)IT'

where

sampling variance
(sampling variance) + (process variance) .
When the sampling variance is large or when the process variance is
small (the process is relatively stable), W will be close to unity, and most
of the weight will be on the long-run average, as should be expected.
Conversely, when the process variance is large, W will be small, and most
of the weight will be on the current sample index.
In practice, W must be estimated from the data. In addition, an interval estimate of the current population index is needed for rating purposes
to compare the best measure (~T) against the quality standard.
A Gamma distribution is used to approximate the current produced
quality (see Hoadley 1981a) as shown in Figure 18-16. This distribution

QUALITY
STANDARD

99 95

5

PERCENTILES

I I
BOX AND WHISKER PLOT

QMP POSTERIOR
DISTRIBUTION OF
PRODUCED QUALITY

o

2

3

4

QUALITY INDEX

Figure 18-16. Estimated distribution of produced quality
and corresponding box and whisker plot.

22 This result is derived from a Bayesian model as in Hoadley 1981a.

Evolution of Products
and Services

Chap. 18

755

can be represented by a "box and whisker" plot, where the box extends
from the 5th to the 95th percentiles, and the whiskers extend to the 1st
and 99th percentiles. This plot forms the basis for quality assurance
reporting, as shown in Figure 18-17. When the value of the 99th percentile exceeds unity, the product is declared below normal, since a posteriori, there is at least a 99-percent chance that aT is larger than one,
that is, that the level of quality of the production in the current rating
period exceeds (is worse than) the standard. Similarly, the rating class is
declared to be on alert when the 95th (but not the 99th) percentile
exceeds unity.
Nonconformance
The quality assurance function is distinct from quality control and usually
cannot serve as a screen of defective products on a lot-by-Iot basis. However, when strong statistical evidence indicates that the quality of a product has fallen substantially below standard, the product is identified as
nonconforming. In this situation, the Bell Laboratories Quality Assurance
Center determines whether the affected product can be shipped to the
Bell operating companies or must be reworked or discarded. The role of
the quality control process is adequate protection for the customer, and
the quality assurance procedures are designed to monitor this control
process.

~BETTER

o

STD

WORSE~

2

PROCESS AVERAGE,

"8

BEST MEASURE OF QUALITY,
NORMAL
CURRENT PERIOD INDEX,I
T

ALERT

5

4

3

x

BELOW
NORMAL

QUALITY INDEX

Figure 18-17. Quality reporting criteria.

fi'T

756

Environment and Evolution

Part 5

Evidence for a nonconformance usually originates from the audit;
nonconformance procedures are thus subsidiary to and a by-product of
the audit procedures. Such procedures are useful primarily where there
is a gross quality problem that the producer is unaware of, has
misevaluated, or has chosen to ignore. It provides only marginal protection for the customer on a lot-by-Iot basis, since all lots may not be sampled in the quality assurance audit, the audit subsample chosen from a
selected lot may be small, and the trigger threshold used in declaring the
nonconformance is usually severe in order to control the false-alarm
probability.
Trigger thresholds, called allowance numbers, are specified for the various product lines, frequently for major defect types and minor defect
types separately, and sometimes individually for various categories of
defects.
When a nonconforming condition is identified, an attempt is usually
made to determine the extent of the problem, so that large quantities of a
defective product do not appear in the field.
It should be emphasized that the primary objective of the nonconformance procedure is to stimulate corrective action in the material inspection process, the manufacturing or assembly process, and the quality control process. A nonconformance is in effect until the cause of the defects
is identified and corrected, with independent verification by the
appropriate quality assurance organization.
lS.7.3 QUALITY ASSURANCE MONITORING
Quality assurance auditing alone is not always sufficient to ensure
adequate quality of the final product; hence, the audit is supplemented by
various quality-monitoring activities. As shown in Figure 18-12, monitoring occurs throughout all stages of the development, manufacture, operating, and repair processes to verify that adequate controls have been maintained. In addition, numerous tracking studies are conducted in the Bell
operating companies to determine the performance of the products quantitatively under actual field conditions. The quality assurance activities
that continue throughout the life of a product are described below.
Surveillance
Continuing surveillance usually occurs when available information indicates that chronic quality problems exist in some phase of the production
processes or when a formal audit of the end product is either not
appropriate or is, by itself, deemed not to provide an adequate measure of
quality. Examples might be components or piece parts manufactured at
one location and subject to only a partial inspection after assembly into a
finished product at another location.

Chap. 18

Evolution of Products
and Services

757

Surveillance ordinarily includes:
• reviewing engineering and manufacturing information with regard to
com pleteness and clarity
• verifying that all engineering requirements are checked or otherwise
accounted for in a satisfactory manner, including process requirements
and generally accepted standards of good workmanship
• verifying that test sets are calibrated, maintained, and operated in
accordance with specifications
• reviewing the quality control program and its effectiveness
• reviewing the adequacy of manufacturing facilities (machines, tools,
ovens, test sets, gauges, etc.) and their maintenance
• reviewing packaging and shipping operations.
Surveys
Quality assurance surveys comprise I-time, objective, in-depth reviews of
factors affecting initial quality, performance, and service life of products
or services provided to the Bell operating companies. They are most
often applied to areas where severe quality problems are known to exist
or may be anticipated, including situations where extensive field problems have been reported or recurrent poor quality levels are evident (for
example, from the audit). Quality assurance surveys may also be conducted on newly introduced products or when manufacture of a product
is transferred to a different location. The intent is to determine if all
necessary steps have been taken to obtain and maintain satisfactory quality and to identify specific conditions that may affect the adequacy of
quality. Surveys are usually the joint effort of Western Electric and Bell
Laboratories personnel directly involved with the affected product lines.

Field Representatives
To expedite the flow of information between Bell Laboratories and the
Bell operating companies, the Bell Laboratories Quality Assurance Center
maintains a staff of eighteen field representatives and fifteen assistants at
the company locations shown in Figure 18-18. They constantly monitor
the operating side of the business and participate in field evaluations and
reliability studies. Their activities take them into all parts of the telephone plant, bring them into close contact with personnel at all levels,
and expose them to service problems, operating company needs, and
equipment performance.
One of their major functions is alerting Bell Laboratories and Western
Electric to quality problems in the field. For situations that require

OMAHA.

•

DENVER

INDIANAPOLIS •
ST. LOUIS.

Figure 18-18. Locations of Bell Laboratories field representatives.

extended investigations and possibly coordination of several organizations, Bell operating companies submit engineering complaints (discussed
below) that are reviewed and tracked by field representatives. In addition, field representatives are available for technical consultation concerning the engineering, operation, and maintenance of Bell System communications equipment. In their interface role, they represent the Bell
Laboratories viewpoint to the Bell operating companies, and equally as
important, they represent the Bell operating company viewpoint to Bell
Laboratories.
Engineering Complaints
This mechanism is a formal feedback path between the user and the
designer or manufacturer. It serves as an additional source of information concerning field problems and supports the standard supply contract
between the Bell operating companies and Western Electric (see Section 1.2.3).
Currently, about eight thousand complaints are submitted annually;
about 70 percent are handled by Western Electric and the rest by Bell
Laboratories. Complaints sent to Bell Laboratories are reviewed and distributed to the responsible development organizations by the Quality
Assurance Center. Bell Laboratories designers report their findings and
corrective action in a final report, which is reviewed by the Quality
Assurance Center and the appropriate field representative before distribution to the operating company. As part of the complaint operation, the
758

Evolution of Products
and Services

Chap. 18

759

Quality Assurance Center issues, through the field representatives, summary reports generated from the computerized engineering complaint
data base. These documents alert the Bell operating companies to design
problems encountered throughout the system, thus giving each the
benefit of the experience of the others.

Field Performance Tracking
While the engineering complaint routine provides a great deal of useful
information about field problems, additional quantitative information is
often required, particularly for high-volume products such as circuit
packs and customer-premises equipment. For these products, detailed
feedback is obtained through special field studies (frequently conducted
with the assistance of field representatives) and by means of formal Product Performance Surveys (PPSs), which continuously monitor the quality of
critical products.
There may be several dozen tracking studies being conducted concurrently under the aegis of the Quality Assurance Center. These special
studies are frequently implemented, with the cooperation of one or more
Bell operating companies, when important new products are first introduced to the field. Quantitative data are gathered for a long enough
period of time, under actual operating conditions, either to confirm that
the quality level is satisfactory or to identify specific performance
deficiencies. Such studies are especially critical in the presence of
decreasing intervals between design concept and sale to the Bell operating companies. In addition, the field studies may identify the need for
changes in the audit.
Product Performance Surveys provide continuous monitoring to supplement the normal laboratory testing, field trials, and appraisal studies
associated with the introduction of new or modified equipment. The
basic functions of PPSs are shown in Figure 18-19. Unforeseen quality

Figure 18-19. Product Performance
Survey functions.

760

Environment and Evolution

Part 5

problems can appear as a result of well-intentioned cost reductions,
design changes, process modifications, and material substitutions. Quality
control and quality assurance at the manufacturing or repair stage can
prevent quality problems only if the consequences of these changes are
anticipated and if measurement techniques are feasible and economically
justified.
By providing extensive information on in-service reliability and initially defective units, the PPS quantifies product quality as seen by the
Bell operating companies. The continuous tracking of about one million
telephone sets in service and hundreds of thousands of installations, for
example, has resulted in the collection of extensive detailed defect data.
Analyses of these data have provided the means for quickly and accurately identifying problems. Since the PPS is a continuing activity, the
modified product can be monitored to ensure the validity of the "fixes"
implemented, and the cycle repeated as shown in Figure 18-19.
In a sense, Figure 18-19 summarizes the nature of all the major quality
assurance functions included under auditing and monitoring: They are
on-going activities intended to identify quality problems, call them to the
attention of the appropriate organizations, and then verify that the
corrective actions have been effective.

AUTHORS
K. J. Cohen
B. L. Hanson

C. E. Johnson
C. H. King
D. C. Krupka
V. O. Mowery
R. Sherman

References and Additional Reading

ASQC
BLR
BSTJ
CCITT
EIA
FCC
ICC
ICCC
IEEE
ISO
USITA

American Society for Quality Control
Bell Laboratories Record
The Bell System Technical Journal
Comite Consultatif International
Telegraphique et Telephonique
Electronics Industries Association
Federal Communications Commission
International Conference on Communications
International Conference on Computer
Communications
Institute of Electrical and Electronics
Engineers
International Organization for Standardization
United States Independent Telephone
Association

CHAPTER 1. STRUCTURE AND ACTIVITIES

References
AT&T. 1983. 1982 Statistical Report. New York.
Bell Laboratories. 1982.
Hills, NJ.

Facts About Bell Laboratories.

12th ed. Short

- - - . 1977. Engineering and Operations in the Bell System. Murray Hill,
NJ.
761

References
Additional Reading

762

Fagen, M. D., ed. 1975/1978. A History of Engineering and Science in the
Bell System: vol. 1, The Early Years (1875-1925); vol. 2, National Service in
War and Peace (1925-1975). Murray Hill, NJ: Bell Laboratories.
Lustig, 1. K., ed. 1981. Impact: A Compilation of Bell System Innovations in
Science and Engineering. 2nd. ed. rev. Short Hills, NJ: Bell Laboratories.
Mueser, R., ed. 1979. Bell Laboratories Innovation in Telecommunications.
Murray Hill, NJ: Bell Laboratories. (A second edition, updated, is
scheduled for publication in 1984.)
USITA. 1983. Independent Telephone Statistics for the Year 1982. Vol. 1,
1982 ed. Washington, DC.

CHAPTER 2. SERVICES

Reference
AT&T Long Lines. 1982. Basic Packet Switching Service. Illustrative Tariff
FCC No. 270. November 15.

CHAPTER 3. INTRODUCTION TO THE NETWORK

Reference
Skoog, R. A., ed. 1980. The Design and Cost Characteristics of Telecommunications Networks. Murray Hill, NJ: Bell Laboratories.

Additional Reading
Manhire, 1. M. 1978. Physical and Transmission Characteristics of Customer Loop Plant. BSTJ 57:35-59.
Noweck, H. E. 1961.
39:312-316.

The Versatility of TOUCH-TONE Calling.

BLR

763

Chapters 4 and 5

CHAPTER 4. NETWORK STRUCTURES AND PLANNING

References
AT&T Long Lines. 1982. The World's Telephones-A Statistical Compilation
as of January, 1981. Morris Plains, NJ.
Skoog, R. A., ed. 1980. The Design and Cost Characteristics of Telecommunications Networks. Murray Hill, NJ: Bell Laboratories.

Additional Reading
Bell Laboratories. 1978. BSTJ, vol. 57, no. 4 (April). A special issue on
the loop plant.
Katz, S. 5.; Lifchus,1. M.; and
Switched Service. BLR 57:38-45.

Skeer, M. H. 1979.

A

Sophisticated

Lutz, K. J.; Pecsvaradi, T.; and Waninski, J. E. 1983. The Integrated Special Services Network. Paper read at the IEEE ICC '83. Boston.

CHAPTER 5. TRAFFIC

References
Barnes, D. H. 1976. Extreme Value Engineering of Small Switching
Offices. Proceedings of the 8th International Teletraffic Congress, Melbourne,
10-17 November. Paper no. 242.
Cooper, R. B. 1981. Introduction to Queueing Theory. 2nd ed. New York:
North Holland Press.
Elsner, W. B. 1977. A Descent Algorithm for the Multihour Sizing of
Traffic Networks. BSTJ 56:1405-1430.
Feller, W. 1966. An Introduction to Probability Theory and Its Applications.
Vol. 2, 2nd ed. New York: John Wiley & Sons.
- - - . 1968. An Introduction to Probability Theory and Its Applications.
Vol. I, 3rd ed. New York: John Wiley & Sons.

764

Keterences
Additional Reading

Fuchs, E., and Jackson, P. E. 1970. Estimates of Distributions of Random
Variables for Certain Computer Communications Traffic Models. Communications of the Association for Computing Machinery 13:752-757.
Hill, D. W., and Neal, S. R. 1976. Traffic Capacity of a ProbabilityEngineered Trunk Group. BSTJ 55:831-842.
Wilkinson, R. I. 1956. Theories for Toll Traffic Engineering in the USA.
BSTJ 35:421-514.

CHAPTER 6. TRANSMISSION

References
AT&T. 1977. Telecommunications Transmission Engineering. 2nd ed. 3 vols.
Lisle, IL: Bell System Center for Technical Education.
Bell Laboratories. 1982. Transmission Systems for Communications. 5th ed.
Holmdel, NJ.
Greene, E. S. 1962. Principles of Physics. New York: Prentice-Hall.
Lucky, R. W.; Salz, J.; and Weldon, E. J., Jr. 1968. Principles of Data Communication. New York: McGraw-Hill Book Company.
Martin, J. 1976. Telecommunications and the Computer. 2nd ed. Englewood
Cliffs, NJ: Prentice-Hall.

Additional Reading
Abate, J. E.; Rosenberger, J. R.; and Yin, M. 1981. Keeping the Integrated
Services Digital Network in Sync. BLR 59:217-220.
AT&T. 1980. Notes on the Network. Section 7.
Campbell, L. W., Jr. 1970. The PAR Meter: Characteristics of a New
Voiceband Rating System. IEEE Transactions on Communications Technology
COM-18:147-153.
Cavanaugh, J. R.; Hatch, R. W.; and Sullivan, J. L. 1976. Models for the
Subjective Effects of Loss, Noise, and Talker Echo on Telephone Connections. BSTJ 55:1319-1371.

765

Chapters 6, 7, and 8

Jacobs, I. 1979.
BLR 57:298-304.

Lightwave Communications Begins Regular Service.

Lewinski, D. A. 1964. A New Objective for Message Circuit Noise. BSTJ
43:719-740.

CHAPTER 7. SWITCHING

Additional Reading
Inose, H. 1979. Introduction to Digital Integrated Communications Systems.
Tokyo: University of Tokyo.
Joel, A. E., Jr. 1977. What Is Telecommunication Switching? Proceedings
of the IEEE 65:1237-1253.
- - - . 1979. Digital Switching-How It Has Developed. IEEE Transactions on Communications COM-27:948-959.
Marcus, M. J. 1977. The Theory of Connecting Networks and Their
Complexity: A Review. Proceedings of the IEEE 65:1263-1271.
McDonald, J. C., ed. 1983. Fundamentals
York/London: Plenum Press.

of

Digitai

Switching.

New

Pearce, J. G. 1981. Telecommunication Switching. New York/London: Plenum Press.
Talley, D. 1975. Basic Electronic Switching for Telephone Systems. Rochelle
Park, NJ: Hayden.
- - . 1979. Basic Telephone Switching Systems. 2nd ed. Rochelle Park,
NJ: Hayden.

CHAPTER 8. SIGNALING AND INTERFACES

References
AT&T. 1978. Interconnection Specification for Digital Cross-Connects. AT&T
Technical Reference no. 34. Issue 2. Basking Ridge, NJ.

766

Kererences
Additional Reading

-,- - . 1980a. Local Switching System General Requirements (LSSGR). Preliminary issue. Bell System Technical Reference PUB 48501.
- - - . 1980b. Operations Systems Network Communications Protocol
Specification BX.25. Bell System Technical Reference PUB 54001. Issue 2.
- - - . 1983. Description of the Analog Voiceband Interface Between the Bell
System Local Exchange Lines and Terminal Equipment. Bell System Technical
Reference PUB 61100.
CCITT. 1981a. Data Transmission at 48 Kilobits Per Second Using 60-108 kHz
Group Band Circuits. Recommendation V.35. Yellow Book. Vol. 8.1. Geneva.
- - - . 1981b. Interface Between Data Terminal Equipment (DTE) and Data
Circuit-Terminating Equipment (DCE) for Terminal Operating in the Packet
Mode on Public Data Networks. Recommendation X.25. Yellow Book.
Vol. 8.2. Geneva.
- - - . 1981c. Terminal and Transmit Call Control Procedures and Data
Transfer System on International Circuits Between Packet-Switched Data Networks. Recommendation X.75. Yellow Book. Vol. 8.3. Geneva.
- - - . 1983. Reference Model of Open Systems Interconnection for CCITT
Applications, Draft Recommendation X.200. CCITT Circular Letter no. 58.
EIA. 1969. Interface Between Data Terminal Equipment and Data Communication Equipment Employing Serial Binary Data Exchange. Standard RS-232-C.
- - - . 1977/1980. General Purpose 37-Position and 9-Position Interface for
Data Terminal Equipment and Data Circuit-Terminating Equipment Employing
Serial Binary Data Interchange. Standard RS-449. With Addendum 1.
FCC. 1975. Engineering and Operations Supplement, FCC Docket 20099.
- - . 1977/1980. Part 68 Rules and Regulations. Vol. 10.
ISO. 1983. Data processing - Open systems interconnection model. Organization Standard 7498.

Basic reference

Additional Reading
AT&T. 1977. Telecommunications Transmission Engineering. 2nd ed. 3 vols.
Lisle,IL: Bell System Center for Technical Education. Vol. 2, Section 1.

767

Chapters 8 and 9

Bell Laboratories. 1978. BSTJ, vol. 57, no. 2 (February). A special issue
on common-channel interoffice signaling.
- - - . 1982. BSTJ, vol. 61, no. 7, part 3 (September). A special issue on
the stored-program control network.
CCITT. 1981. Common Channel Signaling System No.7. Recommendations
Q.701-Q.707, Q.721-Q.725. Yellow Book. Vol 6.6. Geneva.

CHAPTER 9. TRANSMISSION SYSTEMS

Reference
Bell Laboratories. 1982.
5th ed. Holmdel, NJ.

Transmission

Systems

for

Telecommunications,

Additional Reading
Adams, W. B., and Bailey, C. C. 1976. Identifying the Needs of Rural
Networks. BLR 54:184-188.
Anderson, C. D.; Gleason, R. F.; Hutchinson, P. T.; and Runge, P. K.
1980. An Undersea Communication System Using Fiberguide Cables.
Proceedings of the IEEE 68:1299-1303.

AT&T. 1977. Telecommunications Transmission Engineering. 2nd ed. 3 vols.
Lisle, IL: Bell System Center for Technical Education.
Bangert, J. T. 1973. L5-A "Jumbojet" Coaxial System. BLR 51:290-299.
Berger, U. S. 1973.· The Old TD2 Becomes the New TD2C. BLR 51:278284.
Bergholm, J. 0., and Koliss, P. P. 1972. Serving Area Concept-A Plan
for Now with a Look to the Future. BLR 50:212-216.
Billhardt, R. A.; Guarneri, P. J.; Madigan, T. W.; and Tentarelli, K. D.
1980. The Digital Carrier Trunk: a "Smart" Move Toward Increased ESS
Capabilities. BLR 58:341-345.

768

References
Additional Reading

Bleisch, G. W., and Michaud, W. P., Jr. 1971. The A-6 Channel Bank:
Putting New Technologies to Work. BLR 49:251-254.
Brolin, S.; Cho, Y.-S.; Michaud, W. P., Jr.; and Williamson, D. H. 1980.
Inside the New Digital Subscriber Loop System. BLR 58:110-116.
Clark, M., and Porter, G. R. 1976. The Metallic Facility Terminal: Special Help for Special Services. BLR 54:215-219.
Colton, J. R. 1980. Cross-connections-DACS Makes Them Digital. BLR
58:248-255.
Danielson, W. E. 1975.

Exchange Area and Local Loop Transmission.

BLR 53:40-49.

Dodds, S., and Mitchell, W. J. 1979. From Analog to Digital for the No.4
ESS-Economically. BLR 57:288-292.
Drechsler, R. C. 1980. DACS Cross-Connects-And That's Just the Beginning. BLR 58:305-311.
Graczyk, J. F.; Mackey, E. T.; and Maybach, W. J. 1975. TIC Carrier: The
Tl Doubler. BLR 53:256-263.
Haury, P. T., and Romeiser, M. B. 1976. Tl Goes Rural. BLR 54:178-183.
Hull, T. R., and LeBlanc, R. E. 1979. New Life for Short-Haul Analog
Routes. BLR 57:270-274.
Hymel, D. P., and Walker, J. D. 1980. The Difference with DIF? Size,
Intelligence-and Cost. BLR 58:346-349.
Jacobs, I. 1980. Lightwave Communications-Yesterday, Today, and
Tomorrow. BLR 58:2-10.
Johannes, V. I. 1976. The Evolving Digital Network. BLR 54:268-273.
Markle, R. E. 1978. Single Sideband Triples Microwave Radio Route
Capacity. BLR 56:104-110.
O'Neill, E. F. 1975. Radio and Long-Haul Transmission. BLR 53:50-59.
Sipress, J. M. 1975. T4M: New Superhighway for Metropolitan Communications. BLR 53:352-359.

Chapter 10

769

CHAPTER 10. NETWORK SWITCHING SYSTEMS

Additional Reading
Almquist, R. P.; Carney, D. L.; and Estvander, R. A. 1977. 1A ESS:
Newest, Largest-Capacity Local Switch Cuts Over Early. BLR 55: 15-20
Bell Laboratories. 1964. BSTJ, vol. 43, no. 5 (September). A special issue
on the lESS system.
1965. BLR, vol. 43, no. 6 (June). A special issue on the lESS

system.
1969. BSTJ, vol. 48, no. 6 (October). A special issue on the 2ESS

system.
1970. BSTJ, vol. 49, no. 10 (December). A special issue on TSPS

No. 1.
1974. BSTJ, vol. 53, no. 1 (January).
Automatic Intercept System.

A special issue on the

- - - . 1977a. BSTJ, vol. 56, no. 2 (February). A special issue on the 1A
Processor.
- - - . 1977b. BSTJ, vol. 56, no. 7 (September). A special issue on the
4ESS system.

- - - . 1977c. BLR, vol. 55, no. 11 (December). A special issue on the
4ESS system.

- - . 1979. BSTJ, vol. 58, no. 6, part 1 (July-August). A special issue
on TSPS No. 1.
- - . 1981. BSTJ, vol. 60, no. 6, part 2 (July-August). A special issue
on the 4ESS system.
- - . 1982. BSTJ, vol. 61, no. 4 (April). A special issue on the lOA
Remote Switching System.
- - - . 1983a. BSTJ, vol. 62, no. 1, part 2 (January). A special issue on
the 3B20D Computer and DMERT Operating System.
- - . 1983b. BSTJ, vol. 62, no. 3, part 3 (March). A special issue on
TSPS No. lB.

~~r~r~IU;~t;

770

Additional Reading

Bruce, R. A.; Giloth, P. K.; and Siegel, E. H., Jr. 1979. No.4 ESS: A Continuing Evolution. BLR 57:154-161.
Byrne, C. J., and Pilkinton, D. C. 1976. Towards Automated Local Billing. BLR 54:104-109.
Fagan, M. D., ed. 1975. A History of Engineering and Science in the Bell System: vol. 1, The Early Years (1875-1925). Murray Hill, NJ: Bell Laboratories. Chapter 6, pp. 467-714.
Foster, R. W. 1977. No.3 ESS Improves Telephone Service for Country
Customers. BLR 55:230-235.
Garraty, W. C. 1975.
53:395-399.

Introducing CAMA Features to No.1 ESS.

BLR

International Switching Symposium. 1981. No.5 ESS-Overview, System Architecture, Hardware Design, Software Design. Montreal, 21-25
September. Vol. 3, pp. 31A-l,I-31A-4,6.
Joel, A. E., Jr., et al. 1982. A History of Engineering and Science in the Bell
System: Switching Technology (1925-1975). Murray Hill, NJ: Bell Laboratories.
Johnson, J. W.; Kennedy, J. C.; and Warner, J. C. 1981. No. 5 ESSServing the Present, Serving the Future. BLR 59:290-293.
Kleber, J. J., and Perkinson, W. B. 1980. No. lA AMARC: At the Center
of Billing Data Collection. BLR 58:236-243.
Lehder, W. E., Jr., and Whitemyer, J. G. 1979. Introduction of AMA to
No.3 ESS. BLR 57:174-178.
Mandigo, P. D. 1976. No.2B ESS: New Features from a More Efficient
Processor. BLR 54:304-309.
Neville, S. M., and Royer, R. D.
Switching Systems. BLR 54:30-33.

1976.

Controlling Large Electronic

Richards, P. C. and Herndon, J. A. 1973. No.2 ESS: An Electronic
Switching System for the Suburban Community. BLR 51:130-135.
Sevcik, R. W., and Smith, D. P. 1980.
Clarksville. BLR 58:63-68.

Custom

Calling

Comes

to

Chapters 10 and 11

771

Telesis. 1978. Vol. 5, no. 10 (August). A special issue on the DMS-10
digital community office.
Winckelmann, W. A. 1968. Automatic Intercept Service. BLR 46:138-143.

CHAPTER 11. CUSTOMER-SERVICES EQUIPMENT AND SYSTEMS

References
Balkovic, M. D.i Klancer, H. W.i Klare, S. W.i and McGruther, W. G. 1971.
1969-70 Connection Survey: High-Speed Voiceband Data Transmission
Performance on the Switched Telecommunications Network. BST]
50:1349-1384.
Kretzmer, E. R. 1973.
51:258-265.

The New Look in Data Communications.

BLR

Larsen, A. B., and Brown, E. F. 1980. Continuous Presence Video Conferencing at 1.5/6 Mbps. Teleconferencing and Interactive Media, pp.391398. [Madison]:University of Wisconsin.
Lucky, R. W.i Salz, J.i and Weldon, E. J., Jr., 1968. Principles of Data Communication. New York: McGraw-Hill Book Company.
Netravali, A. N., and Robbins, J. D. 1979. Motion-Compensated Television Coding: Part I. BST] 58:631-670.

Additional Reading
Arnold, T. F., and Toy, W. N. 1981. Inside the 3B-20 Processor.
59:66-71.

BLR

Bell Laboratories. 1979. BST], vol. 58, no. 1 (January). A special issue on
Advanced Mobile Phone Service.
Borison, V. S. 1975. Transaction Telephone Gets the Facts at the Point of
Sale. BLR 53:376-383.
Brophy, F. J.i Honnold, G. H.i and Thayer, S. T. 1981. DATAPHONE II
Service-New Standard for Data Communications. BLR 59:248-252.

772

References
Additional Reading

Burns, H. S., and Loosme, O. 1981. An Inside Look at DATAPHONE II
Service-Part II: Diagnostic Tools. BLR 59:268-271.
Busala, A. 1960. Fundamental Considerations in the Design of a VoiceSwitched Speakerphone. BSTJ 39:265-294.
Carpenter, J. R.; Dennis, T. M.; Holzman, L. N.; and Tong, K. 1981. An
Inside Look at DATAPHONE II Service-Part I: Data Sets. BLR 59:264268.
Daugherty, T. H. 1970. Digitally Companded Delta Modulation for Voice
Transmission. IEEE International Symposium on Circuit Theory, Digest of
Technical Papers, pp. 17-18.
Douglas, V. A. 1964. The MJ Mobile Radio Telephone System. BLR
42:382-389.
Ehlinger, J. C., and Stubblefield, R. W. 1983. No.1 PSS: Number 1
Packet-Switching System-Service Capabilities and Architecture. Paper
read at the IEEE ICC '83. Boston.
Ham, J. H., and West, F. 1963. A TOUCH-TONE Caller for Station Sets.
IEEE Transactions on Communication and Electronics, no. 65:17-24.
Handler, G. J., and Snowden, R. L. 1982. Planning Packet Transport
Capabilities in the Bell System. Paper read at the 6th ICCC. London.
Huff, D. L., and Pennotti, R. J. 1980.
Ahead. BLR 58:91-96.

Mobile Phone Service Moves

Inglis, A. H., and Tuffnell, W. L. 1951. An Improved Telephone Set.
BSTJ 30:239-270.

Levy, R. P., and Magee, F. R. 1983. An Internal Packet Network Protocol
and Buffer Management Scheme for an X.25 Based Network. Paper read
at IEEE INFOCOM '83. San Diego.
Mansell, J. J., and Stubbs, R. D., II. 1982. Systems Engineering Considerations in the Bell System Packet-Switching Networks. Paper read at
the 6th ICCC. London.
Matlack, R. C., and Render, D. J. 1977. DATAPHONE Select-a-Station
Service: Improved Transmission Facilities for Alarm Systems. BLR
55:243-247.

Chapters 11, 12, 13, and 15

773

Mounts, F. W. 1969. A Video Encoding System With Conditional
Picture-Element Replenishment. BSTJ 48:2545-2554.
O'Brien, J. A. 1978. Final Tests Begin for Mobile Telephone System. BLR
56:171-174.
Rodriguez, E. J. 1982. Architectural Considerations in Bell System
Packet-Switching Networks. Paper read at the 6th ICCC. London.

CHAPTER 12. COMMON SYSTEMS

Additional Reading

New Equipment-Building System (NEBS)-General Equipment
Requirements. Bell System Technical Reference PUB 51001. Issue 2, rev.

AT&T. 1981.

Pferd, W. 1979. The Evolution and Special Features of Bell System Telephone Equipment Buildings. BSTJ 58:427-466.

CHAPTER 13. OVERVIEW OF TELEPHONE COMPANY
OPERATIONS

Reference
Martz, L. M., and Osofsky, A. J. 1982. Operations Impact of New Technology: A Case Study of the Digital Access and Cross-Connect System.
IEEE ICC '82, Conference Record. Philadelphia.

CHAPTER 15. OPERATIONS PLANNING

Additional Reading
Johnson, D. W.; Varma, G. K.; and Waninski, J. E. 1982. Network Operations Planning for the Bell System. IEEE ICC '82, Conference Record.
Philadel phia.

774

References
Additional Reading

CHAPTER 16. EVALUATION OF SERVICE AND PERFORMANCE

References
Cavanaugh, J. R.; Hatch, R. W.; and Sullivan, J. L. 1976. Models for the
Subjective Effects of Loss, Noise, and Talker Echo on Telephone Connections. BSTJ 55:1319-1371.
Hester, S. 1982. Taking the Pulse of the Network. BLR 60:70-74.
McDade, S. 1979. Measurement for Measurement's Sake, Past Tense. Bell
Telephone Magazine 58 (February):31-32.
Militzer, K. H. 1980. Macro versus Micro Input/Output Ratios. Management Review 69 (June):8-15.

Additional Reading
Alexander, A. A.; Gryb, R. M.; and Nast, D. W. 1960. Capabilities of the
Telephone Network for Data Transmission. BSTJ 39:431-476.
Balkovic, M. D.; Klancer, H. W.; Klare, S. W.; and McGruther, W. G. 1971.
1969-70 Connection Survey: High-Speed Voiceband Data Transmission
Performance on the Switched Telecommunications Network. BSTJ
50:1349-1384.
Duffy, F. P., and Thatcher, T. W., Jr. 1971. 1969-70 Connection Survey:
Analog Transmission Performance on the Switched Telecommunications
Network. BSTJ 50:1311-1347.
Duffy, F. P.; McNees, G. K.; Nasell, I; and Thatcher, T. W., Jr. 1975. Echo
Performance of Toll Telephone Connections in the United States. BSTJ
54:209-243.
Fennick, J. H., and Nasell, I. 1966. The 1963 Survey of Impulse Noise on
Bell System Carrier Facilities. IEEE Transactions on Communications Technology COM-14:520-525.
Fleming, H. C., and Hutchinson, R. M., Jr. 1971. 1969-70 Connection
Survey: Low-Speed Data Transmission Performance on the Switched
Telecommunications Network. BSTJ 50:1385-1405.

Chapters 16 and 18

775

Gresh, P. A. 1969. Physical and Transmission Characteristics of Customer Loop Plant. BSTJ 48:3337-3386.
Kessler, J. E. 1971. The Transmission Performance of Bell System Toll
Connecting Trunks. BSTJ 50:2741-2777.
N~sell,

I. 1964. The 1962 Survey of Noise and Loss on Toll Connections.
BSTJ 43:697-718.

- - - . 1968. Some Transmission Characteristics of Bell System Toll
Connections. BSTJ 47:1001-1018.
N~sell, I.; Ellison, C. R., Jr.; and Holmstrom, R. 1968. The Transmission
Performance of Bell System Intertoll Trunks. BSTJ 47:1561-1613.

Spang, T. C. 1976. Loss-Noise-Echo Study of the Direct Distance Dialing
Network. BSTJ 55:1-36.

CHAPTER 18. EVOLUTION OF PRODUCTS AND SERVICES

References
AT&T. 1983. Purchased Products Quality Handbook. Bell System Information Publication IP 10450.
Belfi, J. 1.; Glenn, F. W.; and Keyser, C. J. 1978. Quality Assurance for
Components. BLR 56:296-302.
Bell Labs News. March 1, 1982.

Comella, W. K. 1978. Quality Assurance for ESS: Advanced Measures
for Advanced Technology. BLR 56:288-295.
Dodge,

H. F. 1928.

A

Method

of Rating

Manufactured

Product.

BSTJ 7:350-368.

Dodge, H. F., and Torrey, M. N. 1956. A Check Inspection and Demerit
Rating Plan. Industrial Quality Control 13:5-12.
Dreyfuss, H. 1959. The Measure of Man; Human Factors in Design.
York: Whitney Library of Design.

New

776

Kererences
Additional Reading

French, W. L., and Godfrey, B. 1978. Quality Assurance: The New Audit
for Cables. BLR 56:276-281.
Fuchs, E., and Howard, B. T. 1978. Quality Assurance: Tradition and
Change. BLR 56:226-231.
Godfrey, B., and Hoadley, B.
Assurance. BLR 56:233-237.

1978.

Statistical Methods in Quality

Green, P. E., and Tull, D. S. 1978. Research for Marketing Decisions. 4th ed.
Englewood Cliffs, NJ: Prentice-Hall.
Hiering, V. S., and Hooke, J. A. 1978. Product Performance Surveys:
Field Tracking for Station Sets. BLR 56:238-240.
Hoadley, B. 1981a. The Quality Measurement Plan (QMP). BST] 60:215273.
- - - . 1981b. The Universal Sampling Plan. Transactions of the 35th
Annual Quality Congress, May 27-29.
Kennedy, K. W., and Bates, C. 1965. Development of Design Standards for
Ground Support Consoles. Aerospace Medical Research Laboratories report
no. AMRL-TR-65-163. Wright-Patterson Air Force Base, Ohio.
Mowery, V. O. 1978.
BLR 56:271-275.

A New Look at Customer Premises Products.

Peters, R. A., and Karraker, I. O. 1975. Why You Can Trust Your Telephone. Industrial Research 17:47-51.

Scientific American. 1977. A special issue on microelectronics. Vol. 237
(September).
Shewhart, W. A. 1926. Correction of Data for Errors of Averages
Obtained from Small Samples. BST] 5:308-319.
Urban, G. L., and Hauser, J. R. 1980. Design and Marketing of New Products. Englewood Cliffs, NJ: Prentice-Hall.
Van Cott, H. P., and Kinkade, R. G. 1972. Human Engineering Guide to
Equipment Design. Washington: American Institutes for Research.
Woodson, W. E., and Conover, D. W. 1964. Human Engineering Guide for
Equipment Designers. 2nd ed. Berkeley: University of California Press.

Chapter 18

777

Additional Reading
AT&T. 1977. Engineering Economy: A Manager's Guide to Economic Decision
Making. 3rd ed. New York: McGraw-Hill Book Company.
Dodge, H. F., and Romig, H. G. 1929. A Method of Sampling Inspection.
BSTJ 8:613-631.
- - - . 1959. Sampling Inspection Tables: Single and Double Sampling. 2nd ed.
New York: John Wiley & Sons.
Fagen, M. D., ed. 1975. A History of Engineering and Science in the Bell System: vol. 1, The Early Years (1875-1925). Murray Hill, NJ: Bell Laboratories. Chapter 9.
Hoadley, B. 1979. An Empirical Bayes Approach to Quality Assurance.
33rd Annual Technical Conference Transactions of ASQC, pp. 257-263.
Hughes, G. D. 1978. Marketing Management: A Planning Approach. Reading, MA: Addison-Wesley.
Kolter, P. 1980. Marketing Management: Analysis, Planning and Control.
4th ed. Englewood Cliffs, NJ: Prentice-Hall.
Liebesman, B. S. 1979. The Use of MIL-STD 1050 to Control Average
Outgoing Quality. Journal of Quality Technology 11:36-43.
McCarthy, E. J. 1978. Basic Marketing: A Managerial Approach. 6th ed.
Homewood, IL: Richard D. Irwin.
McCormick, E. J. 1964. Human Factors Engineering. 2nd ed. New York:
McGraw-Hill Book Company.
Murphy, R. B. 1956. The Role of Quality Assurance in the Bell System.
BLR 34:241-245.
Patterson, E. G. D. 1956. An Overall Quality Assurance Plan. Industrial
Quality Control 12:32-37.
Phadke, M. S. 1979. Sequential Empirical Bayes Sampling via Microcomputers. 33rd Annual Technical Conference Transactions of ASQC, pp. 833-841.
Shewhart, W. A. 1927. Quality Control. BSTJ 6:722-735.

778

References
Additional Reading

- - - . 1931. Economic Control of Quality of Manufactured Product. New
York: D. Van Nostrand Co.
- - - . 1958. Nature and Origin of Standards of Quality. BSTJ 37:1-22.
Wagner, H. M. 1975. Principles of Operations Research.
wood Cliffs, NJ: Prentice-Hall.

2nd ed.

Engle-

Glossary

Words set in SMALL CAPS are defined elsewhere in this glossary. A few of the definitions in the glossary have been
taken, with permission, from the sources listed below. A
bracketed letter code that corresponds to the specific
source follows each of these definitions.

DST

McGraw-Hill Dictionary of Scientific and Technical
Terms. 1974. New York: McGraw-Hill.

IEC/ITU

Definitions agreed upon by Working Group C
of the International Electrotechnical Commission (lEC)/lnternational Telecommunications
Union (lTU) Joint Coordination Group. 1982.

IEEE

IEEE Standard Dictionary of Electrical and Electronics Terms. 1977. New York: Institute of
Electrical and Electronics Engineers.

TC

Martin, J. 1976. Telecommunications and the
Computer. 2nd ed. Englewood Cliffs, NJ:
Prentice-Hall, Inc.

Address. (1) A sequence of numbers that identifies the telephone or other
customer-premises EQUIPMENT to which a call is directed. The address
is usually a 7- or 10-digit number, depending on whether the destination is inside or outside the NUMBERING PLAN AREA from which the
call originated. (2) Digital information (a combination of bits) that
identifies a location in a storage device or equipment unit.

779

780

Glossary

Addressing. Specifying to the NETWORK the, destination of a call.
Address Signal. A SIGNAL used to convey call destination information such
as telephone STATION NUMBER, CENTRAL OFFICE CODE, and area code. Some
forms of address signals are called pulses; for example, dial pulses, multifrequency pulses.
Alerting. Generating an audio (or visual) SIGNAL to indicate that a call is
being made.
Alerting Signal. A SIGNAL sent to a customer, a PRIVATE BRANCH EXCHANGE, or
a SWITCHING SYSTEM to indicate an incoming call. A common form is
the signal that rings a bell in the station set being called.
All-Number Calling. The system of telephone numbering that uses only
numbers and replaces the 2-letter plus 5-number (2L+5N) NUMBERING
PLAN. All-number calling offers more usable combinations of numbers
than the 2L+5N numbering plan and is now the nationwide
standard.
Alternate Routing. A means of selectively distributing TRAFFIC over a
number of routes that ultimately lead to the same destination.
Amplitude/Frequency Distortion. Distortion in which the relative magnitudes of the different frequency components of a SIGNAL are altered
during amplification or TRANSMISSION over a CHANNEL. [DS11
Amplitude Modulation (AM). One way to modify a SIGNAL to make it
"carry" information. The amplitude of a CARRIER SIGNAL is modified in
accordance with the amplitude of the information signal.
Analog Channel. A TRANSMISSION path that accepts a band of frequencies
and is compatible with the transmission of ANALOG SIGNALS.
Analog Signal. A SIGNAL, such as voice or music, that varies in a continuous manner. An analog signal may be contrasted with a DIGITAL SIGNAL,
which represents only discrete states.
Answer Delay. The time from the beginning of RINGING until the called
station or an operator answers.
Area Code. See NUMBERING PLAN AREA.
Attempt. A call initiation or bid for service in which at least one digit is
received by the originating SWITCHING SYSTEM. For some purposes, an
attempt is said to occur as soon as an originating station goes OFF-HOOK
and causes some response in the originating switching system.
Attempts per Circuit per Hour (ACH). A running count of the number
of TRUNK requests (new calls). Used in network management as an
indicator of calling pressure. See CONNECTIONS PER CIRCUIT PER HOUR.

Glossary

781

Attenuation. A decrease in SIGNAL amplitude during TRANSMISSION from one
point to another, usually expressed in DECIBELS (dB).
Automatic Identified Outward Dialing (AIOD). An arrangement provided with CENTREX service whereby stations associated with the service can be identified automatically when originating TOLL calls or for
other purposes.
Automatic Intercept Center (AI C). A centrally located set of equipments
that is a part of an AUTOMATIC INTERCEPT SYSTEM. STORED-PROGRAM CONTROL is used to inform the calling customer (by means of a recorded
or electronically assembled announcement) why connection to the
called number cannot be completed.
Automatic Intercept System (AIS). A type of TRAFFIC service system consisting of one or more AUTOMATIC INTERCEPT CENTERS and a centralized
intercept bureau for handling INTERCEPT CALLS.
Automatic Message Accounting (AMA). The automatic collection, recording, and processing of information relating to calls for billing
purposes.
Automatic Number Identification (ANn. The automatic identification of
a calling station by a SWITCHING SYSTEM, usually for AUTOMATIC MESSAGE
ACCOUNTING.
Automatic Voice Network (AUTOVON). A PRIVATE NETWORK serving the
Department of Defense. AUTOVON employs automatic switching
and handles both voice and data TRAFFIC. It is worldwide; the continental United States portion is known as CONUS AUTOVON.
Average Busy Season Busy Hour (ABSBH) Load. The expected OFFERED
LOAD for which a TRUNK GROUP is engineered. Estimated by averaging
the measured loads for a given peak hour of the day (BUSY HOUR) during each weekday of the peak season (BUSY SEASON). Normally, measurements of load on 20 weekdays are averaged.
Bandlimited. Applied to a SIGNAL or a CHANNEL that is limited in frequency
content.
Bandpass Filter. A CIRCUIT designed to allow only frequencies within a
specific range to pass. The cutoff frequencies must be finite and
nonzero. The band of frequencies between the cutoff frequencies is
called the passband.
Bandwidth. The range of frequencies that can be transmitted by a communications CHANNEL, a TRANSMISSION FACILITY, or a transmission
medium, expressed in hertz (Hz).

782

Glossary

Baseband Channel. A CHANNEL that carries a SIGNAL faithfully without
requiring MODULATION, in contrast to a passband channel.
Baseband Signal. The original form of a SIGNAL, unchanged by MODULATION.
Bell Administrative Network Communications System (BANCS). A
computer system consisting of modules that form a message SWITCHING
NETWORK for business communications. BANeS handles computer-tocomputer, interactive terminal-to-computer, and store-and-forward
output DISTRIBUTION traffic.
Bit Rate. The speed at which digital signals are transmitted, usually
expressed in bits per second (bps).
Blocked Calls Cleared (BCC). Designation for a queuing system in which
demands (calls) that find no idle seTvers leave the system immediately. Commonly used to model systems for which waiting positions
are not provided, such as TRUNK GROUPS.
Blocked Calls Delayed (BCD). Designation for a queuing system in
which demands (calls) that find no idle servers wait until an idle
server becomes available (that is, they never give up). Commonly
used to model systems for which customers are not overly impatient,
such as digit receivers.
Blocking. The inability of the calling party to be connected to the called
party because either (1) all suitable TRUNK paths are busy or (2) a
path between a given inlet and any suitable free outlet of the SWITCHING NETWORK of a SWITCHING SYSTEM is unavailable.
Bridged Tap. A cable pair connected in parallel with a customer LOOP.
The connection (tap) may occur at the CENTRAL OFFICE or at some point
along a cable route.
Bridging Connection. A parallel CONNECTION that draws some of the SIGNAL
energy from a CIRCUIT, often with imperceptible effect on the circuit's
normal operation.
Broadband Channel. A transmission CHANNEL with a BANDWIDTH wider than
that required for transmitting voice SIGNALS, for example, 48 kilohertz
(kHz), 240 kHz.
Business Information Systems Programs (BISP). One of many computerbased systems for performing voluminous business and administrative operations associated with the provision of telephone service by
an OPERATING COMPANY.
Busy Hour. The hour(s) of the day during which TRAFFIC normally peaks.
Hours during which there are peaks for abnormal reasons (holidays,
special events) are not considered. Traffic systems are typically sized
for busy hour demand levels.

Glossary

Busy Season. A period during the year when
highest.

783

BUSY HOURS

are at their

Busy Season Busy Hour (BSBH). The hour of the business day that, on
the average, is the busiest hour during the BUSY SEASON.
Busy Tone. An audible
because the called
minute.

indicating that a call cannot be completed
is busy. The tone is applied 60 times per

SIGNAL
LINE

Cable Entrance Facility (CEF). The entrance area in a telephone EQUIPMENT
building for all types of OUTSIDE PLANT cables carrying subscriber LINES
and interoffice TRANSMISSION FACILITIES. The typical cable entrance facility
is a vault-like, below-grade area, 15 feet high and 12 feet wide, that
runs the length of the building directly under the main DISTRIBUTING
FRAME(S).

Capacity Expansion. A NETWORK PLANNING activity that determines types,
sizes, locations, and timing of SWITCHING SYSTEM and TRANSMISSION FACILITY installations.
Carried Load. The average number of busy servers in a TRAFFIC system. In
BLOCKED-CALLS-DELAYED systems, carried load equals OFFERED LOAD since all
calls are served eventually. In BLOCKED-CALLS-CLEARED systems, carried
load is less than offered load since some calls are denied service.
Carrier-Derived Channel. One of a number of

CHANNELS

provided by a

CARRIER SYSTEM.

Carrier Frequency. A sinusoidal waveform with constant amplitude that
undergoes MODULATION by an information SIGNAL to shift the information signal frequencies to a higher frequency band.
Carrier Frequency Shift. The change in frequencies between transmitting
and receiving terminals in a nonsynchronized CARRIER SYSTEM.
Carrier Signal. A SIGNAL suitable for MODULATION by an information signal.
It may be a CARRIER FREQUENCY, or it may be a stream of constantamplitude pulses as in PULSE-CODE MODULATION.
Carrier System. A TRANSMISSION system in which one or more CHANNELS of
information are processed and converted to a form suitable for the
transmission medium used by the system. Common types of carrier
systems are frequen,cy division, in which each information channel occupies an assigned portion of the frequency spectrum, and time division,
in which each information channel uses the transmission medium for
periodic, assigned time intervals.
Centralized Automatic Message Accounting (CAMA). A process using
centrally located EQUIPMENT (including a switchboard or a TRAFFIC SERVICE POSITION), associated with a tandem or TOLL switching office, for

784

Glossary

automatically recording billing data for customer-dialed extracharge
calls originating from several local CENTRAL OFFICES. A tape record is
processed at an electronic data-processing center. See TANDEM SWITCHING SYSTEM.
Centralized Automatic Reporting on Trunks (CAROT). An OPERATIONS
SYSTEM that automatically schedules tests of TRUNKS, performs the tests,
and analyzes and records the results. CAROT also performs tests of
trunks on demand to verify trouble and repair conditions.
Centralized Automatic Trouble Locating and Analysis System (CATLAS). An OPERATIONS SYSTEM designed as a maintenance aid for
Stored-Program Control Systems (SPCSs). When trouble is detected
and diagnosed, CATLAS automates trouble location procedures that
identify faulty circuit packs. See TOTAL NETWORK DATA SYSTEMS.
Centralized System for Analysis and Reporting (CSAR). An OPERATIONS
SYSTEM that measures the accuracy, timeliness, and completeness of
the TOTAL NETWORK DATA SYSTEM data flow and the consistency of its
record bases.
Central Office. Usually used to refer to a LOCAL SWITCHING SYSTEM that connects LINES to lines and lines to TRUNKS. It may be more generally
applied to any network switching system. The term is sometimes used
loosely to refer to a telephone company building in which a SWITCHING SYSTEM is located and to include other EQUIPMENT (such as transmission system terminals) that may be located in such a building.
Central Office Code. A 3-digit identification under which up to 10,000
station numbers are subgrouped. EXCHANGE AREA boundaries are associated with the central office code, which accordingly has billing
significance. Several central office codes may be served by a CENTRAL
OFFICE.
Central Office Equipment Engineering System (COEES). A time-sharing
OPERATIONS SYSTEM that assists in the planning and engineering of local
swi tching EQUIPMENT.
Centrex. A service for customers with many stations that permits stationto-station dialing, one listed directory number for the customer,
DIRECT INWARD DIALING to a particular station, and station identification
on outgoing calls. The switching functions are performed in a CENTRAL OFFICE.
Channel. A TRANSMISSION path between two points. May refer to a I-way
path or, when paths in the two directions of transmission are always
associated, to a 2-way path. It is usually the smallest subdivision of a
transmission system by means of which a single type of communications service (that is, a voice, teletypewriter, or data channel) is
provided.

Glossary

785

Channel Bank. TERMINAL EQUIPMENT used to combine (MULTIPLEX) CHANNELS
on a frequency-division or time-division basis. Voice channels are
combined into 12- or 24-channel groups.
Circuit. Frequently used interchangeably with CHANNEL to designate a
communications path between two or more points. Other meanings
include: (1) a configuration of interconnected NETWORK equipment
that provides a TRANSMISSION capability and (2) a closed path through
which current can flow.
Circuit Administration Center (CAC). An OPERATIONS CENTER that administers
the TRUNK
network.
The
functions
of
the
CAC
include: (1) determining demand and BUSY-SEASON trunk requirements
and issuing MESSAGE TRUNK orders to provide the required
trunks, (2) developing forecasts of trunk requirements for the NETWORK for 1 to 5 years, and (3) network routing.
Circuit Design. The OPERATING COMPANY process that specifies the types of
NETWORK equipment required to be interconnected to satisfy a functional capability.
Circuit Order. The document used to transmit engineering design of a
TRUNK or SPECIAL-SERVICES CIRCUIT for the PUBLIC SWITCHED TELEPHONE NETWORK to the department that implements the design.
Circuit Order Control (COC). A component system of the TRUNKS
INTEGRATED RECORDS KEEPING SYSTEM (TIRKS) that controls telephone company MESSAGE TRUNK, SPECIAL-SERVICES, and CARRIER SYSTEM orders. Reports
produced by cac provide management with the current status of CIRCUIT ORDERS. cac provides data to other TIRKS component systems to
update the assignment status of EQUIPMENT, FACILITIES, and CIRCUITS as
orders are processed.
Circuit Provision Center (CPC). An OPERATIONS CENTER that assigns EQUIPMENT and FACILITIES and prepares and distributes work documents for
MESSAGE TRUNK circuits, designed SPECIAL-SERVICES CIRCUITS, and CARRIER SYSTEMS. It also generates and maintains CIRCUIT RECORDS and inventory
and assignment records for all interoffice facilities and equipment.
Circuit Provisioning. The OPERATING COMPANY process that responds to
needs for TRUNKS and SPECIAL-SERVICES CIRCUITS. It includes CIRCUIT DESIGN,
assignment of specific components, and generation of work documents for the required installation work.
Circuit Record. The OPERATING COMPANY document that records the specific
configuration of EQUIPMENT assigned to a CIRCUIT.
Circuit Routing. A NETWORK PLANNING activity that determines the most
efficient configuration of TRANSMISSION FACILITIES to provide the
required CIRCUITS.

786

Glossary

Circuit Switched Digital Capability (CSDC). A NETWORK capability that
provides a high-speed, digital path over portions of the PUBLIC
SWITCHED TELEPHONE NETWORK. CSDC provides services such as audiographics TELECONFERENCING and bulk data transport.
Coaxial Cable. A type of cable made up of coaxial units, or tubes. Each
tube contains an inner conductor that is centered within an outer
conductor through the use of insulating disks spaced about 1 inch
apart. The outer conductor forms a cylinder around the disks. Each
cable contains from four to twenty-two of these coaxial tubes.
Common Carrier. A supplier in an industry that undertakes to "carry"
goods, services, or people from one point to another for the public in
general or for specified classes of the public. In telecommunications,
such carriage relates to the provision of TRANSMISSION capabilities over
the telecommunications NETWORK. A common carrier that offers communications services to the public is subject to regulation by federal
and state regulatory commissions.
Common-Channel Interoffice Signaling (CCIS). A SIGNALING system,
developed for use between SWITCHING SYSTEMS with STORED-PROGRAM CONTROL, in which all of the signaling information for one or more
groups of TRUNKS is transmitted over a dedicated high-speed data LINK,
rather than on a per-trunk basis. CelS can reduce call-setup time and
save money compared to individual trunk signaling.
Common Control. An automatic arrangement in which items of control
equipment in a SWITCHING SYSTEM are shared; they are associated with a
given call only during the periods required to accomplish the control
functions. All Bell System crossbar and ELECTRONIC SWITCHING SYSTEMS
have common control.
Common-Control Switching Arrangement (CCSA). An arrangement in
which SWITCHING for a PRIVATE NETWORK is provided by one or more
COMMON-CONTROL switching systems. The SWITCHING SYSTEMS may be
shared by several private networks and may also be shared with the
PUBLIC SWITCHED TELEPHONE NETWORK. This service provides uniform dialing to customers who use a NETWORK of dedicated TRANSMISSION FACILITIES
between geographically dispersed locations.
Common-Language Equipment Identification (CLEI) Code. An alphanumeric code that identifies NETWORK equipment by family, subfamily,
and type.
Common-Language Location Identification (CLLI) Code. A lO-character
designation that identifies any physical location within the Bell
System.

Glossary

787

Common System. A system that provides common power, INTERCONNECTION, or environmental support for NETWORK elements associated with
TRANSMISSION and SWITCHING; for example, power systems, DISTRIBUTING
FRAMES, and equipment building systems.
Common Update/Equipment (CU /EQ). A master record base that stores
the TRAFFIC measurement requests generated by OPERATING COMPANY
personnel.
Common Update/Trunking (CU /TK). A record base system that contains
trunking NETWORK information and other information required by
TRUNK SERVICING and TRUNK FORECASTING SYSTEMS.
Community Dial Office (CDO). A small, electromechanical SWITCHING SYSTEM that serves a separate EXCHANGE AREA and, ordinarily, has no
operating or maintenance force located in its own building; operation
is handled and maintenance is directed from a remote location.
Compandor. An abbreviation of compressor-expandor, a device used to
compress the range of talker volumes at the input to a CARRIER SYSTEM
(in particular, to increase low-level talker volumes) and to expand the
received volumes at the output of the carrier system (to provide the
complementary function and to make the transmission system transparent). This technique improves the SIGNAL-TO-NOISE RATIO for lowlevel talkers and provides a substantially reduced received NOISE level
during the so-called quiet intervals.
Computer Subsystem (CSS). The computer system used by a SWITCHING
CONTROL CENTER (SCC) or by several SCCs. It interfaces with the ESS
SWITCHING SYSTEMS supported by SCCs.
Computer System for Mainframe Operations (COSMOS). A WIRE CENTER
administration system operating in real time for subscriber services.
The objective of COSMOS is to minimize congestion and long crossconnects on main DISTRIBUTING FRAMES while maintaining load balance
across the SWITCHING equipment in the wire center.
Concentrated Range Extension with Gain (CREG). A design method that
enables the increased utilization of finer gauge cables in the LOOP
plant through the use of switched range extension shared by many
customers.
Concentration. (1) Applies to a SWITCHING NETWORK (or portion of one) that
has more inputs than outputs. (2) In a TRAFFIC NETWORK, combining
calls arriving on many LINES or TRUNKS to convey them more efficiently
to other TRANSMISSION or SWITCHING equipment. (3) Locating as much
EQUIPMENT as possible at a given place to achieve economies in such
things as building costs, power arrangements, and maintenance.

788

Glossary

Connection. (1) Generally, in terms of a telephone connection, a 2-way
voiceband CIRCUIT completed between two points by means of one or
more SWITCHING SYSTEMS. It contains two LOOPS and may contain one or
more TRUNKS. (2) A point where a junction of two or more conductors
is made.
Connections per Circuit per Hour (CCH). A running count of the
number of TRUNK connections between SWITCHING SYSTEMS. Used in NETWORK management as an indicator of switching congestion. See
ATTEMPTS PER CIRCUIT PER HOUR.
Coordinate Network. A SWITCHING NETWORK connecting stages of COORDINATE
SWITCHES.
Coordinate Switch. A switch with contacts or crosspoints arranged in a
matrix, or gridlike, structure. The crosspoints are usually fine-motion
electromechanical elements or electronic switching elements.
CORNET. A PRIVATE NETWORK serving Western Electric and Bell Laboratories. CORNET (a contraction of corporate network) is an example of
an ENHANCED PRIVATE SWITCHED COMMUNICATION SERVICE.
Country Code. The 1-, 2-, or 3-digit number that, in the world NUMBERING
PLAN, identifies each country or integrated numbering plan in the
world. The initial digit is always the WORLD ZONE NUMBER. Any subsequent digits in the code further define the designated geographic
area (normally identifying a specific country). On an international
call, the country code is dialed before the national number.
Craft Force. OPERATING COMPANY personnel with specialized technical training who install and maintain EQUIPMENT.
Crossbar Switch. A form of COORDINATE SWITCH and the basic element of
any crossbar system. A crossbar switch is a relay mechanism consisting
of ten horizontal paths and ten or twenty vertical paths. Any horizontal path can be connected to any vertical path by means of magnets. A 2-stage operation is used to close any crosspoint. First, a
selecting magnet shifts all selecting fingers in a horizontal row; then
a holding magnet shifts a vertical actuating card to close the selected
contacts.
Crosstalk. Interference in a communications CHANNEL caused by a SIGNAL
traveling in an adjacent channel. Telephone crosstalk may be intelligible or unintelligible to the parties engaged in conversation.
Customer-Line Signaling. The interaction between the customer and the
SWITCHING SYSTEM that serves the customer.
Customer-Premises Equipment. See EQUIPMENT.

Glossary

789

Cutover. A brief interval in an overall conversion period when operation
actually changes from an existing to a new system or system
configuration. In some cases, the change occurs almost instantaneousl y and may be called a flash cut.
Data Circuit. A NETWORK equipment configuration that provides a capability for data services.
DATAPHONE Digital Service. A service in which calls are placed over

the PUBLIC SWITCHED TELEPHONE NETWORK in the normal manner or
automatically, and after a connection is established, DATA TERMINALS are
connected at both ends for exchange of data. The term applies to
PRIVATE-LINE SERVICE as well.
Data Set. Equipment that converts SIGNALS (usually DIGITAL SIGNALS) from
data processors or other TERMINAL EQUIPMENT into signals suitable for
TRANSMISSION over telephone lines and controls the connection. Data
sets can be transmitters, receivers, or both. That portion of a data set
that converts terminal signals for transmission (modulator) and
received line signals for delivery to the terminal (demodulator) is
called a MODEM (a contraction of modulator/demodulator). The terms
data set and modem are often used interchangeably.
Data Terminal. A device that is used with a computer system for data
input and output. If it is situated at a location remote from the computer system, it requires data TRANSMISSION. Examples of data terminals
include teletypewriters and magnetic tape readers. The term also
applies to devices for terminal-to-terminal communications.
Data Under Voice (DUV). An arrangement for transmitting 1.544megabits per second (Mbps) pulse streams in the BANDWIDTH available
underneath the portion of the baseband used for voice CHANNELS on
existing microwave systems.
Decibel (dB). A logarithmic measure of the ratio between two powers:
~

dB = 10 loglO

P2

p-;.

Named for Alexander Graham Bell.
Demarcation Point. In the INTERCONNECTION environment, the physical and
electrical boundary between EQUIPMENT or FACILITIES provided by an
OTHER COMMON CARRIER and Bell System facilities.
Demodulation. The process of restoring a SIGNAL to its original form at the
receiving end of a TRANSMISSION system.
Dial. The part of a station set that generates a coded SIGNAL to control the
CENTRAL OFFICE switching equipment in accordance with the digits

790

Glossary

dialed. It may be either a rotary or pushbutton device. (See STATION
EQUIPMENT.) The term is sometimes used as an adjective, as in dial
administration, the process of short-term rearranging and performance
monitoring in a central office SWITCHING SYSTEM.
DIAL-IT Network Communications Service. Any of several services in
which customers dial advertised telephone numbers to reach an
announcement. Examples of DIAL-IT services are Public Announcement Service (Sports-Phone, Dial-a-Joke, etc.) and Media Stimulated
Calling (telephone voting, telethons, and media promotions).

Dial Tone. An audible tone sent from an automatic SWITCHING SYSTEM to a
customer to indicate that the EQUIPMENT is ready to receive dial SIGNALS.
Dial-Tone Delay. A measure of time required to provide DIAL TONE to customers. This measures one aspect of the performance of a SWITCHING
SYSTEM.
Differential Phase-Shift Keying (DPSK). A MODULATION technique for
transmitting digital information in which that information is conveyed by selecting discrete phase changes of the CARRIER SIGNAL. At
the receiving end, phase changes are detected by comparing the
phase of each SIGNAL element with the phase of the preceding signal
element.
Digital Carrier Trunk (OCT). An internal INTERFACE that combines certain
T-carrier TRANSMISSION functions and electronic SWITCHING SYSTEM control
functions.
Digital Channel. A transmission CHANNEL that carries SIGNALS in digital
form.
Digital Data System (DDS). A nationwide, PRIVATE-LINE, synchronous data
communications NETWORK formed by interconnecting digital TRANSMISSION FACILITIES and providing special maintenance and testing capabilities. Customer CHANNELS operate at 2.4, 4.8, 9.6, or 56 kilobits per
second (kbps).
Digital Signal. A SIGNAL that has a limited number of discrete states prior
to TRANSMISSION. A digital signal may be contrasted with an ANALOG SIGNAL, which varies in a continuous manner and may be said to have
an infinite number of states.
Digital Signal (OS) Level. One of several TRANSMISSION rates in the TIMEDIVISION MULTIPLEX hierarchy. For example, the DSl level is 1.544
megabits per second (Mbps).
Digital System Cross-Connect. An internal INTERFACE that acts as a central
point for cross-connecting, rearranging, patching, and testing digital
EQUIPMENT and FACILITIES.

791

Glossary

Digital Transmission. A mode of TRANSMISSION in which all information is
transmitted in digital form, that is, as a serial stream of pulses. Any
ANALOG SIGNAL -such as voice-can be converted into a DIGITAL SIGNAL.
Digroup. A digitally multiplexed group of twenty-four CHANNELS. Digroup
usually refers to the T1 carrier line SIGNAL of 1.544 megabits per
second (Mbps); however, the term is also used to refer to the digital
CHANNEL BANK that provides the 24-channel MULTIPLEXING function.
Direct Distance Dialing (DOD). The automatic establishment of TOLL calls
in response to SIGNALS from the dialing device of the originating
customer.
Direct Inward Dialing (DID). A feature that permits incoming calls to
stations served by a PRIVATE BRANCH EXCHANGE or by a CENTREX to be
dialed directly; the call need not go through an attendant.
Direct Services Dialing Capability (DSDC). A set of service-independent
NETWORK capabilities that will allow the creation of specific services to
meet specific customer needs. The capabilities are provided by a set
of primitives in a SWITCHING SYSTEM that can be summoned into use with
any service. Examples of primitives are: route the call, make a billing record, play an announcement.
Direct Trunk. A

TRUNK

between two class 5 offices
hierarchy.

(END OFFICES)

in the

PUB-

LIC SWITCHED TELEPHONE NETWORK

Distributing Frame. A manually operated hardware system used to interconnect NETWORK elements (OUTSIDE PLANT cables, SWITCHING and
TRANSMISSION equipment, etc.) to provide telecommunications services.
Distribution. In a SWITCHING NETWORK, distribution refers to the capability of
connecting an input to anyone of several outputs. In a TRAFFIC NETWORK, distribution refers to separating calls on incoming TRUNK GROUPS at
a tandem TOLL OFFICE and recombining them on other outgoing trunk
groups.
Double-Sideband Amplitude Modulation (DSBAM). Amplitude MODULATION in which the modulated wave is accompanied by both of the
sidebands resulting from modulation; the upper sideband
corresponds to the sum of the carrier and modulation frequencies,
whereas the lower side band corresponds to the difference between
the carrier and modulation frequencies. [DST]
Dual-Tone Multifrequency (DTMF). A means of SIGNALING that uses a
simultaneous combination of one of a lower group of frequencies and
one of a higher group of frequencies to represent each digit or
character.

792

Glossary

Echo. A wave that has been reflected or otherwise returned with
sufficient magnitude and delay to be perceived in some manner as a
wave distinct from that directly transmitted. Note: Echoes are frequently measured in DECIBELS (dBs) relative to the directly transmitted
wave. See TALKER ECHO. [IEEE]
Echo Canceler. A device that detects transmitted speech signals, generates
a SIGNAL that is a replica of the ECHO, and subtracts this signal from the
actual echo, thereby canceling the echo.
Echo Suppressor. A device that detects speech signals transmitted in
either direction on a 4-wire CIRCUIT and introduces LOSS in the opposite
direction of speech TRANSMISSION to suppress echoes.
Economic CCS (ECCS). The load that should be carried on the last TRUNK
of a HIGH-USAGE GROUP to minimize the total cost of routing the offered
TRAFFIC, assuming that overflow from the high-usage route is offered
to an alternate route engineered to meet an objective BLOCKING probability. See CARRIED LOAD.
Economic Evaluation. Analyzing the economic impact on a company's
overall financial position of designing, producing, or implementing a
product or service.
800 Service. Also called INWATS.

See WIDE AREA TELECOMMUNICATIONS

SERVICES.
Electromechanical Switching System. An automatic SWITCHING SYSTEM in
which the control functions are performed principally by devices,
such as relays and servos, that are electrically operated and have
mechanical motion.
Electronic Switching System. A class of modern SWITCHING SYSTEMS in
which the control functions are performed principally by electronic
devices.
Electronic Tandem Switching (ETS). A PRIVATE NETWORK service that provides customers with a uniform NUMBERING PLAN and numerous callrouting features. The electronic tandem switching functions are furnished by the SWITCHING equipment that provides PRIVATE BRANCH
EXCHANGE or CENTREX service.
Electronic Translator. The equipment in No. 4A and No. 5A Crossbar
Systems that functions primarily to translate ADDRESS codes, by means
of electronic circuitry and STORED-PROGRAM CONTROL, into information
required by the system to select an available route toward the CENTRAL
OFFICE of the called customer. See CROSSBAR SWITCH.
E&M Lead Signaling. A specific form of INTERFACE between a SWITCHING SYSTEM and a TRUNK in which the SIGNALING information is transferred

Glossary

793

across the interface via 2-state voltage conditions on two leads, each
wi th a ground return, separate from the leads used for MESSAGE information. The message and signaling information are combined (and
separated) by a signaling system appropriate for application to the
TRANSMISSION FACILITY. The term E&M lead signaling is used also in
some SPECIAL-SERVICES applications.
End Office. A local SWITCHING office where LOOPS are terminated for purposes of INTERCONNECTION to each other and to TRUNKS. End offices are
deSignated class 5 offices.
End Office Toll Trunk. A HIGH-USAGE TRUNK from an END OFFICE that carries
TOLL traffic only. It may be either a DIRECT TRUNK to another end office
or a trunk to a TOLL CENTER in another toll center area (that is, not the
toll center on which the end office homes).
Engineered Capacity. The highest load level for a TRUNK GROUP or a SWITCHING SYSTEM at which SERVICE OBJECTIVES are met.
Engineering and Administrative Data Acquisition System (EADAS). An
OPERATIONS SYSTEM in which TRAFFIC data are measured at SWITCHING SYSTEMS by electronic devices, transmitted to a centrally located minicomputer, and recorded on magnetic tape in a format that is suitable
for computer processing and analysis.
Engineering and Administrative Data Acquisition System/Network
Management (EADAS/NM). An OPERATIONS SYSTEM that monitors
SWITCHING SYSTEMS and TRUNK GROUPS that have been designated by network managers. EADAS/ NM reports existing or anticipated congestion on a display board at operating company NETWORK MANAGEMENT
CENTERS.
Engineering Period. A particular time period during which service is
measured and compared to the objective GRADE OF SERVICE. Sufficient
EQUIPMENT must therefore be provided (engineered) to meet SERVICE OBJECTIVES during this period.
Enhanced Private Switched Communication Service (EPSCS). A PRIVATE
NETWORK service that, like the COMMON-CONTROL SWITCHING ARRANGEMENT,
provides a uniform dialing plan for customers with geographically
dispersed locations. EPSCS offers 4-wire TRANSMISSION (to improve
transmission quality) within the private network and a Customer
Network Control Center, which can be used by the customer to control some network operations and to obtain private network usage
and status information.

EPLANS Computer Program Service. Software systems used by OPERATING
COMPANY engineering and related personnel to support their planning, recordkeeping, implementation, scheduling, ordering, NETWORK

794

Glossary

performance evaluation, network characterization, and other similar
activities. The programs are Western Electric products a.nd are
offered as time-share or batch-run computer services by Western Electric or, in some cases, are run in telephone company data centers.
Equipment. A term broadly applied to hardware components that
includes customer-premises equipment (ePE) and SWITCHING and
TRANSMISSION (and other) components located in telephone company
buildings.
Equipment and Facility Recovery. The reduction of capital investment
for additional EQUIPMENT accomplished by providing accurate records
of what capital equipment is already available to meet a current need.
Equipment-to-Equipment Interface. Any INTERFACE between EQUIPMENT
units on a user premises that is not considered a NETWORK interface.
Erlang. A dimensionless unit of TRAFFIC intensity used to express the average number of calls underway or the average number of devices in
use. One erlang corresponds to the continuous occupancy of one
traffic path. Traffic in erlangs is the sum of the holding times of paths
divided by the period of measurement. The term erlang can be used
to express the capacity of a system; for example, a TRUNK GROUP of
thirty TRUNKS, which, in a theoretical peak sense, might carry 30
erlangs of traffic, would have a typical capacity of perhaps 25 erlangs
averaged over an hour. Named for A. K. Erlang, the founder of the
traffic theory.
Erlang B. One of the basic TRAFFIC models and related formulas used in
the Bell System. The assumptions are POISSON input, negative
exponential holding times, and BLOCKED CALLS CLEARED. U sed for TRUNK
engineering. Also called Erlang's Loss Formula.
I

Erlang C. One of the basic TRAFFIC models and related formulas used in
the Bell System. The assumptions are POISSON input, negative
exponential holding times, and BLOCKED CALLS DELAYED. U sed for
COMMON-CONTROL engineering. Also called Erlang's Delay Formula.
Error-Second. A measurement of system performance for digital TRANSMISSION FACILITIES. An error-second is a I-second interval during which
one or more bit errors occur.
Exchange Area. Traditionally, an area within which there is a single uniform set of charges for telephone service. An exchange area may be
served by a number of CENTRAL OFFICES. A call between any two points
within an exchange area is a local call.
Expanded 800 Service. An improvement over the basic 800 SERVICE, which
uses DIRECT SERVICES DIALING CAPABILITY to provide customers with more

Glossary

795

options in defining service areas and determining the treatment a call
receives.
Expansion. The term applied to a SWITCHING NETWORK (or portion of one)
that has more outputs than inputs.
Facilities Assignment and Control System (FACS). An on-line dataprocessing system that maintains inventories and provides assignment of OUTSIDE PLANT and CENTRAL OFFICE facilities.
Facilities Network. The aggregate of TRANSMISSION systems, SWITCHING SySTEMS, and STATION EQUIPMENT; it supports a large number of traffic NETWORKS.
Facility. Anyone of the elements of physical telephone plant that are
needed to provide service. Thus, SWITCHING SYSTEMS, cables, and
microwave radio TRANSMISSION systems are examples of facilities. Facility is sometimes used in a more restricted sense to mean TRANSMISSION
FACILITY.
Facility and Equipment Planning System (FEPS). A component system
of the TRUNKS INTEGRATED RECORDS KEEPING SYSTEM (TIRKS) that supports
the current planning process. FEPS helps planners use information
in the TIRKS data bases along with forecasts of future growth to allocate existing inventories efficiently, to determine future EQUIPMENT and
FACILITY requirements, and to update planning designs.
Federal Communications Commission (FCC). A board of commissioners,
appointed by the president of the United States under the Communi- '
cations Act of 1934, having the power to regulate interstate and
foreign communications originating in the United States by wire and
radio.
Feeder Route. A network of LOOP cable extending from a WIRE CENTER into a
segment of the area served by the wire center.
Field Representative. A member of the Bell Laboratories Quality
Assurance Center responsible for monitoring the operation of Bell
System telecommunications EQUIPMENT in the field and participating in
field evaluations and reliability studies.
Final Group. A TRUNK GROUP that acts as a final (last-chance) route for
TRAFFIC. Traffic can overflow to a final group from HIGH-USAGE GROUPS
that are busy. Calls blocked on a final group are not offered to
another route.
Final Trunk. A TRUNK in a FINAL GROUP.
Flat Rate. A rate-setting principle for local service in which customers in
a specific group or area are all charged the same rate for local calling

796

Glossary

regardless of the number of local calls they make or the length of the
calls.
Focused Overload. Abnormal calling from many points to one point; for
example, after a disaster or when a radio or television station
encourages mass calling.
Forecasting. A NETWORK PLANNING activity that provides estimates of future
demands for existing and new services.
Foreign Exchange (FX) Service. A service that provides a CIRCUIT between
a customer's MAIN STATION or PRIVATE BRANCH EXCHANGE and a CENTRAL
OFFICE other than the one that normally serves the EXCHANGE AREA in
which the customer is located.
Foreign Numbering Plan Area (FNPA). Any NUMBERING PLAN AREA (NPA) outside the boundaries of the home NPA.
Frame. (1) A segment of an ANALOG SIGNAL or DIGITAL SIGNAL that has a
repetitive characteristic in that corresponding elements of successive
frames represent the same things. Examples are a television frame,
which represents a complete scan of a picture, or a TELEMETRY frame,
which represents values of a number of parameters in a specific
order. In a TIME-DIVISION MULTIPLEX system, a frame is a sequence of time
slots, each containing a sample from one of the CHANNELS served by
the multiplex system; the frame is repeated at the sampling rate, and
each channel occupies the same sequence position in successive
frames. (2) An assembly of EQUIPMENT units.
Framing Bit. A non-information-carrying bit introduced into a bit stream
to facilitate the separation of characters at the receiving end of a
TRANSMISSION.

Frequency Content. The band of frequencies or specific frequency components contained in a SIGNAL. For example, the frequency content of
a voiceband signal includes components between 200 and 3500 hertz
(Hz).

Frequency-Division Multiplex (FDM). A method of providing a number
of simultaneous CHANNELS over a common TRANSMISSION path by using a
different frequency band for the transmission of each channel.
Frequency Modulation (FM). One way to modify a SIGNAL to make it
"carry" information. The frequency of a CARRIER SIGNAL is modified in
accordance with the amplitude of the information signal.
Frequency-Shift Keying (FSK). A MODULATION technique for transmitting
digital information having two, or possibly more, discrete states.
Each of the discrete states is represented by an associated frequency.
The most common form is binary FSK, which uses two frequencies to
represent the two states.

Glossary

797

Fundamental Planning. A NETWORK PLANNING activity that develops longrange plans for changes and growth in NETWORK structure.
Gain. An increase in SIGNAL power during TRANSMISSION from one point to
another, usually expressed in DECIBELS (dBs). Also called amplification.
[Te]

General Trade. A Bell System term for manufacturers and suppliers of
telecommunications EQUIPMENT other than Western Electric.
Generic Program. A set of instructions for an electronic SWITCHING SYSTEM
that is the same for all offices using that type of system. Detailed
differences for each individual office are listed in a separate parameter table.
Grade of Service (GOS). (1) An estimate of customer satisfaction with a
particular aspect of service (such as NOISE or ECHO). It combines the
distribution of subjective opinions of a representative group of people with the distribution of performance for the particular aspect
being graded. For example, with a specified distribution of noise, 95
percent of the people may judge the noise performance to be good or
better; the noise grade of service is then said to be 95 percent good or
better. (2) In traffic NETWORKS, the proportion of calls that receive no
service (BLOCKING) or poor service (long delay).
High-Usage (HU) Group. A TRUNK GROUP between two SWITCHING SYSTEMS
that is designed for high average occupancy. To provide an overall
acceptable probability of BLOCKING, calls blocked on a high-usage group
are offered to other routes.
High-Usage (HU) Trunk. A TRUNK in a HIGH-USAGE GROUP.
Home Numbering Plan Area (HNPA). The NUMBERING PLAN AREA within
which the calling line appears at a local (class 5) switching office (END
OFFICE).

Human Factors. A scientific discipline that takes human behavioral
characteristics and physical capabilities into account when designing
products to be used or tasks to be performed.
Hundred Call Seconds (CCS) Per Hour. A unit of TRAFFIC used to express
the average number of calls in progress or the average number of
devices in use. Numerically, it is 36 times the traffic expressed in
ERLANGS.

Hybrid. A NETWORK having four ports and designed so that when the
ports are properly terminated, the SIGNAL input tOrany particular port
splits equally between the two adjacent ports with essentially no signal coupled to the opposite port. Hybrids are used to couple 4-wire
CIRCUITS to 2-wire circuits.

798

Glossary

Impulse Noise. Short-duration, high-amplitude bursts, or spikes, of NOISE
energy, much greater than the normal peaks of MESSAGE CIRCUIT NOISE
on a transmission CHANNEL.
Inband Tone Signaling. SIGNALING that uses the same path as a MESSAGE
and in which the signaling frequencies are in the same band used for
the message.
Independent Telephone Company. A telephone company, not affiliated
with the Bell System, that has its own "independent" territory. In
1981, there were more than 1450 independent telephone companies in the
United States.
Individual Circuit Analysis (lCAN) Program. Part of the TOTAL NETWORK
DATA SYSTEM that detects ELECTROMECHANICAL SWITCHING SYSTEM equipment
faults by identifying abnormal load patterns on individual CIRCUITS
within a circuit group. These faults, for example, include defective
circuits that prevent customer calls from being completed. lCAN produces reports used by the NETWORK ADMINISTRATION CENTER.
Intercept Calls. Calls directed by a customer to an improper telephone
number that are redirected to an operator or a recording. The caller·
is told why the call could not be completed and, if possible, given
the correct number.
Intercity Facility Relief Planning System (lFRPS). An OPERATIONS SYSTEM
that aids in FACILITY planning for a TOLL network. Its function is similar to the METROPOLITAN AREA TRANSMISSION FACILITY ANALYSIS PROGRAM.
Intercom Calling. Intralocation calling; calls between stations on the
same customer premises.
Interconnection. The direct electrical connection, acoustical coupling, or
inductive coupling of user-premises TERMINAL EQUIPMENT (including terminal equipment that is a part of a separate communications system)
to the telephone NETWORK. It also includes the direct electrical connection of OTHER COMMON CARRIER facilities to the telephone network.
Interface. A common boundary between two systems or pieces of EQUIPMENT where they are joined.
Interface Specification. A technical requirement that must be met at an
INTERFACE.
International Telecommunications Satellite Consortium (lNTELSAT).
An international organization established in 1964 to govern a global
commercial communications satellite system to provide communications between many countries. Membership is in excess of eighty
countries. The Communications Satellite Corporation (COMSAT) acts
as manager for lNTELSAT and also represents the United States.

Glossary

799

Interoffice Call. A call between two SWITCHING SYSTEMS.
Interoffice Facilities Network. Part of the nationwide FACILITIES NETWORK,
consisting of interoffice TRANSMISSION FACILITIES, TANDEM SWITCHING SYSTEMS,
and LOCAL SWITCHING SYSTEMS. See LOCAL FACILITIES NETWORK.
Interoffice Trunk Signaling. The exchange of call-handling information
between SWITCHING offices within the NETWORK.
Intertoll Trunk. A TRUNK between two TOLL OFFICES.
Intraoffice Call. A call involving only one SWITCHING SYSTEM.
Jumpers. Temporary wires used on a DISTRIBUTING FRAME to cross-connect
the TERMINATION points of cables from particular EQUIPMENT and FACILITIES
to provide service to a customer.
Key Telephone Set (KTS). A telephone set with buttons, or keys, located
on or near the telephone, used as part of a KEY TELEPHONE SYSTEM.
Key Telephone System. An arrangement of KEY TELEPHONE SETS and associated circuitry, located on a customer's premises, that provides combinations of certain voice communications arrangements such as call
pickup, call hold, call line status lamp signals, and INTERCONNECTION
among on-premises stations without the need for connection through
the CENTRAL OFFICE or PRIVATE BRANCH EXCHANGE.
Large-Scale Integrated (LSI) Circuit. An integrated CIRCUIT containing 100
gates or more on a single chip, resulting in an increase in the scope
of the function performed by a single device.
License Contract. The legal agreement that governed the relationship
between AT&T and a Bell OPERATING COMPANY. Each license contract
described the reciprocal services, licenses, and privileges that existed
between the parties. All license contracts terminated with divestiture
of the Bell operating companies.
Line. (1) From a SWITCHING viewpoint, the LOOP-, STATION EQUIPMENT-, and
CENTRAL OFFICE-associated EQUIPMENT assigned to a customer. (2) From a
TRANSMISSION viewpoint, the transmission path between a customer's
station equipment and a SWITCHING SYSTEM. In this sense, it is also
called a LOOP. (3) In CARRIER SYSTEMS, the portion of a transmission system between two terminal locations. The line includes the transmission media and associated line REPEATERS. (4) The side of a piece of
CENTRAL OFFICE equipment that connects to or toward the OUTSIDE PLANT;
the other side of the equipment is called the drop side. (5) A family
of equipment or apparatus designed to provide a variety of styles, a
range of sizes, or a choice of service features, for example, a "product
line."

800

Glossary

Linear Distortion. Distortion resulting from a CHANNEL having a linear
filter characteristic different from an ideal linear LOW-PASS or BANDPASS
FILTER; in particular, amplitude-versus-frequency characteristics that
are not flat over the passband, and phase-versus-frequency characteristics that are not linear over the passband. See BANDPASS FILTER.
Line Build-Out (LBO) Network. Amplifiers (REPEATERS) in a cable
TRANSMISSION system may be designed to compensate for distortion
over a specific length of cable. When the length of cable between
amplifiers is less than that for which the amplifier is designed, one or
more line build-out networks are used to bring the distortion to approximately the design level.
Line Fill. The ratio of assigned CIRCUITS to total capacity in a FACILITY or
EQUIPMENT unit.
Link. A TRANSMISSION FACILITY in the telecommunications NETWORK.
Load Balance System (LBS). A CENTRAL OFFICE reporting system and part of
the TOTAL NETWORK DATA SYSTEM. LBS helps assure network administrators that TRAFFIC loads in SWITCHING SYSTEMS are uniformly distributed. For example, reports generated by LBS are used to determine
the "lightly loaded" line groups to which new subscriber LINES can be
assigned.
Load Balancing. Procedures mainly used for equalizing the TRAFFIC load
on customer TERMINATION groups in SWITCHING NETWORKS.
Load Lost. In a TRAFFIC system, the portion of OFFERED LOAD that is not
served because all servers are busy and all waiting positions (if any
are provided) are occupied.
Local Automatic Message Accounting (LAMA). A process using EQUIPMENT located in a local office for automatically recording billing data
for MESSAGE RATE SERVICE calls (bulk billing) and for customer-dialed
station-to-station TOLL calls. The tape record is sent to and processed
at an electronic data-processing center.
Local Facilities Network. Part of the nationwide FACILITIES NETWORK, consisting of local SWITCHING SYSTEMS located at WIRE CENTERS and the loop
TRANSMISSION FACILITIES through which customers are connected to local
switching systems. Local switching systems are considered the conceptual boundary between the local facilities network and the INTEROFFICE FACILITIES NETWORK and, in this sense, belong to both networks.
See INTEROFFICE FACILITIES NETWORK.
Local Switching System. A SWITCHING SYSTEM that performs END OFFICE (class
5) functions. Local switching systems connect customer LINES directly to
other customer lines, or customer lines to TRUNKS.

801

Glossary

Long-Haul Trunk. A FINAL TRUNK or HIGH-USAGE TRUNK that interconnects
two regions in the PUBLIC SWITCHED TELEPHONE NETWORK hierarchy.
Long-Route Design. A codification of design practices used to plan customer LOOPS that exceed the resistance design limit of the serving CENTRAL OFFICE.

Loop. A

CHANNEL between a customer's terminal and a
loop may also be called a LINE.

CENTRAL OFFICE.

A

Loop Assignment Center (LAC). An OPERATIONS CENTER that assigns customer LOOP facilities, telephone numbers, CENTRAL OFFICE line EQUIPMENT,
and miscellaneous central office equipment.
Loop-Reverse-Battery. A method of SIGNALING over interoffice TRUNKS in
which dc changes, including directional changes associated with battery reversal, are used for SUPERVISION. This technique provides 2-way
signaling on 2-wire trunks; however, a trunk can be seized at only
one end; that is, it cannot be seized at the office at which battery is
applied. Also called reverse battery signaling.
Loop Signaling. A method of SIGNALING over dc CIRCUIT paths that utilizes
the metallic LOOP formed by the LINE or TRUNK conductors and terminating circuits.
Loss. In the Bell System, the term refers to insertion loss, a quantity that
represents a specific relationship between the input and output of a
NETWORK (for example, a customer connection or a CIRCUIT). The basic
insertion loss calculation determines the difference in DECIBELS (dBs)
between power applied to a load directly and power applied to a
load through a network.
Loss Objective. An objective for the amount of LOSS that can be tolerated
in NETWORK components and still maintain a satisfactory GRADE OF
SERVICE.

Loudness Loss. A measure used to express the LOSS of communications
paths in a manner that reflects loudness perception. For partial and
overall telephone CIRCUITS, loudness loss is the ratio of suitably
weighted output SIGNAL levels to input signal levels. (The signals
may be electric or acoustic.)
Low-Pass Filter. A filter having a single TRANSMISSION band extending
from zero to some finite cutoff frequency.
Main Station. A telephone that is connected directly to a CENTRAL OFFICE by
either an individual or shared LINE. Main stations include the principal telephone of each party on a party line. They do not include
telephones that are manually or automatically connected to a central
office through a PRIVATE BRANCH EXCHANGE or extension telephones (that

802

Glossary

is, telephones that have been added to an individual or shared line to
extend telephone service to other parts of the subscriber's home or
business premises).
Marker (Crossbar). The heart of COMMON-CONTROL crossbar CENTRAL OFFICE
equipment. Markers perform the following functions in a No. 5
Crossbar SWITCHING SYSTEM: (1) determine terminal locations of calling
LINES, incoming TRUNKS bidding for service, called lines, and outgoing
trunks in the EQUIPMENT; (2) determine the proper route for the call,
establish the connection within the office, and pass routing information to the senders; (3) determine the calling line class of service and
provide charge classification; (4) recognize line busy, trouble, intercept, and vacant-line conditions; and (5) call in a trouble recorder
when necessary. See CROSSBAR SWITCH.
Market. The set of actual or potential buyers of a product or service.
Marketing. Those activities that influence the flow of goods and services
from producers to consumers.
Marketing Mix. The set of variables that can be adjusted to attract MARKETS
to a product or service. These include variables associated with the
product itself and those related to price, product distribution, and
promotion of the product.
Market Segmentation. The division of a MARKET into submarkets differing
in such characteristics as customer needs and buying behavior.
Master Control Center (MCC). A FRAME of EQUIPMENT in an ELECTRONIC
SWITCHING SYSTEM office with lamps that show the current state of the
office equipment and with keys for operating controls.
Measurement Plan. A compilation of service and performance measurements for operational units (for example, OPERATIONS CENTERS).
Message. (1) In telephone communications, a successful call attempt that
is answered by the called party and followed by some minimum
period of CONNECTION. (2) In data communications, a set of information, typically digital and in a specific code such as the American
Standard Code for Information Interexchange (ASCII), to be carried
from a source to a destination. A header, with ADDRESS and other
information regarding handling, may be considered part of or
separate from the message.
Message Circuit Noise. The short-term average NOISE level as measured
with a 3A noise measuring set or its equivalent. This set includes
frequency weighting and time constants to make the set most sensitive to noise that will impair TRANSMISSION quality in telephone CIRCUITS
used for speech.

Glossary

803

Message Rate Service. Telephone service for which a charge is made in
accordance with a measured amount of usage, referred to as message

units.
Message Telecommunications Service (MTS). Non-PRIVATE-LINE intrastate
and interstate long-distance telephone service.
Message Trunk. A TRUNK carrying MESSAGE TELECOMMUNICATIONS SERVICE traffic
on the PUBLIC SWITCHED TELEPHONE NETWORK.
Metropolitan Area Transmission Facility Analysis Program (MATFAP).
A program that aids in facility planning for metropolitan NETWORKS.
Using various measures, MATFAP analyzes the alternatives available
for future TRANSMISSION FACILITIES and EQUIPMENT and identifies what
transmission plant is needed at various locations and when it will be
needed. MA TF AP also determines the economic consequences of
selecting particular facilities and/or equipment and of selecting particular routes and provides the least-cost assignment of CIRCUITS to
each facility as a guide to the CIRCUIT-PROVISIONING process.
Microprocessor. The control and processing portion of a microcomputer
built with LARGE-SCALE INTEGRATED CIRCUITRY, usually on one chip.
Microprocessors can handle both arithmetic and logic functions under
control of a program stored in a memory chip.
Minimum-Cost Routing. A CIRCUIT-ROUTING scheme that determines a path
through the NETWORK for each point-to-point demand for each year, so
that, when point-to-point demands are provided on these paths and
the resulting CAPACITY EXPANSION problem is solved, the total cost of
TRANSMISSION FACILITIES is minimized.
Mobile Telephone Service. One of a class of services that uses radio CHANNELS to provide telephone service to customers on the move. Mobile
telephone services include land mobile telephone service, BELLBOY
personal signaling set paging service, air / ground service, marine
radiotelephone services, and high-speed train telephone services.
Modem. A contraction of the words modulator and demodulator signifying
an EQUIPMENT unit that performs both of these functions. See DATA
SET.
Modular Engineering. The expression of EQUIPMENT quantities (typically,
TRUNK GROUPS) as an integer multiple of some basic unit, the unit
depending on physical constraints. Reflects the reality that certain
equipment items can only be added in multiples, rather than one
item at a time.
Modulation. The process by which the amplitude, frequency, or phase of
a CARRIER SIGNAL is varied in accordance with one of the characteristics
of an information SIGNAL.

804

Glossary

Multifrequency (MF) Signaling. An inband, interoffice, ADDRESS SIGNALING
method in which ten decimal digits and five auxiliary SIGNALS are
each represented by selecting two frequencies out of the following
group: 700, 900, 1100, 1300, 1500, and 1700 hertz (Hz).
Muldem. A contraction of the words multiplexer and demultiplexer signifying an EQUIPMENT unit that performs both of these functions.

a

Multiplex(ing). The EQUIPMENT or process for combining number of individual CHANNELS into a common frequency band or into a common bit
stream for TRANSMISSION. The converse equipment or process for
separating into individual channels is called demultiplex(ing).
Narrowband Channel. A transmission CHANNEL with a BANDWIDTH narrower
than that required for transmitting voice SIGNALS.
Nationwide Rate Averaging. A rate-setting method that has been used to
establish rates for interstate telephone service; like services and like
calling distances carry the same charges even though the costs of
interstate service to the COMMON CARRIER may differ for routes in lowdensity, high-cost areas and for routes in high-density, low-cost
areas.
Negative Exponential Distribution. A probability distribution function
used in many queuing models to describe the distribution of the
length of completed telephone calls.
Network. (1) The FACILITIES NETWORK is the aggregate of TRANSMISSION systems, SWITCHING SYSTEMS, and STATION EQUIPMENT; it supports a large
number of traffic networks. (2) A TRAFFIC NETWORK is an arrangement of
CHANNELS, such as LOOPS and TRUNKS; associated SWITCHING arrangements;
and station equipment, designed to handle a specific body of TRAFFIC.
A traffic network is a subset of the facilities network. (3) An
electrical/electronic CIRCUIT, usually packaged as a single piece of
apparatus or on a printed circuit pack. Examples are a transformer
network and an equalization network. (4) The switching stages and
associated INTERCONNECTIONS of a switching system are collectively
called the SWITCHING NETWORK.
Network Administration Center (NAC). An OPERATIONS CENTER with
administrative responsibility for local and TANDEM SWITCHING SYSTEMS.
The functions performed by an NAC include data administration,
dail y surveillance of service and load in the SWITCHING NETWORK, EQUIPMENT utilization, machine and TRUNK GROUP performance analysis and
monitoring, and participation in Job Contact Committees associated
with equipment additions, replacements, and rearrangements.
Network Channel-Terminating Equipment. EQUIPMENT located on user
premises that is a part of the telephone NETWORK facility.

Glossary

805

Network Control Point (NCP). A NODE in the STORED-PROGRAM CONTROL network that connects to a SIGNAL TRANSFER POINT in the COMMON-CHANNEL
INTEROFFICE SIGNALING (CCIS) network. The NCP's associated data base
contains customer-service data accessed by SIGNALS routed over the
CCIS network and used to support extended NETWORK services.
Network Data Collection Center (NDCC). An OPERATIONS CENTER that
administers NETWORK data collection. Primarily, the NDCC supervises
the operation and maintenance of the Engineering and Administrative Data Acquisition System (EADAS), lA EADAS central control
units, data LINKS, and data collection apparatus.
Network Interface (NI). In the INTERCONNECTION environment, the physical
and electrical boundary between two separately owned telecommunications capabilities. It also serves as the boundary for administrative
and maintenance activities.
Network Management Center (NMC). An OPERATIONS CENTER responsible
for surveillance and control of TRAFFIC flow in a specific geographic
area. Control is an ongoing activity in response to OVERLOADS, especially from peak day calling, mass calling, NETWORK system failure, or
network rearrangements. The NMC also plans strategies for potential
overload situations.
Network Operations Center (NOC). The OPERATIONS CENTER, located in
Bedminster, New Jersey, that oversees and coordinates management
of the North American message NETWORK. It monitors the status of
the intertoll network among the regions and coordinates the use of
interregional SWITCHING SYSTEM and trunking network capacity that is
temporarily spare. The NOC also directs the network management of
international TRAFFIC flows for the United States and monitors the
status of TOLL facilities and the effect of any problems on the interregional intertoll network. In case of problems in the toll facility
network, the NOC sets restoration priorities and coordinates restoration activities.
Network Operations Center System (NOCS). The primary support system for the NETWORK OPERATIONS CENTER located in Bedminster, New Jersey. NOeS provides near real-time surveillance of major SWITCHING
SYSTEMS and their associated trunking at a nationwide level.
Network Planning. A multifaceted discipline that encompasses the functions involved in planning the evolution and implementation of the
nationwide NETWORK, designing and engineering the configuration of
the network, and managing the total network investments.
Network Service Center (NSC). An OPERATIONS CENTER responsible for
ensuring the overall quality of NETWORK service, keeping management

806

Glossary

informed about levels and trends in the quality of service provided
by the network, and stimulating improvement activities when
instances of substandard service are identified.
New Equipment-Building System Standards (NEBS). A set of integrated
specifications for both telecommunications EQUIPMENT and equipment
buildings.
No.5 Crossbar Central Office Equipment Reports (5XB COER). A CENTRAL OFFICE equipment reporting system for No.5 Crossbar offices.
The system analyzes TRAFFIC data in support of NETWORK administration
and design functions.
Node. A SWITCHING office or
tions NETWORK.

FACILITY

junction point in the telecommunica-

Noise. An unwanted disturbance introduced in a communications CIRCUIT.
It may partially or completely obscure the information content of a
desired SIGNAL. On telephone circuits, noise may be an annoyance
during quiet intervals as well as when speech is present.
Nonlinear Distortion. A type of distortion in which the output SIGNAL
amplitude does not have the desired linear relationship to the input
signal amplitude.
Numbering Plan. A numbering system for a switched telephone NETWORK
that identifies each MAIN STATION by a unique ADDRESS that is convenient, readily understandable, and similar in format to that of
other main stations connected to the network.
Numbering Plan Area (NPA). The familiar area code, defining a geographic division within which telephone directory numbers are
subgrouped. In North America, a 3-digit NO / IX code is assigned to
denote each NPA, where:

N = any digit 2 through 9
X = any digit 0 through 9
0/1 = 0 or 1.
Occupancy. The fraction of time that a CIRCUIT or an EQUIPMENT unit is in
use, expressed as a decimal. Numerically, it is the ERLANGS carried per
circuit. Occupancy typically includes both MESSAGE time and call-setup
time.
Offered Load. The demand placed on a TRAFFIC system, defined by the
product of two parameters: the average rate at which customers place
demands on the system and the average length of time they require
service.

Glossary

807

Off-Hook. Station switchhook contacts closed, resulting in LINE current, or
whatever SUPERVISION condition is indicative of the in-use or requestfor-service state.
One-Way Trunk. A TRUNK that can be seized for use by the SWITCHING
equipment at one end only. Once a trunk is seized, 2-way TRANSMISSION may occur.
On-Hook. Station switchhook contacts open or whatever SUPERVISION condition is indicative of the EQUIPMENT-idle state.
Operating Company. A regulated telephone company whose primary
business is providing telephone service to customers.
Operations Center. A group of people, reporting to a common manager,
who perform a set of closely related functions for a specific geographic area, group of customers, or service.
Operations Planning. An ongoing activity to ensure that changes in the
roles and responsibilities of people are linked to changes in the telephone business and that operations-related functions are assigned to
people and OPERATIONS SYSTEMS in ways that realize potentials for
greater efficiency and better customer service.
Operations Process. The sequence of interactions between OPERATIONS SYSTEMS and OPERATIONS CENTERS that are required to perform a particular
operation, such as adding a TRUNK to the NETWORK or maintaining a
SWITCHING SYSTEM.
Operations System. A computer-based system that OPERATING COMPANY
employees use to support operations activities. An operations system
does not directly provide telecommunications service to customers,
but supports operating company personnel in the performance of
their duties, such as testing TRUNKS or maintaining SWITCHING SYSTEMS.
Operations Systems Network. The collection of OPERATIONS SYSTEMS, communications terminals at OPERATIONS CENTERS, and the switched and
direct communications LINKS interconnecting them and connecting
them to a variety of telecommunications systems (for example, an
ELECTRONIC SWITCHING SYSTEM).
Operator Code. A code of the form lXX, 1 lXX, or llXXX that allows outward TOLL operators to reach inward, directory assistance, or other
operators in distant cities.
Operator Number Identification (ONI). Operator identification of a calling station, usually for billing purposes, when automatic
identification from a local office is not available.
Operator Services. A variety of services normally performed by operators. These include completing or helping customers to complete

808

Glossary

TOLL calls and assistance calls; preparing billing inputs on those calls;
providing directory assistance; intercepting and helping customers
with calls to changed or nonworking numbers; providing SPECIAL SERVICES, such as person-to-person, coin, calling card, collect, mobile,
PICTUREPHONE MEETING SERVICE, and conference calls; and giving onthe-job consultation to business customers.
Optical Fiber. A glass fiber that provides a I-way path for light SIGNALS.
Used in lightguide cable.
Other Common Carrier (OCC). A telecommunications COMMON CARRIER
authorized by the FEDERAL COMMUNICATIONS COMMISSION (FCC) to provide
a variety of services. The FCC refers to these carriers as domestic
satellite carriers, miscellaneous common carriers, and specialized common
carriers.
Outpulsing. Sending ADDRESS or other SIGNALING information over a LINE or
TRUNK.
Outside Plant. The part of the telephone system that is located physically
outside of telephone company buildings. Outside plant includes
cables, supporting structures, and certain EQUIPMENT items; it does not
include microwave towers, antennas, and cable system REPEATERS.
Outstate Facility Network Planning System (OFNPS). An interactive
computer system that aids in facility planning for rural NETWORKS. It
is similar in function to the METROPOLITAN AREA TRANSMISSION FACILITY
ANALYSIS PROGRAM.
Overflow. A count of all calls that are offered to a TRUNK GROUP but are not
carried (see PEG COUNT); usually measured for one hour.
Overload. (1) In TRANSMISSION, a load greater than that which a device is
designed to handle; may cause overheating of power-handling components and distortion in SIGNAL circuits. [DSl1 (2) For telecommunications TRAFFIC, an overload is an increase in OFFERED LOAD beyond trre
capacity for which NETWORK components (for example, TRUNKS and
SWITCHING SYSTEMS) are engineered.
Packet. A group of bits that is switched as an integral unit. Typically, a
packet contains data, destination and origination information, and
control information, arranged in a particular format.
Packet-Switching Network. A NETWORK that is designed to transport and
switch data in PACKET form.
Pair Gain. Referring to a system that uses digital or analog carrier techniques to serve many customers over a few pairs between the CENTRAL
OFFICE and a remote electronic terminal.

Glossary

809

Peakedness. A telecommunications TRAFFIC term signifying the ratio of the
variance of the load to the mean of the load. For random traffic (POISSON arrivals, negative exponential holding times), the peakedness of
the load is 1.0, that is, the variance equals the mean. For traffic
overflowing a HIGH-USAGE GROUP, peakedness exceeds I, reflecting the
fact that OVERFLOW occurs in "bursts" or peaks.
Peak Load. Denotes a higher-than-average quantity of TRAFFIC; usually
expressed for a I-hour period and as any of several functions of the
observing interval, such as peak hour during a day, average of daily
peak hours over a 20-day interval, maximum of average hourly traffic
over a 20-day interval. Significantly higher peak loads occur infrequently as the result of catastrophes and on Mother's Day and Christmas Day.
Peg Count. A count of all calls offered to a TRUNK GROUP, usually measured
for 1 hour. As applied to units of SWITCHING SYSTEMS with COMMONCONTROL, peg count, or carried peg count, means the number of calls
actuall y handled.
Performance Measurement. A measurement intended to reflect whether
an operational unit (for example, an OPERATIONS CENTER) is meeting
objectives or whether NETWORK service characteristics and performance
parameters are at levels required to meet SERVICE OBJECTIVES.
Performance Objective. A statement of the operational objective to be
met by a component of the NETWORK or the network as a whole. For
example, the MESSAGE CIRCUIT NOISE objective for customer LOOPS is stated
in terms of an upper limit of 20 DECIBELS above reference noise, Cmessage weighted (dBrnC).
Per-Trunk Signaling. A method of SIGNALING in which the SIGNALS pertaining to a particular call are transmitted over the same TRUNK that carries the call. Interoffice signaling other than COMMON-CHANNEL INTEROFFICE SIGNALING falls into this category.

PICTUREPHONE Meeting Service. A service, supported by High-Speed
Switched Digital Service, that allows people in two distant, specially
equipped rooms to hold a fully interactive audio and video conference.
Plant. All of the FACILITIES (such as land, buildings, machinery, apparatus,
instruments, and fixtures) needed to provide telecommunications services. See OUTSIDE PLANT.
Plug-In Inventory Control System/Detailed Continuing Property
Records (PICS/DCPR). An OPERATIONS SYSTEM that inventories PLUG-IN
UNITS of EQUIPMENT in CENTRAL OFFICES. The DCPR portion serves as an
investment data base that supports a telephone company's accounting
records for all types of central office equipment.

810

Glossary

investment data base that supports a telephone company's accounting
records for all types of central office equipment.
Plug-In Unit. A prewired modular assembly of electronics components
designed to be plugged into a permanently wired receptacle. Most
contemporary telecommunications EQUIPMENT consists of many plug-in
units to perform a variety of functions.
Poisson. In TRAFFIC theory, Poisson refers to a distribution or a process
resulting in a distribution of events such that the intervals between
adjacent events are independent, random variables that are members
of identical exponential distributions. Under certain conditions, the
arrival of telephone calls to be routed over a TRUNK GROUP can be
approximated by a Poisson distribution. Named after a 19th-century
French mathematician.
Preferential Assignment. A method of attaining more short JUMPERS on a
DISTRIBUTING FRAME than would be possible if jumpers were assigned
randomly from lists of spares. The distributing frame is administered
in zones with the objective of running jumpers between adjacent
zones.
Prefix. Any SIGNAL dialed prior to the ADDRESS. Prefixes are used to place an
address in proper context, to indicate service options, or both. An
example is the prefix 0 used before an address where operator assistance or intervention is requested, such as for collect calls.
Premises Information System (PREMIS). An OPERATIONS SYSTEM that provides rapid access to information about EQUIPMENT located on customers' premises.
Present Worth. An adjustment made to amounts of money associated
with different times in the future to make the amounts equivalent to
present values for use in ECONOMIC EVALUATION.
Primary Center. A class 3 office in the hierarchy of TOLL switching offices.
See TOLL OFFICE.
Private Branch Exchange (PBX). A private SWITCHING SYSTEM, either manual
or automatic, usually serving an organization, such as a business or a
government agency, and usually located on the customer's premises.
Telephones served by the PBX are called stations. Calls from one station to another or to an external network such as the PUBLIC SWITCHED
TELEPHONE NETWORK may be handled manually or automatically,
depending on the type of PBX. DIRECT INWARD DIALING and AUTOMATIC
IDENTIFIED OUTWARD DIALING service (formerly called CENTREX-CU) can be
provided by some PBXs. TIE TRUNKS are commonly used between PBX
systems of a single customer.
Private Line. A circuit used for PRIVATE-LINE SERVICE.

Glossary

811

Private-Line Service. A service in which the customer leases a CIRCUIT, not
interconnected with the PUBLIC SWITCHED TELEPHONE NETWORK, for the customer's exclusive use. The PRIVATE LINE may be used for transmission
of voice, teletypewriter, data, television, etc.
Private Network. A NETWORK made up of CIRCUITS and, sometimes, SWITCHING arrangements, for the exclusive use of one customer. These networks can be nationwide in scope and typically serve large corporations or government agencies. An example of a private voiceband
network is the Bell System corporate network, CORNET.
Product Life Cycle. The length of time from introduction of a product or
service into the MARKET until the product or service becomes obsolete
and is replaced by a newer product. It typically has four stages:
introduction, growth, maturity, and decline.
Progressively Controlled Network. A SWITCHING NETWORK in which calls
are set up by making a series of connections, stage by stage, based on
the digits dialed.
Protector Devices. Fuse-like devices mounted on DISTRIBUTING FRAMES to
guard telecommunications EQUIPMENT against spurious voltages and
currents that might result from lightning strikes or power-line
crosses in the OUTSIDE PLANT.
Protocol. A strict procedure for the initiation and the maintenance of
data communications.
Public Communications Services. Services provided through telephones
installed at locations where a public need exists, such as airports, bus
and train stations, hotel lobbies, large business offices, public streets,
and highways.
Public Switched Telephone Network (PSTN). The portion of the total
NETWORK that supports PUBLIC SWITCHED TELEPHONE NETWORK SERVICES. It
provides the capability for interconnecting virtually any home or
office in the country with any other. Public is the key word; any
equipment indeterminately shared by more than one customer is part
of the PSTN. May also be called the public switched network, public telephone network, or the DIRECT DISTANCE DIALING (DDD) network.
Public Switched Telephone Network Services. Services provided by the
PUBLIC SWITCHED TELEPHONE NETWORK including MESSAGE TELECOMMUNICATIONS SERVICE (MTS), WIDE AREA TELECOMMUNICATIONS SERVICES (WATS), and
DATAPHONE DIGITAL SERVICE among others.
Public Utilities Commission (PUC). An agency charged with regulating
communications services as well as other public utility services, usually within a state.

812

Glossary

Pulse-Amplitude Modulation (PAM). A MODULATION technique in which
the amplitude of each pulse is related to the amplitude of an ANALOG
SIGNAL. Used, for example, in TIME-DIVISION MULTIPLEX arrangements in
which successive pulses represent samples from the individual
VOICEBAND CHANNELS; also used in time-division SWITCHING SYSTEMS of
small and moderate size.
Pulse-Code Modulation (PCM). Conversion of an ANALOG SIGNAL, such as
voice, to a digital format, ordinarily in terms of binary-coded pulses
representing the quantized amplitude samples of the analog signal.
Pulse Rate. The number of pulses transmitted per unit of time. May also
be called the baud rate. When the pulses or symbols have only two
possible values (binary), the pulse rate is also the BIT RATE.
Quality Assurance. Continuous independent verification of satisfactory
performance of products and services from the user's viewpoint.
Quality Assurance System. In the Bell System, a series of audits and continuous monitoring to determine the effect of QUALITY CONTROLS used
by the designers and producers of a product or service.
Quality Control. A set of procedures used by designers and producers of
products and services to provide sufficient control of machinery, person:uel, and material necessary to meet acceptable quality criteria in
an economic manner.
Regeneration. The process of receiving a DIGITAL SIGNAL and reconstructing
it in a form in which the amplitudes, waveforms, and timing of the
SIGNAL elements are constrained within specified limits. [IEC/ITU]
Regional Center. A class 1 office in the hierarchy of TOLL switching
offices; the highest level TOLL OFFICE.
Register. A part of an automatic SWITCHING SYSTEM that receives and stores
SIGNALS from a calling device or other source for interpretation and
action, some of which is carried out by the register itself.
Remote Memory Administration System (RMAS). An OPERATIONS SYSTEM
that changes translations in SWITCHING SYSTEMS.
Reorder Tone. A tone applied 120 times per minute that indicates all
SWITCHING paths are busy, all TOLL trunks are busy, EQUIPMENT blockages,
unassigned code dialed, or incomplete registration of digits at a tan~
dem or TOLL OFFICE; also called channel busy or fast busy tone.
Repeater. EQUIPMENT, essentially including one or several amplifiers
and/ or regenerators and associated devices, inserted at a point in a
TRANSMISSION. The medium may operate in one or both directions of
transmission. [IEe/ITU]

Glossary

813

Repertory Dialer. A piece of STATION EQUIPMENT that permits a user to dial
telephone numbers automatically from a preprogrammed directory.
Reserve Capacity. The amount of additional capacity (beyond that
needed to meet SERVICE OBJECTIVES) that should be provided in a longterm forecast to minimize the cost of underestimating the demand.
Reducing reserve capacity in the long-term forecast generates shortterm costs (for example, servicing during BUSY SEASONS). The
"optional" reserve capacity balances the long-term capital cost against
the short-term servicing cost.
Resistance Design. A design method for customer LOOPS in which an
attempt is made to employ cable having the highest gauge (smallest
wire) that will ensure a loop resistance less than the SIGNALING limit of
the CENTRAL OFFICE serving the loop.
Revenue Accounting Office (RAO). A telephone company center utilizing large, mainframe computers for billing and other data processing.
Functions performed include receipt and processing of AUTOMATIC MESSAGE ACCOUNTING (AMA) data, preparation of the customer's bill, directory preparation, MARKETING support, internal reports, and payroll and
inventory management.
Ringing. The process of alerting the called party by the application of an
intermittent 20-hertz (Hz) SIGNAL to the appropriate LINE; this produces
a ringing sound at the called telephone set. When the ringing signal is
applied to the called line, an intermittent signal called an audible ring
is sent to the calling telephone to indicate that ringing is taking place.
Sectional Center. A class 2 office in the hierarchy of TOLL switching
offices. See TOLL OFFICE.
Seize, Seizure. An action of a SWITCHING SYSTEM in selecting an outgoing
TRUNK or other component for a particular call.
Separation. A rate-setting requirement that specified that the costs of providing long-distance telephone service be appropriately distributed
between interstate services, which are under the jurisdiction of the
FEDERAL COMMUNICATIONS COMMISSION, and intrastate services, which are
under the jurisdiction of the state PUBLIC UTILITIES COMMISSIONS.
Service Code. A code, typically of the form NIl (N = any digit 2 through
9), that defines a connection for a service (for example, 411 for directory assistance).
Service Evaluation. The process of determining what customers and the
Bell System expect of a service, setting appropriate objectives based
on the expectations, and assessing compliance with objectives.

814

Glossary

Service Measurement. A measurement reflecting aspects of operations
perceivable by the customer.
Service Objective. A statement of the quality of service that is to be provided to the customer; for example, no more than 1.5 percent of customer calls should encounter a delay of more than 3 seconds for DIAL
TONE during the average BUSY HOUR. See GRADE OF SERVICE.
Service Order. An order prepared in the commercial department of an
OPERATING COMPANY, at the request of a customer, to establish, a service,
to change an existing service, or to terminate a service. The resultant
document contains all the information required to meet the customer's needs.
Service Representative. An individual in the business office of an OPERATING COMPANY who typically deals with customers.
Serving Area Interface. A rearrangeable cross-connect point between
feeder and DISTRIBUTION cables in the LOOP plant.
Settlement. An accounting procedure based on the total investment in
telephone EQUIPMENT; the total investment of a company determines
the base for the allowed earnings (called the rate base). Settlements
define how revenue from a single call is distributed among different
companies, both Bell and independent, involved in that connection.
Sidetone. The portion of the SIGNAL from a telephone transmitter that
appears at the receiver of that telephone. Some sidetone appears to be
desirable to assure the customer that the telephone is working and to
help the talker adjust the level of speech.
Signal. An electrical, optical, or other representation of information for
(1) MESSAGES; for example, voice, data, television; (2) NETWORK control;
for example, call routing, network management; (3) internal operation of network elements; for example, timing and control of SWITCHING SYSTEMS.
Signaling. The transmission of ADDRESS, SUPERVISION, or other SWITCHING
information between stations and SWITCHING SYSTEMS and between
switching systems, including any information required for billing.
Signal-to-Noise Ratio. The ratio of the average SIGNAL power at any point
in a TRANSMISSION path to the average NOISE power at that same point,
often expressed in DECIBELS (dBs).
Signal Transfer Point (STP). A switching NODE in the COMMON-CHANNEL
INTEROFFICE SIGNALING network. STPs operate under STORED-PROGRAM CONTROL to connect signaling LINKS to network SWITCHING SYSTEMS and other
STPs. They may also connect directly to NETWORK CONTROL POINTS.

Glossary

815

Single-Frequency (SF) Signaling. A method of conveying dial-pulse and
SUPERVISION signals from one end of a TRUNK or LINE to the other, using
the presence or absence of a single specified frequency. A 2600-hertz
(Hz) tone is commonly used.
Single-Sideband Amplitude Modulation (SSBAM). AMPLITUDE MODULATION
in which only one of the sidebands resulting from MODULATION is
selected for TRANSMISSION by a BANDPASS FILTER. A precise and stable CARRIER FREQUENCY is inserted at the receiving terminal for DEMODULATION.
Small Office Network Data System (SONDS). An OPERATIONS SYSTEM that
collects TRAFFIC data from small step-by-step offices, processes the data,
and provides reports to NETWORK administrators.
Span. A collection of span LINES between two offices. The term is also
used to refer to the collection of all SPAN LINES in a particular cable, all
span lines on a particular route, or all span lines between two offices.
Span Line. A repeatered Tl LINE section between two CENTRAL OFFICES (not
necessarily contiguous offices). A Tl CARRIER SYSTEM is made up of a
tandem combination of span lines, plus a digital CHANNEL BANK at each
terminal.
Speakerphone. An audio terminal, consisting of transmitter and
loudspeaker units, used with a telephone set for TELECONFERENCING.
Special Services. Services requiring special treatment with respect to
TRANSMISSION, SIGNALING, SWITCHING, billing, or customer use. Examples
are PRIVATE BRANCH EXCHANGE (PBX) service; WIDE AREA TELECOMMUNICATIONS SERVICE (WATS); FOREIGN EXCHANGE (FX) SERVICE; and PRIVATE-LINE SERVICES
such as CIRCUITS for voice, data, teletypewriter, and television.
Special-Services Circuit. A TRANSMISSION path used to provide SPECIAL SERVICES to a specific customer.
Standard Supply Contract. The legal agreement that has governed the
relationship between Western Electric and a Bell OPERATING COMPANY.
The supply contract required Western Electric either to manufacture or
to purchase materials that the operating companies might reasonably
require, which they might order from Western Electric. The supply
contract did not, however, obligate the operating companies to purchase these materials from Western Electric. The supply contract terminated at the time of divestiture.
Station Equipment. EQUIPMENT that allows a customer to access the NETWORK and the available services. The most common station equipment
is the ordinary single-line telephone set.

816

Glossary

Station Number. The final four digits of a standard 7- or 10- digit ADDRESS
that define a connection to a specific customer's line within a CENTRAL
OFFICE. See CENTRAL OFFICE CODE.
Step-by-Step (SXS) System. An automatic SWITCHING SYSTEM using step-bystep switches. In most such systems, a call is extended progressively,
step-by-step, to the desired terminal under direct control of pulses
from a customer's DIAL or from a sender.
Stored-Program Control (SPC). A form of SWITCHING SYSTEM control in
which system operations are controlled by a stored program executed
by one or more processors. Operation of the system can be altered
significantly by changing programs.
Stored Program Control System/Central Office Equipment Reports
(SPCS/COER). A series of time-shared programs that analyzes TRAFFIC
data for ELECTRONIC SWITCHING SYSTEM offices and produces reports.
Suffix. Any SIGNAL dialed after the ADDRESS.
ample, to indicate the end of dialing.

Used by operators, for ex-

Supervision. The constant monitoring and controlling of the status of a
call.
Switching. (1) Refers to the process of connecting appropriate LINES and
TRUNKS to form a desired communication path between two station
sets. Included are all kinds of related functions, such as sending and
receiving SIGNALS, monitoring the status of CIRCUITS, translating
ADDRESSES to routing instructions, alternate routing, testing circuits for
busy condition, and detecting and recording troubles. (2) Designates
a field of work, such as system development, planning, or engineering, involving the application of switching technology in telecommunications NETWORKS. (3) Refers, in a more restricted sense, to the
technology associated with any circuit that operates discretely, particularly logic and memory.
Switching Control Center (SCC). An OPERATIONS CENTER responsible for the
centralized installation and maintenance of a group of SWITCHING SYSTEMS in a geographic area.
Switching Control Center System (SCCS). The Computer Subsystem
(CSS) and the EQUIPMENT units that remote the MASTER CONTROL CENTER
capability of an ELECTRONIC SWITCHING SYSTEM. The sees provides for
the administration, control, and maintenance of electronic switching
systems from central locations.
Switching Network. SWITCHING stages and their INTERCONNECTIONS within a
SWITCHING SYSTEM.

Glossary

817

Switching System. An electromechanical or electronic system for connecting LINES to lines, lines to TRUNKS, or trunks to trunks. The term
includes PRIVATE BRANCH EXCHANGE switching systems and centrally
located NETWORK switching systems. See SWITCHING.
System Code. A 3-digit code, usually of the form lXX but including OXX
(X = any digit 0 through 9) assignments, available only to operators
or to SWITCHING equipment for use as part of a special or modified
ADDRESS to influence route selection. These codes are reserved for
system-wide use; that is, they are the same across all NUMBERING PLAN
AREAS.
Talker Echo. An ECHO of a talker's voice that is returned to the talker.
When there is delay between the original SIGNAL and the echo, the
effect is disturbing, unless the echo is attenuated to a tolerable level.
Tandem Switching System. A broad functional category representing
systems that connect TRUNKS to trunks. Tandem switching divides
into two applications: Those offices that connect trunks within a
metropolitan area are referred to as local tandem offices; Those offices
that connect trunks in the TOLL network portion (class 1 to class 4) of
the PUBLIC SWITCHED TELEPHONE NETWORK are called TOLL OFFICES.
Tandem Trunk. A TRUNK that connects WIRE CENTERS through a local tandem
office.
Tariff. The published rates, regulations, and descriptions governing the
provision of communications service.
T -Carrier Administration System (TCAS). An OPERATIONS SYSTEM responsible for T-carrier alarms.
Teleconferencing. Voice telephone service between a group of people
and one or more other groups or individuals.
Telemetry. The method or EQUIPMENT used to transmit status information
such as that represented by the operation of keys or by lamp displays
to a remote location.
Terminal Equipment. In the INTERCONNECTION environment, any separately
housed EQUIPMENT unit or a group of equipment units located on user
premises on the user side of a network INTERFACE.
Termination. (1) The points on a SWITCHING NETWORK to which a TRUNK or a
LINE may be attached. (2) An item that is connected to the terminals
of a CIRCUIT or piece of EQUIPMENT. (3) An impedance connected to the
end of a circuit being tested.
Termination Layout Mask. A plan that reserves space on a DISTRIBUTING
FRAME for different TERMINATION categories of EQUIPMENT and FACILITIES.

818

Glossary

Thus, as the office grows, there will be room on the
derly addition of new terminations.

FRAME

for the or-

Throughput. (1) The total useful information processed or communicated
during a specified time period. [Te] (2) A measure of the effective
rate of TRANSMISSION of data by a communications system. [DSn
Tie Cable. Cable that interlinks DISTRIBUTING FRAMES.
Tie Trunk. A
EXCHANGES

SPECIAL-SERVICES CIRCUIT connecting two
or equivalent SWITCHING SYSTEMS.

PRIVATE

BRANCH

Time-Division Multiplex (TDM). A method of serving a number of
simultaneous CHANNELS over a common TRANSMISSION path by assigning
the transmission path sequentially to the various channels, each
assignment being for a discrete time interv,al.
Time-Multiplexed Switch (TMS). An element of a time-division SWITCHING NETWORK that effectively operates as a very high-speed spacedivision switch whose input-to-output paths can be changed in every
time slot.
Time Sharing. The use of a FACILITY or piece of EQUIPMENT for more than
one purpose or function or for repetition of the same function within
the same overall time period. This is accomplished by interspersing
or interleaving the required actions in time.
Time-Slot Interchange (TSI). An element of time-division SWITCHING that
separates and switches SIGNALS from multiple calls that are presented
in a TIME-DIVISION MULTIPLEXED format.
Tip and Ring Conductors. The two conductors associated with a 2-wire
cable pair. The terms tip and ring derive their names from the physical characteristics of an operator's cord switchboard plug in which
these two conductors terminated in the days of manual switchboards.
Use of the names tip and ring has extended throughout the plant.
The cord switchboard plug also had a sleeve, and the name is occasionally used for a third conductor associated with tip and ring.
Toll. A term describing service that is a part of public telephone service
but under a TARIFF separate from the EXCHANGE AREA tariff. Also used to
describe components of the FACILITIES NETWORK that are used principally
for toll service.
Toll Center. A class 4 office in the hierarchy of toll switching offices; the
lowest level TOLL OFFICE.
Toll Center Code. A 3-digit code of the form OXX (X = any digit 0
through 9) that identifies a specific TOLL CENTER and is available only
for OPERATING COMPANY use.

Glossary

819

Toll Charge. A charge for telephone service for calls outside the designated local EXCHANGE AREA. TOLL service calls are billed individually.
Toll Connecting Trunk. A TRUNK between an END OFFICE and a TOLL OFFICE.
Toll Office. Those offices that connect TRUNKS in the TOLL network portion
(class 1 to class 4) of the PUBLIC SWITCHED TELEPHONE NETWORK. See T ANOEM SWITCHING SYSTEM.
Total Network Data System (TNDS). A coordinated family of OPERATIONS
SYSTEMS. TNDS consists of both manual procedures and computer systems that provide OPERATING COMPANY managers with comprehensive,
timely, and accurate NETWORK information. It supports OPERATIONS
CENTERS responsible for administration of the trunking network, network data collection, daily surveillance of the load on the SWITCHING
NETWORK, and design of local and CENTRAL OFFICE switching equipment
to meet future service demands.
Total Network Operations Plan (TNOP). A Bell System operations plan
that describes the OPERATIONS PROCESSES, OPERATIONS CENTERS, and OPERATIONS SYSTEMS to be used in administering and provisioning the
telecommunications NETWORK in the Bell OPERATING COMPANIES.
Traffic. The flow of information or MESSAGES through the NETWORK. This
information flow may be generated by telephone conversations or
may be the result of providing data, audio, and video services.
Traffic Data Administration System (TDAS). Part of the TOTAL NETWORK
DATA SYSTEM (TNDS) that formats and temporarily stores TRAFFIC data
for other TNDS systems.
Traffic Engineering. A NETWORK PLANNING activity that determines the
number and type of CHANNELS or communication paths required
between SWITCHING points and the call-handling capacity of the
switching points.
Traffic Network. An arrangement of CHANNELS (such as LOOPS and TRUNKS,
associated SWITCHING arrangements, and STATION EQUIPMENT) designed to
handle a specific body of TRAFFIC. Traffic networks are provided by the
FACILITIES NETWORK.
Traffic Service Position (TSP). A cordless console that is associated with
either a crossbar tandem office or a TRAFFIC SERVICE POSITION SYSTEM,
equipped so that operators can provide assistance, if needed, on
station-to-station calls, special TOLL calls, public telephone calls, and
all local and toll assistance TRAFFIC. The operators provide assistance
in completing these calls and ensure that correct data are recorded in
the centralized AUTOMATIC MESSAGE ACCOUNTING equipment or in the
Traffic Service Position System equipment. They also supervise coin

820

Glossary

deposits for calls originating at public telephones. The position is
arranged for automatic display of both the calling and called
numbers, as well as certain other information.
Traffic Service Position System (TSPS). A type of TRAFFIC service system,
with STORED-PROGRAM CONTROL that provides for the processing and
recording of special TOLL calls, public telephone toll calls, and other
types of calls requiring operator assistance. It includes TRAFFIC SERVICE
POSITIONS arranged in groups called Operator Office Groups, where
operators are automatically connected in on calls to perform the functions necessary to process and record the calls correct! y.
Traffic Theory. A branch of applied probability theory that produces
models used to determine the capacity requirements to meet SERVICE
OBJECTIVES of systems with nondeterministic demands.
Translation. The operation of converting information from one form to
another. In SWITCHING SYSTEMS, the process of interpreting all or part of
a destination code to determine the routing of a call.
Transmission. (1) Designates a field of work, such as EQUIPMENT development, system design, planning, or engineering, in which electrical
communication technology is used to create systems to carry information over a distance. (2) Refers to the process of sending information
from one point to another. (3) Used with a modifier to describe the
quality of a telephone connection: good, fair, or poor transmission.
(4) Refers to the transfer characteristic of a CHANNEL or NETWORK in
general or, more specifically, to the amplitude transfer characteristic.
One may hear the phrase, "transmission as a function of frequency."
Transmission Facility. An element of physical telephone PLANT that performs the function of TRANSMISSION; for example, a multipair cable, a
COAXIAL CABLE system, or a microwave radio system.
Transmission Level Point (TLP). A specification, in DECIBELS (dBs), of the
relative level at a· particular point in a TRANSMISSION system as referred
to a zero transmission level point (0 TLP). The TLP value does not
specify the absolute power that will exist at that point.
Transmission Objedive. Electrical performance characteristics for communication CIRCUITS, systems, and EQUIPMENTS based on both economic
and technical considerations of telephone facilities and on reasonable
estimates of the performance desired. Characteristics for which
objectives are stated include LOSS, NOISE, ECHO, CROSSTALK, frequency
shift, ATTENUATION distortion, envelope delay distortion, etc.
Trouble Ticket. A form containing either symptoms or detailed information about malfunctioning EQUIPMENT. It is given to a craftsperson
whose job is to locate and repair the equipment.

Glossary

821

Trunk. A communication CHANNEL between two SWITCHING SYSTEMS. The
term switching system includes CENTRAL OFFICE types, TOLL switching systems, PRIVATE BRANCH EXCHANGES, KEY TELEPHONE SYSTEMS, manual and
automatic switchboards, concentrators, etc.
Trunk Circuit. A CIRCUIT, part of a SWITCHING SYSTEM, associated with the
connection of a TRUNK to the switching system. It serves to convert
between the SIGNAL formats used internally in the switching system
and those used in the TRANSMISSION circuit, and it performs logic and
sometimes memory functions associated with SUPERVISION.
Trunk Forecasting System (TFS). Part of the TOTAL NETWORK DATA SYSTEM
that forecasts MESSAGE TRUNK requirements for 5 years in the future.
Trunk Group. A number of TRUNKS that can be used interchangeably
between two SWITCHING SYSTEMS.
Trunk Servicing System (TSS). Part of the TOTAL NETWORK DATA SYSTEM
that processes TRAFFIC data from the TRAFFIC DATA ADMINISTRATION SySTEM, computes OFFERED LOAD, and calculates TRUNK requirements. It is
used by trunk administrators to maintain the MESSAGE TRUNK
network.
Trunks Integrated Records Keeping System (TIRKS). An OPERATIONS SYSTEM for maintaining the inventory and assignment of the FACILITIES
and EQUIPMENT used to establish TRUNKS of all kinds.
Two-Way Trunk. A TRUNK that can be seized for use by the SWITCHING
equipment located at either end.
Unigauge Design. A design method for customer LOOPS that provides for
the exclusive use of 26-gauge cable on all loops within 30 kilofeet of
the CENTRAL OFFICE. Requires range extension equipment developed
specifically for the unigauge system.
Universal Telephone Service. The goal of establishing affordable and
available nationwide telephone service.
Usage-Sensitive Rate. A rate-setting principle that relates directly to customer use of EQUIPMENT and service: Those who use less, pay less.
Value-of-Service Pricing. A rate-setting principle that relates directly to
the customer density in a local calling area, the frequency of use, and
importance of service to the customer.
Voiceband Channel. A transmission CHANNEL with a nominal 4-kilohertz
(kHz) BANDWIDTH suitable for voice TRANSMISSION.

822

Glossary

Voice-Frequency (VF) Facility. An analog FACILITY that provides one
VOICEBAND CHANNEL and carries the information in the vOice-frequency
band.
Volatile Memory. A computer memory in which stored information is
lost if the power supply for the memory fails or is turned off.
Wide Area Telecommunications Services (WATS). A service that permits
customers to make (OUTWATS) or receive (lNWATS or 800 SERVICE)
long-distance calls and to have them billed on a bulk basis rather
than individually. WAYS is provided within selected service areas, or
bands, by means of special private-access LINES, which are connected
to the PUBLIC SWITCHED TELEPHONE NETWORK through WAYS-equipped CENTRAL OFFICES. A single access line permits inward or outward service
but not both.
Wire Center. The location of one or more local SWITCHING SYSTEMS; a point
at which customer LOOPS converge. May be loosely used to mean the
CENTRAL OFFICE building at that location.
Wire Center Area. The area surrounding a WIRE CENTER containing all customers, other than those with FOREIGN EXCHANGE SERVICE, whose LOOPS
are connected to a CENTRAL OFFICE at that wire center.
Wire Pair Cable. Cables composed of twisted pairs of wires rather than
coaxial tubes, fibers, etc.
Work Package. Material sent to OPERATING COMPANY field forces that
describes work to be performed.
World Numbering Plan. See COUNTRY CODE.
World Zone Number. A I-digit number that, in the world numbering
plan, identifies a geographic zone. The world zone number is the initial number in a COUNTRY CODE.

Acronyms and Abbreviations

The acronyms and abbreviations listed here reflect usage in this book.
They may be used differently in other contexts.

ABSBH
ACD
ACH
ACP
ACTS
ACU
ADS
AIC
AIOD
AIS
ALBO
ALGOL
AM
AMA
AMARC
AMAT
ANI
APS
ASCII
ASPEN

ATRS
AT&T

average busy season busy
hour
automatic call distributor
attempts per circuit per hour
action point
Automated Coin Toll Service
automatic calling unit
Auxiliary Data System
Automatic Intercept Center
automatic identified
outward dialing
Automatic Intercept System
automatic line buildout
Algorithmic Computer
Language
amplitude modulation
automatic message
accounting
Automatic Message Accounting Recording Center
automatic message accounting transmitter
automatic number
identification
automatic protection
switching
American Standard Code for
Information Interexchange
Automatic System for Performance Evaluation of the
Network
Automated Trouble Reporting System
American Telephone and
Telegraph Company

AUTOVON
BANCS

automatic voice network
Bell Administrative Network
Communications System
BCC
blocked calls cleared
BCD
blocked calls delayed
billing data transmitter
BDT
Business Information SysBISP
terns Programs
Bell operating company
BOe
BOSS
Billing and Order Support
System
bits per inch
bpi
bps
bits per second
Basic Packet-Switching
BPSS
Service
busy season busy hour
BSBH
B6ZS
bipolar with six zero
substitution
BSP
Bell System Practice
BSRFS
Bell System Reference
Frequency Standard
Btu
British thermal unit
Circuit Administration Center
CAC
Customer Administration
Center
CAMA
centralized automatic message accounting
CAMA-ONI centralized automatic message accounting-operator
number identification
CAROT
Centralized Automatic
Reporting on Trunks
Centralized Automatic
CATLAS
Trouble Locating and
Analvsis System

823

824

CCH
CCIR

CCIS
CCITT

CCS
CCSA
COA
COCF
COO
COT
CEF
CEV
CLEI
CLL!
CMDF
CMOS
CMS
CNAB
CNCC
CO
COC
co dec
COEES
COER
COMSAT
CONN
CONUS
CORNET
COSMIC

COSMOS
CPC

Acronyms and Abbreviations

connections per circuit per
hour
Comite Consultatif International des Radiocomm unications
common-channel interoffice
signaling
Comite Consultatif International Telegraphique et
Telephonique
hundred call seconds
common-control switching
arrangement
call data accumulator
cumulative discounted cash
flow
community dial office
call data transmitter
cable entrance facility
controlled environment
vault
common-language equipment
iden tification
common-language location
identification
combined main distributing
frame
Centralized Message Data
System
Call Management System
Customer Name and
Address Bureau
Customer Network Control
Center
central office
circuit order control
coder-decoder
Central Office Equipment
Engineering System
Central Office Equipment
Report
Communications Satellite
Corporation
connector
continental United States
corporate network
(Bell System)
Common Systems Main
Interconnection Frame System
Computer Sy&tem for
Mainframe Operations
Circuit Provision Center

CPO
CPE
CPU
CRC
CREG
CRIS
CRS
CRT
CSAR
CSOC
CSO
CSS
CSU
CUCRIT
CU/EQ
CU/TK
DACS
dB
DBAS
dBmO
dBrnC

DCE
DCM
DCP
DCPR

DCPSK
OCT
DDC
ODD
DDS
DERP
DEW
DFI
DIC
DID
DIF

central pulse distributor
customer-premises
equipment
central processing unit
Customer Record Center
concentrated range extension
with gain
Customer Records Information System
Centralized Results System
cathode-ray tube
Centralized System for
Analysis and Reporting
circuit-switched digital
capability
central services organization
Computer Subsystem
channel service unit
Capital Utilization Criteria
Common Update/Equipment
(System)
Common Update /Trunking
(System)
Digital Access and CrossConnect System
decibel
Data Base Administration
System
decibels with reference to a
power of 0 milliwatt
decibels above reference
noise, using C-message
weighting
data circuit-terminating
equipment
digital carrier module
duplex central processor
Detailed Continuing
Property Records
(PICS/DCPR)
differential coherent
phase-shift keying
digital carrier trunk
Direct Department Calling
direct distance dialing
Digital Data System
Defective Equipment
Replacement Program
Distant Early Warning (Line)
digital facility interface
digital interface controller
direct in ward dialing
digital interface frame

Acronyms and Abbreviations

DIM
DIU
DLTU
DOC
DOJ
DaM
DOV
DP
DPAC
DPP
DPSK
DR
OS
DSBAM
DSDC
DSn
DSS
DSU
DSX
DT

DTC
DTE
DTMF
DTU
DUV
EADAS

EADAS/NM

EBCDIC
ECCS
ECPT
EEl
EFRAP
EIA
EISS
E911

data in the middle
digital interface unit
digital line / trunk unit
dynamic overload control
Department of Justice
data on mastergroup
data over voice
demarcation point
dedicated plant assignment
card
discounted payback period
differential phase-shift
keying
data receive
digital signal
double-sideband amplitude
modulation
direct services dialing
capability
digital signal (level) n
data station selector
data service unit
digital system cross-connect
data transmit
digroup terminal
digroup terminal controller
data terminal equipment
dual-tone multifrequency
digroup terminal unit
data under voice
Engineering and Administrative Data Acquisition
System
Engineering and Administrative Data Acquisition
System/Network
Management
Extended Binary-Coded
Decimal Interchange Code
economic hundred call
seconds
electronic coin public
telephone
equipment-to-equipment
interface
Exchange Feeder Route
Analysis Program
Electronics Industries
Association
Economic Impact Study
System
Enhanced 911 (Emergency
Service)

EPL
EPROM
EPSCS
ETS
FACS
FCAP
FCC
FCG
FDM
FDX
FEPS
FIFO
FIP
5XB COER

FM
FNPA
FSK
FTS
FX
GHz
GND
GOS
GRPMOD
GSAT

HCSDS
HCTDS
HDX
HNPA
HSSDS
HU
Hz
IC
ICAN
ICC
IDDD

825

Electronic Switching System
Programming Language
erasable programmable
read-only memory
Enhanced Private Switched
Communication Service
electronic tandem switching
electronic translator system
Facilities Assignment and
Control System
Facility Capacity
Federal Communications
Commission
false cross or ground
frequency-division
multiplex
full duplex
Facility and Equipment
Planning System
first in, first out
facility interface processor
No.5 Crossbar Central
Office Equipment
Reports (System)
frequency modulation
foreign numbering plan area
frequency-shift keying
Federal Telecommunications
Service
fO,reign exchange
gigahertz
ground
grade of service
group modulator
General Telephone and
Electronics Satellite
Corporation
High-Capacity Satellite
Digital Service
High-Capacity Terrestrial
Digital Service
half duplex
home numbering plan area
High-Speed Switched
Digital Service
high usage
hertz
interexchange carrier
Individual Circuit Analysis
Interstate Commerce
Commission
international direct distance
dialing

826

IDF
IEEE
IF
IFRPS
1M
INTELSAT

INWATS
lOP
IPLAN
ipm
IRC
IROR
ISDN
ISO
ITU
JMX
K
kbps
kft
kHz
KP
KSR
KTS
LAC
LAMA
LATA
LBO
LBS
LCIE
LED
LF
LIFO
LLN
LMMS
LMOS
LOCAP
LRAP

Acronyms and Abbreviations

intermediate distributing
frame
Institute of Electrical and
Electronics Engineers
intermediate frequency
Intercity Facility Relief
Planning System
interface module
International Telecommunications Satellite
Consortium
inward Wide Area Telecommunications Services
input / output processor
Integrated Planning and
Analysis
interruptions per minute
international record carrier
internal rate of return
integrated services digital
network
International Organization
for Standardization
International Telecommunication Union
jumbogroup multiplex
kilobit
kilo bits per second
kilofeet
kilohertz
key pulse
keyboard send-receive
key telephone system
Loop Assignment Center
local automatic message
accounting
local access and transport
area
line buildout
Load Balance System
lightguide cable
interconnection equipment
light-emitting diode
line finder
last in, first out
line link network
Local Message Metering
System
Loop Maintenance
Operations System
low capacitance
Long Route Analysis
Program

LRS
LRSS
LSI
LSRP
LTF
MAS
MATFAP

Mbps
MCC
MELD

MET
MF
MFJ
MFT
MGT
MHz
MLT
MMGT
MMX
modem
MSC
MTSO
muldem
MUX
NAC
NCP
NCTE
NDCC
NEBS
NI
NMC
NOC
NOCS
NORGEN
NOTIS

line repeater station
Long Range Switching
Studies
large-scale integration
Local Switching Replacement Planning (System)
lightwave terminating
frame
Mass Announcement System
Metropolitan Area
Transmission Facility
Analysis Program
megabits per second
master control center
Mechanized Engineering
and Layout for
Distributing Frames
multibutton electronic
telephone
m ultifrequency
Modification of Final
Judgment
metallic facility terminal
mastergroup translator
megahertz
mechanical loop testing
multimastergroup translator
mastergroup multiplex
modulator-demodulator
Media Stimulated Calling
mobile telecommunications
switching office
multiplexer-demultiplexer
multiplex
Network Administration
Center
network control point
network channelterminating equipment
Network Data Collection
Center
New Equipment-Building
System
network interface
Network Management
Center
Network Operations Center
Network Operations Center
System
Network Operations Report
Generator
Network Operations Trouble
Information System

Acronyms and Abbreviations

NPA
NPV
NSC
NSCS
NSPMP
NTSC
OCC
OCE
OCM
OCU
OFNPS
ONI
OPS
OS
OSC
OSI
OUTWATS
PABX
PACE
PAM
P/AR
PAS
PBC
PBX
PCM
PCO
PE
PIA
PIC
PICS
PICS/DCPR

PIN
PNPN
PPS
PREMIS

numbering plan area
net present value
Network Service Center
Network Service Center
System
Network Switching Performance Measurement Plan
National Television
Standards Committee
other common carrier
other common carrier
channel equipment
office carrier module
office channel unit
Outstate Facility Network
Planning System
operator number
identification
off-premises station
outstate
oscillator
Open Systems
In terconnection
outward Wide Area Telecommunications Service
private automatic branch
exchange
Program for Arrangement
of Cables and Equipment
pulse-amplitude modulation
peak-to-average ratio
Public Announcement
Service
peripheral bus computer
private branch exchange
pulse-code modulation
peg count and overflow
peripheral equipment
Plug-In Administrator
plastic-insulated cable
Plug-In Inventory Control
System
Plug-in Inventory Control
System / Detailed Con tin uing Property Records
personal identification
number
positive-negative-positivenegative (devices)
Product Performance
Surveys
Premises Information
System

827

PREMIS/LAC Premises Information
System/Loop Assignment Center
PSAP
public safety answering
point
PSK
phase-shift keying
PSTN
public switched telephone
network
PUC
public utilities commission
QAM
quadrature-amplitude
modulation
QMP
Quality Measurement Plan
QSS
Quality Surveillance System
RAM
random-access memory
R&SE
Research and Systems
Engineering
RAO
Revenue Accounting Office
RASC
Residence Account Service
Center
RBOC
regional Bell operating
company
RCC
radio common carrier
RCM
remote carrier module
RCVR
receiver
ROES
Remote Data Entry System
RDS
Radio Digital System
Radio Digital Terminal
ROT
remote equipment module
REM
radio frequency
RF
RMAS
Remote Memory
Administration System
root-mean-square
rms
receive only
RO
ROM
read-only memory
rn
reference noise
RSM
remote switching module
RSS
remote switching system
remote trunk arrangement
RTA
RTM
remote test module
read/write memory
R/WM
Street Address Guide
SAG
Swit<;hed Access Remote
SARTS
Test System
Switching Control Center
SCC
Switching Control Center
SCCS
System
selector control unit
SCU
Specific Development and
SD&D
Design
selective dynamic overload
SDOC
controls
selector
SEL
single frequency
SF

828

SI
SMAS

Acronyms and Abbreviations

status indicator
Switched Maintenance
Access System
SMDF
subscriber main distributing
frame
SMSA
standard metropolitan
statistical area
SONDS
Small Office Network Data
System
SP
signal processor
SPC
stored-program control
SPCS
Stored-Program Control
System
SPCS/COER Stored-Program Control
System/Central Office
Equipment Reports
SSAS
Station Signaling and
Announcement Subsystem
SSB
single sideband
SSBAM
single-sideband amplitude
modulation
SSTTSS
space-space-time-time-spacespace (network)
ST
start
STP
signal transfer point
STS
space-time-space (network)
SXS
step-by-step
TASC
Telecommunications Alarm
Surveillance and Control
(System)
TASI
Time Assignment Speech
Interpolation (System)
TCAS
T-Carrier Administration
System
TCSP
Tandem Cross Section
Program
TDAS
Traffic Data Administration
System
TDM
time-division multiplex
TE
terminal equipment
transverse electric
TELSAM
Telephone Service Attitude
Measurement
TERM
terminal
TFLAP
T-Carrier Fault-Locating
Applications Program
TFS
Trunk Forecasting System
3ACC
3A central control

TIRKS
TLN
TLP
TM
TMDF
TMS
TNDS
TNOP
TI/OS
TPMP

TRMTR
TSI
TSORT
TSP
TSPS
TSPS/ACTS

TSS
TSST
TST
TSTS
TTY
TWT
UCD
USITA
USP
VF
VHF
VIU
VNL
VNLF
VSB
WATS
XB
XBT

Trunks Integrated Records
Keeping System
trunk link network
transmission level point
transverse magnetic
trunk main distributing
frame
time-multiplexed switch
Total Network Data System
Total Network Operations
Plan
TI (carrier) outstate
Total Network Data System
Performance Measurement
Plan
transmitter
time-slot interchange
Transmission System
Optimum Relief Tool
traffic service position
Traffic Service Position
System
Traffice Service Position
System/ Automated Coin
Toll Servic~
Trunk Servicing System
time-space-space-time
(network)
time-space-time (network)
time-space-time-space
(network)
teletypewri ter
traveling-wave tube
Uniform Call Distribution
United States Independent
Telephone Association
Universal Sampling Plan
voice frequency
very high frequency
voiceband interface unit
Via Net Loss (Plan)
via net loss factor
vestigial sideband
modulation
Wide Area Telecommunications Services
crossbar
crossbar tandem

Index

"Above-890" Ruling, 689, 693
Access code, 94n
Access line(s), 39
engineering of, 188, 190
Access unit, 493
Accounting measures, in project evaluation,
719,726
ACCUNET Packet Service, 71 n
see also Basic Packet-Switching Service
ACCUNET Reserved 1.5 Service, 73n
see also High-Speed Switched Digital
Service
ACCUNET Tl.5 Service, 73n
see also High-Capacity Terrestrial Digital
Service
Acoustic signal, 97, 194
see also Speech signal(s)
Action point (ACP), 509, 510, 512, 513
Adaptive transversal equalizer, 478-79
Address
assignment of, by PREMIS, 620
data base, 617
definition of, 115
history and evolution, 114-19
signaling, 268, 269, 275-76, 290
Addressing, definition of, 85
see also Signaling
Address input devices, 117
Administration, 572
customer-service, 577-79
billing, 577.-78
switching systems, 578-79
network, 592-96
data, 592-93
equipment utilization, 594
network management, 595-96
office status evaluation, 594
operator-services, 593
problem analysis, 594
transition management, 595

Advance Calling, human factors in design
of,739-40
Advanced Mobile Phone Service, 521-24
cellular concept, 21, 521-23
regulation, 524
system operation, 523-24
system plan, diagram, 522
see also AUTOPLEX cellular radio system
Advanced Mobile Phone Service, Inc., 5, 10
Advanced Research Projects Agency,
private data network of, 91
Advertising organizations, 713
Air Force, U.S., 19
Air / ground service, 74
Airline reservations
automatic call distributor, 503, 55-56
Basic Packet-Switching Service, 72
DA TAPHONE digital service, 70
Alarm service bureaus, use of DATAPHONE
Select-a-station service, 538-40
Alarm systems, in telephone equipment
building, 541, 567
Alerting, 85,265,274,285
electronic telephone, 469
ALGOL,433
All-number calling, 118n
Allocation area(s), in local facilities
network, 122-23 .
Allowance numbers, in quality assurance,
756
Alternate routing, 92, 110, 166, 169,259,630
automatic, 106, 400
busy hour, 173
cancellation of, 181
cost function for, graph, 171
during trunk congestion, 177
economics of, 169-72
in ESPCS, 69
in No. 4A Crossbar System, 400
in private switched networks, 113

829

Index

830

Alternating current (ac), 211, 541, 543-49
power system, telephone equipment
building, 561
American Bell Inc., 5, 10,23, 26, 27, 34,

488n,689,703
cellular radio equipment, 524
American Bell Telephone Company, 689,
700
American Information Technologies
Corporation (Ameritech), 36
American Speaking Telephone Company,
689,700
American Standard Code for Information
Interexchange (ASCII), 318, 483
American Telephone and Telegraph
Company, see AT&T
American Transtech, 34
Amplifier, in COM KEY telephone system,
490
see also Handicapped, aids for the;
Repeater(s)
Amplitude I frequency distortion, see
Distortion
Amplitude modulation (AM), see
Modulation
AMPS, 521n
Analog carrier transmission, 340-69
circuit-miles by type, tables, 395, 396
4-wire interface, 295-99
frequency-division multiplex terminals,
362-69
long-haul interoffice, 347-62
analog microwave radio, 352-56
L-carrier, 347-50
satellite, 356-62
undersea coaxial cable, 350-51
loop, 340-41
metropolitan interoffice, 341, 343
outstate interoffice, 343-47
N-carrier,343-46
short-haul microwave radio, 345-47
Analog channel(s)
carrier transmission, 200, 337
data transmission impairments, 237
definition of, 198
digital data signal transmission, 46, 189n,
197,217-18,272
4-wire carrier interface, 295
impairments, effect of, 223
standard, Bell System, 219
Analog data transmission, 529
Analog facility, 128, 198
data communications, 43-44
voice-frequency facility, 199
see also Channel(s); Transmission facilities
Analog microwave radio, 352-56
Analog radio, circuit-miles of, 395
Analog radio system(s), compatibility with
digital systems, 383

Analog signal(s), 97,193,194, 198
data sets, 476
impairments of, 224-36
modulation, 211
multiplexing, 299
noise, 216
pulse-code modulation, 214, 216
see also Program signal; Speech signal(s);
Video signal
Analog-to-digital conversion, 97, 133
Analog transmission
electronic switching network, 413
4ESS switch, 428-29
impairments and objectives, 223, 224
multiplexing, 221
see also Analog channel(s); Analog
facility; Analog signal(s);
Voiceband channel
Anderson, Philip, 22
ANIK domestic satellites, 357
Announcement machine, for Automatic
Intercept System, 444
Announcement system, see Mass
Announcement System
Answer delay, 270
Answer entry, LAMA, 449
Answering service, telephone, see
Telephone answering systems
Answer supervision, 97
Mass Announcement System, 515
Antenna(s), 564
long-haul microwave radio systems, 35255
diagram, 354
satellite communications systems, 358,
359,361-62
satellite microwave radio, 209-10
short-haul microwave radio systems, 346,
347
terrestrial microwave radio, 207-8
picture, 208
Antisidetone circuit, 468
Antisidetone network, diagram, 468
Antitrust suit, see Justice, Department of,
U.S.
Apollo, Program, 19
Applied research (central services
organization), 37
Area code, 115, 116, 117, 118, 512n
principal city concept, 120
table, 119
see also Numbering plan area
AR6A long-haul analog microwave radio
system, 355,356
Artificial larynx, 49, 694n
picture, 50
ASCII (American Standard Code for
Information Interexchange), 318,
483

831

Index

Asynchronous transmission, 477
data sets, 480
data terminals, 482, 484
AT&T
postdivestiture
American Transtech, 34
AT&T Bell Laboratories, 34
AT&T Communications, 32
AT&T Information Systems, 34
AT&T International, 10,34
AT&T Technologies, 32
AT&T Western Electric, 33
corporate structure, 32-34
diagram, 33
predivestiture
Bell Laboratories, 20-27
Research and Systems Engineering,
funding, 23
corporate functions, 6
corporate history, 691, 694, 700
table, 689
corporate structure, 7-10
diagram, 8
income statement, 726
mature product management, 708
non-Bell System relations, 27-28
operating companies, 11-14
operations planning, role in, 646, 647,
648,656,659,662
Quality Surveillance System, 743
resources, 28-29
service, volume of, 28-29
Western Electric, 14-20
see also Bell Laboratories; Bell System;
Legislation; Litigation;
Modification of Final Judgment;
Regulation; Telephone
company(ies); Western Electric
Company, Incorporated
AT&T Bell Laboratories, 33, 34
see also Bell Laboratories
AT&T Communications, 32-33
AT&T Information Systems, 33, 34
see also American Bell Inc.
AT&T International, 10,33,34
AT&T Technologies, 32, 33
AT&T Western Electric, 33
see also Western Electric Company,
Incorporated
Attempts per circuit per hour (ACH),
179
Attenuation, 223
coaxial cable, graph, 348
definition of, 201n
lightguide cable, 207
open-wire lines, 201
rain, 209, 210, 360
satellite, 210
terrestrial microwave radio, 207

VOice-frequency transmission, 333-34
loaded cable, graph, 334
nonloaded cable, graph, 333
waveguides, 204
Attenuation-versus-frequency characteristic,
of E6 repeater, 339
A-type channel bank, 296, 364-65
diagram, 364
tables, 394, 395
Audible ring, 97, 98,267,268,412,419
Audio program network, 92
Audio system, PICTUREPHONE meeting
service, 528
Audio tape recorder, with GEMINI 100
Electronic Blackboard System, 485
Audio tone pulses, in mobile telephone
systems signaling, 519-20
Audit, quality assurance, 743,745-56
appraising, 746-49
checking, 746
defects, table, 748
nonconformance, 755-56
rating and reporting, 746, 752-55
sampling, 746, 751-52
standards, establishing, 746, 749-51
Audit point, 744, 745
Authorization code entry
electronic tandem switching, 70
EPSCS,69
Automated Calling Card Service, 61, 64, 78,
122,293,443,472,475,655
human factors in design of, 740
stored-program control network, 511-12
Automated Coin Toll Service (ACTS), 77,
443,471-72,475
human factors in design of, 740
Automated Trouble Reporting System
(ATRS),597n

Automatic alternate routing, 106,400
Automatic back-up line, 600
Automatic call distributor (ACD), 53, 55-56,
407,503-4
common-control systems, 260
in ESS switching systems, 411
Automatic calling and answering, 44
Automatic calling units, 44, 45
Automatic dialing, 47-48, 49, 69, 470
Automatic dial operation, in mobile
telephone systems, 519
Automatic Electric Company, 398, 490n,
494
Automatic identified outward dialing
(AIOD),501
dial PBX, 494
Automatic Intercept Center (AIC), 443-45
Automatic Intercept System (AIS), 111,44345,590,637,655
diagram, 444
SCCS,637

832

Automatic line buildout (ALBO), in digital
carrier transmission, 378, 381
see also Line buildout network(s)
Automatic message accounting (AMA), 94,
99,175,421,445-47,448-57,678
recorded volume projection, graph, 446
remote recording, 453-56
store and forward, 456-57
see also Billing; Centralized automatic
message accounting; Local
automatic message accounting
Automatic Message Accounting Recording
Center (AMARC), 436n, 453-54,
455,456,458
No.1 AMARC, 453, 455
No. lA AMARC, 454, 456
Automatic message accounting transmitter
(AMAT),457
Automatic nondial trunk, 285
Automatic number identification (ANI),
451-52
stored-program control network, 510
Automatic overload control, 163
Automatic protection switching (APS), 240
digital carrier transmission, 378, 379, 380,
381,391
Automatic route selection, 54, 55
Automatic switching, 398
Automatic System for Performance
Evaluation of the Network
(ASPEN), 668-69, 683
diagram, 669
Automatic voice network (AUTOVON), 91,
399
AUTOPLEX cellular radio system, 522
staged growth of, diagram, 523
see also Mobile telephone service(s)
Auxiliary Data System, 427, 428
Auxiliary sets, data, 45
Availability
data circuit objective, 237
performance objectives, 676, 678
signaling, 294
Average busy season busy hour (ABSBH),
153, 166, 167
2ESS switch call capacity, 420
Average call arrival rate, 148, 156
data traffic, 184
Average call holding time, 148, 156
data traffic, 184, 187
Average transaction arrival rate, 186, 187
Awards, see Prizes and awards
Back-up line, carrier system, 600
Balanced pair, 296n
Bandlimited signal
pulse-code modulation, 214
speech,194

Index

Band number, signal transfer point
architecture, 293
Bandpass filter
analog carrier transmission, 341, 364, 365
SSBAM transmission, 213
Bandwidth
amplitude/frequency distortion, 234
analog channels, 198,218
channel capacity, 189n
data sets, 44
definition of, 194
DSBAM transmission, 213
lightguide cable, 207
linear distortion, prevention of, 238
long-haul interoffice facilities, 332
program network, 92
program signals, 194
pulse-code modulation, 217
space-division networks, 245
speech signals, 194
SSBAM transmission, 213
waveguides, 204
Bardeen, John, 22
Baseband channel, definition of, 199
Baseband signal
digital carrier transmission, 373
DSBAM,212
long-haul microwave radio systems, 353,
356
pulse-code modulation, 214, 215, 217
Base period study, in quality assurance,
749
Basic group, frequency-division
multiplexing, 362, 363, 364
Basic 911 (B911) service, see 911 Emergency
Service
Basic Packet-Switching Service (BPSS), 7172,535
characteristics, table, 322
interfaces, 320n, 322
No.1 PSS packet switch, 535-38
Basic service, see Service(s)
Battery, electric, 546, 547, 549
Battery voltage, 272
Bell, Alexander Graham, 687, 700
Bell Administrative Network
Communications System, 660
Bell Atlantic Corporation, 36
BELLBOY personal signaling set paging
service, see Paging
Bellcomm, 19
Bell Journal of Economics, 28
Bell Laboratories, 4, 5, 9, 14, 20-27, 162, 671,
693
administration of central office
equipment, 612
antitrust (1974),703
branch laboratories, 25, 27
table, 27

Index

Business Information Systems Programs,
642
-.
contributions to telecommunications
science and technology, table, 21
corporate structure (January 1983), 23-27
COSMOS, 620
C programming language, 438
economic evaluations of projects, 714,
715,717,727,728,729
funding, 23
human factors, 737-42 .
locations and activities (1982), table, 26
maintenance planning organizations, 597
matching new technology to market
needs,713,730
Modification of Final Judgment, 703
network characterization, 665
Nobel prizes, 20-22
operations planning, 572, 646-48, 662
operations systems development, 23, 64243,658
enhancement, 654
network, 660
PICS/DCPR, 615

postdivestiture, 34
PREMIS, 616
purpose, 22-23
quality assurance, 742
Quality Assurance Center, 25, 743, 745,
747,749,755,757,758,759
"S" Data Analysis System, 669
service evolution, 705
sonar and underwater surveillance
projects, 19-20
switching systems, engineering of, 241
traffic engineering methods, 148n
UNIX operating system, 668
video telephony, 526
Bell Laboratories Executive Summary, 727
graphs, 728
Bell Laboratories Record, 28, 743
Bell Northern Research, Ltd., 433
Bell of Pennsylvania, 12, 36, 646
Bell operating company (BOC), 29, 30, 32,
36,39
see also Telephone company(ies)
Bell PhoneCenters, 79, 488
Bell Sales Division(s) (Western Electric), 14
Bell Sales-East, 16
Bell Sales-West, 16
corporate functions, predivestiture, 16
regions, 16
map, 17
BellSouth Corporation, 36
Bell System
corporate history, 700-701
table, 689
corporate structure, predivestiture (1982),
diagram, 5

833

marketing organizations, 712, 713
relationships with non-Bell System
companies, 27, 28
resources and volume of service (1982),
28-29
table, 29
services, 39-80
structure and activities, predivestiture
(1982), 3-29
vertical integration, 3-5, 695
see also AT&T; Bell Laboratories;
Divestiture; Legislation; Litigation;
Modification of Final Judgment;
Quality assurance; Regulation;
Telephone company(ies); Western
Electric Company, Incorporated
Bell System Carrier Synchronization
Network, 214, 221
Bell System Corporate Plan, 656n
Bell System network, interconnection with,
305-10
Bell System Practice (BSP), 667
Bell System Reference Frequency Standard
(BSRFS), 221
Bell System Technical Journal, The, 28, 667
Bell Telephone Company, 689, 700
Bell Telephone Company of Nevada, 5n
Bell Telephone Laboratories, Incorporated,
see Bell Laboratories
Billed Number Screening, 64
Billing, 56, 110, 111,243,408,439,576-77,
578
bill format, 446
equipment and systems, 445-57, 458
integrity, 678
rate setting, 697-99
Residence Customer Billing Inquiry
Process, 652
diagram, 653
see also Billing system(s); Traffic Service
Position System
Billing data transmitter (BOT), see Billing
System(s)
Billing Order and Support System (BOSS),
653
Billing system(s), 445-57
automatic message accounting, 448-52
bill format, 446-47
evolution, 447-48
remote recording, 452-57
billing data transmitter, 454, 456, 458
call data accumulator, 454, 458
call data transmitter, 456, 458
using store and forward, 456-57
see also Automatic Message Accounting
Recording Center; Centralized
automatic message accounting;
Electronic translator system; Local
automatic message accounting

834

Bill-to-third number calls, 64, 77, 471
Binary code, 216
Binder group, 201-3
Bipolar flip-flop, 419
Bipolar pulse, 419
Bipolar signal, 374-75, 376
diagram, 374
see also Bipolar with six zero substitution
Bipolar violation, digital multiplexer, 392
Bipolar with six zero substitution (B6ZS),
381-82
Bit, 196,218
PCM-TDM synchronization, 222
signaling, 291
time-division multiplexing, 220
Bit error rate
data sets, 479, 482
definition of, 601
grade-of-service rating, 676
see also Error objective
Bit rate, 199,218,369,666
channel capacity, 189n
visual systems, 529
see also Digital signal level
Blocked calls cleared (BCC), 149, 153
Erlang's model, 158-59
see also Erlang's Loss Formula
Blocked calls delayed (BCD), 150, 160
Erlang's model, 156-58
operator-services provisioning, 589
see also Erlang's Delay Formula
Block error rate, 479, 482
Blocking
common control, 259
direct progressive control system, 259
estimates, 182
4ESS switch, 427
performance objectives, 676
probability, 149
space-division switching network, 245,
246
B911 service, see 911 Emergency Service
Book depreciation, 724, 725
Bootstrap loader, 433
Box and whisker plot, in quality assurance,
755
Branch feeder cable, 327
Branch laboratories (Bell Laboratories), 25
Bell Laboratories specific development
and design functions, 25, 27
locations and activities, table, 27
Brattain, Walter, 22
Bridge, telephone answering systems,
diagram, 505
Bridged tap, 329, 338, 339n
Broadband channel, 198,200
data communications, 44, 46, 477, 478
multiplexing, 218, 219
private-line channel, 67
protection channel, 237 n

Index

Broadband serial data sets, 478
Buffer circuit, 410
Buffer storage, definition of, 222
Building Engineering Standards, 564
Bulk bill format, 446-47
Burglar alarms, use of telegraph channels,
91
Burleson, A. S., 692
Bus, time-division, 253,500,502
Business customers
order negotiation and marketing, 574-75
rates, 698
service administration, 578-79
switching services, 52-56
Transaction telephones, 475-76
Business Information Systems Programs
(BISP), 642, 643
Business Marketing organization (AT&T),
corporate functions,
predivestiture, 7, 8
business office services, 79
Business organization(s), corporate
functions, predivestiture
AT&T,7,8
telephone company(ies), 14
Business products, manufacturing of, table,
15
Business service(s), 40, 52-56
EPSCS,113
exchange business services, 58-59
private switched networks, 113
network diagram, 113
see also Customer switching systems
Business Services organization (AT&T),
corporate functions,
predivestiture, 7,8
Busy hour(s), 152,153, 166,629
data networks, 190
electronic switching systems, 414, 422,
426
non coincidence, 173, 184
stored-program control systems, 509
capacity, table, 459
Busy/idle status, 112,289,411,513,519
see also Off-hook; On-hook
Busy season, 152, 166,629,639
Busy season busy hOJ,lr (BSBH), 152-55
Busy signal (busy tone), 94, 97, 268, 282,
402,412
see also Recorder tone
BX.25 interface, 320, 322, 323
C, programming language, 438
Cable, 27
digital carrier system, 372, 376, 379, 381
digital facilities network, 132
distributing frame, 549, 550, 551
diagram, 554
inductive interference, 235

Index

interoffice facilities network, 128
inventory, 608
key telephone system, 490
loaded,333-34,338,372
attenuation, graphs, 333, 334
effect of loading, graph, 335
local facilities network, 124
products, manufacturing of, table, 15
quality assurance, 746
small communications systems, 491
telephone answering system, 505
telephone equipment building, 560
picture, 562
VOice-frequency transmission, 333, 338
see also Coaxial cable; Lightguide cable;
Multipair cable; Paired cable
Cable distribution systems, 565, 567
Cable entrance facility (CEF), 541, 565
diagram, 567
Cable Pathways Plan, 567
Cable-pressure monitoring, 598
Call annoyance bureau, 638
Call arrivals, 155, 156
Call assembly, 451
definition of, 448
Call-by-call simulator, 178
Call counting, 513, 515
Call coverage, HORIZON communications
system, 492-93
Call cutoff rate, 676
defini tion of, 668
Call data accumulator (CDA), see Billing
system(s)
Call data transmitter (CDT), see Billing
system(s)
Call departure rate, 156
CALL DIRECTOR telephone, picture, 466
Call disconnect, 98
LAMA entry, 449
supervision, 99
Call distributor, 4A, 504
Call Forwarding, 57, 501, 713
numbering, 121
with telephone answering systems, 506-7
Call holding time(s), 155, 161, 185-87
Call identity index, LAMA, 448, 449
Calling card service, see Automated Calling
Card Service; Charge-a-Call
Call Management System (CMS), 80/5 CMS,
504
Call passing, 53
Call processing, stored-program control,
262-64
Call processor, 254
Call queuing, 69, 70
see also Traffic theory
Call rate, 185-87, 640
Call-related time, 190
Call routing
local network, 106

835

principal city concept, 120
private network services, 68
CCSA,68
DSDC,64
EPSCS,69
ETS,70
stored-program control network, 509, 510,
513
toll calls, 109
in a typical telephone call, 94
see also Alternate routing
Call store, 262,415-16,421,431, 639, 640,
641
Call store bus, 415, 416
Call tracing, 282
Call transfer, HORIZON communications
system, 492
Call Waiting, 57, 501
Cancel timing indication, 117
Capacitor, 467
Capacity expansion, 136, 138-39
Capital investment, 719, 721
Capital Utilization Criteria (CVCRIT)
program, 144,718
Carbon transmitter, 465, 470
Card dialer, see Automatic dialing
Card translator, 408
Carried load, 159
congestion, 176, 177, 178-79
see also Offered load
Carrier(s)
domestic,10
interexchange, 30, 31, 703
international,10
interstate, 688
see also Common carrier(s)
Carrier circuit(s), tables, 394, 395, 396
Carrier-derived channel, 199-201
coaxial cable, 203
4-wire channel, diagram, 200
Carrier facility terminal, 200
Carrier frequency, 222
DSBAM transmission, 212, 213
multiplexing, 219
SSBAM transmission, 213
Carrier frequency shift, 221, 235
Carrier group alarm, digital channel bank,
387
Carrier inventory, TIRKS, 608
Carrier Signal, modulation of
DSBAM,211-13
frequency-shift keying, 218
PCM,214
SSBAM,213-14
Carrier system(s), 86, 200, 335-37
circuit order control, 607, 608
definition of, 86n
loop applications, 124, 393-94
loss objectives, 225
maintenance, 600-601

836

Index

Carrier system(s) (contd)
metropolitan network, 128
paired cable, 203
relative economics of, graph, 336
signaling, 269, 272, Z74, 278, 284, 289, 291
Carter Electronics, 694
Carterfone Decision, 310, 311, 479, 694, 701
Cash flow, 140
Cash flow measures, in project evaluation,
719-23
cumulative discounted cash flow, 721-23
diagram, 722
internal rate of return, 723
posttax cash flow, 720-21
pretax funds, 719-20
Cathode-ray tube (CRT), 318, 319, 610
data terminals, 45-46, 480
PREMIS, 618
SCCS, 635, 636
TIRKS, 610
CCITT, see Comite Consultatif International
Telegraphique et Telephonique
CCITT Signaling System No.6, 293
CCITT Signaling System No.7, 294
Cellular radio, see Advanced Mobile Phone
Service
Centralized automatic message accounting
(CAMA), 62, 400, 407, 408, 447,
451-52
CAMA-C, 451, 458
diagram, 452
Mass Announcement System, 515
TSPS, 439
Centralized automatic message accountingoperator number identification
(CAMA-ONI), 111, 112
Centralized automatic message accountingoperator number identification
(CAMA-ONI) operators, 62, 112
Centralized Automatic Reporting on Trunks
(CAROT), 600n, 643, 650, 654, 660,
713
Centralized Automatic Trouble Locating
and Analysis System (CATLAS),
638
Centralized Message Data System (CMOS),
175
Centralized Results System (CRS), 680
Centralized System for Analysis and
Reporting (CSAR), 625, 631
Central office, 87,104,211,327,329,549
battery, 465
noise, 230
call origination, 156
coin service, 275,472
compatibility requirements in design of
station equipment, 487
customer switching services, 52
customer switching systems, 489
diagram, 489

dial-tone delay, 180
DMS-I0 switching system, 433
equipment reporting process, TNOS, 62830
exchange business services, 58, 59
functioning of, in typical telephone call,
92-99,465
inventory and records of equipment, 612
in the metropolitan area, 329
in the outstate area, 330
remote switching system, 87
routing between offices, 169-72
service provisioning, 576
signaling, 85, 266, 271,272,274,468
diagram, 267
TOUCH-TONE service, accommodation
of, 276, 736-37
Central office building, 87n, 104, 105
interconnections, 310
span lines, 376
see also Telephone equipment building
Central office code, 87, 92,94,96,117, 119
definition of, 115
principal city concept, 120
Central Office Equipment Engineering
System (COEES), 638, 639-41
information flow, diagram, 639
Central processing unit (CPU)
DMS-I0 switching system, 431, 432, 433
lESS switching system No.1 CPU, 415
Central pulse distributor (CPO), 418-19
Central services organization (CSO), 30, 33,
34,35,37
functions, technical support, 37
Central stock, 612
Central switching, 82, 84
diagram, 83
Centrex, 40, 55, 59, 161
No.5 Crossbar System, 399, 407
private network services, 67-68, 69, 70
Ceramics, abnormalities, sensing of, 15
Chalk switching, 485
Channel(s), 44, 173
analog carrier transmission, 340-41
capacity,189
carrier-derived,199-201
classifications of, 86
concept, 198-99
interoffice facilities network, 128
L-carrier system, 348
limitations, 223
multiplexing, 218, 219,220, 221
overload, 236
private-line data network, 91
signaling, 266, 271
switching network, 243
voice-frequency, 199-201
see also Circuit(s); Line(s); Trunk(s)
Channelbank,255,295
analog carrier system, 362

Index

analog FDM terminal, 364-65
digital carrier systems, 377, 380, 381, 38789
digital carrier trunk, 302, 303, 305
Digital Data System, 533
DSX-l digital system cross-connect
interface, 299, 300
4ESS switch, 429
4-wire analog carrier interface, 295, 296,
299
diagram, 297 .
multiplexing, 221
numbers of, tables, 394, 395
Channel capacity, 189
Channel separation filter, 290
Channel services unit (CSU),315, 531
Channel unit
digital channel bank, 387-89
digital signaling, 291
Characterization, network, 665-69
Charge-a-Call
service, 9, 471
telephone, 77, 78, 471
picture, 78
Check verification, 49
Chesapeake and Potomac Telephone
Companies, 5n, 11, 12, 36
Cincinnati Bell, Inc., 5n, 12, 29
Circuit(s), 242, 747
antisidetone network, 468
Automatic Intercept Center, 444
busy, 149
carrier circuit miles by type, 394, 395, 396
classifications of, 86, 419
DMS-IO switching system, 431
ICAN, 629-30
long-haul interoffice plant, 332
metropolitan interoffice plant, 329, 330,
339
1/IAESS switches, 419
outstate interoffice plant, 330
provisioning, TIRKS, 605-9, 610, 615
diagrams, 606, 608
order control, 613
private-line data networks, 91
private-line voice networks, 89, 90, 91
range extension, 340
repair, 660
signaling, 271
transmission level points, 296
voice-frequency, 340
see also Channel(s); Special-services
circuit(s); Trunks Integrated
Records Keeping System
Circuit Administration Center, 606, 630,
631
Circuit design, 607
Bell Laboratories Specific Development
and Design, 23
interfaces, 270

837

Circuit order control (COC) system, TIRKS,
607
Circuit Provision Center (CPC), 606, 607,
610,651
Circuit routing, 136, 138
Circuit segregation, 642
Circuit-switched digital capability (CSDC),
61,65-66,530,666
characteristics, table, 322
Circuit switching, 244
Circuit-switching network(s), 244-45
data traffic, 189
Circuit (Cl) system, 607, 608, 609
Circular waveguide, 204, 352
picture, 208
Class 5 office, see End office
Class of switching office, 107-9
Clayton Antitrust Act, 688, 689, 691
Clock interrupt, electronic switching
systems, 411
Cluster busy-hour technique, 173
diagram, 174
Cluster controller, data terminals, 480, 483,
484
C-message weighting, 230, 672
Coastal-harbor [radiotelephone] service, 7475
Coaxial cable, 6, 201, 203, 429, 642
attenuation, graph, 348
diagram, 204
digital carrier systems, 377-78
digital facilities network, 133
interoffice facilities network, 132
see also L-carrier analog carrier system(s);
Undersea cable
Code blocking, 181
Codec
DMS-I0 switching system, 431
PICTUREPHONE meeting service, 527
CODE-COM telephone set, 51
picture, 51
Coded ringing, 285
Code partitioning, 119
table, 119
Code select ringing circuit, 285
Coin disposal test(s), 474
Coin-first
service, 372, 471-75
telephone, 77
Coin hopper, 474
Coinless public telephone, see Charge-a-call
Coin relay, 474
Coin telephone, see Public telephone(s)
Collect calls, 64
from public telephones, 77
Collector, automatic message accounting,457
Color television, video signals, 195-96
Combined main distributing frame (CMDF),
551,555
diagram, 554

838

Comite Consultatif International des Radiocommunications (CCIR), 28, 358
Comite Consultatif International
Telegraphique et Telephonique
(CCITT), 28, 120, 188n, 358, 387
interfaces and protocols, 319-21, 536, 538
CCITT Signaling System No.6, 293
CCITT Signaling System No.7, 294
COM KEY key telephone system, 490
diagram, 491
Command guidance system, 19
Common carrier(s), 40, 211, 689, 693
Modification of Final Judgment, 703
state regulations, 692
see also Carrier(s); Other common carrier
Common-carrier band
long-haul microwave radio systems, 355
short-haul microwave radio systems, 345
Common-channel interoffice signaling
(CCIS), 63, 272, 277, 280-84, 286,
292-94,655
network,507-9,513,596,655
No. 4A Electronic Translator System, 409
operating configuration, diagram, 292
switched special-services signaling, 285
time-division switching, 426
Common control, 246, 259-61, 398
crossbar PBX, 495-96
LAMA, 448, 449
marker-type, diagram, 261
No.5 Crossbar System, 399, 406, 407
step-by-step systems, 400
TNOS,629
TOUCH-TONE service, 736
wired-logic electronic PBXs, 499
see also Crossbar switching system(s);
Progressive control; Storedprogram control
Common-control switching arrangement
(CCSA), 68-69
Common-language equipment
identification (CLEI) code, 614
Common-language location identification
(CLL!) code, 614
Common systems, 541-67
distributing frames, 549-59
equipment building systems, 559-65
miscellaneous systems, 565-67
alarm, 567
cable distribution, 565-67
cable entrance facility, 565, 567
power, 541-49
provisioning, 590-92
diagram, 591
role of, diagram, 542
Common Systems Main Interconnection
Frame System (COSMIC), 553-55,
558
COSMIC II, 553-55
pictures, 555, 558

Index

Common update/equipment (CU /EQ)
system, 625, 628, 629
Common update/trunking (CU /TK) system,
625,630-31
Comm-Stor II communications storage unit,
484
Communications management, 52, 53, 54,
55,59
Communications module, 5ESS switch, 435,
436
Communications satellite(s), 356-62
frequency bands, table, 358
orbit locations, diagram, 357
transmission medium, 209-10
Communications Satellite Corporation
(COMSAT),28
Communications systems, small, 53, 49193
Community dial office (COO), 130, 379
DMS-10 system, 430
number of, in Bell System, 461
power system, 541, 547
3ESS switch, 422
Compandor, N-carrier terminals, 345
Compatibility requirements, station
equipment, 485, 487
Compelled signaling mode, 286, 287
Competition, 687, 689, 699-703, 708
interstate services, 699
product life cycle, 711
Computer(s), 193, 196, 542, 549
number in use in Bell System (1981),
660
3B20 computer, 434, 513
3B200 computer, 536,537
WE 8000 microcomputer, 472
WE 4000 microprocessor, 305, 423
see also Operations system(s)
Computer Inquiry I, 689,702
Computer Inquiry II, 10, 32, 488n, 574n,
689,702-3
Bell System marketing structure, effect
on, 712
retail sales of terminal equipment, 79
Computerization
ASPEN,668
automatic call distributor, 504
billing, 451, 453-54, 455
BISP, 642
carrier system maintenance, 600-601
coin collecting from public telephones,
583
DIMENSION PBX, 499,500, 501, 502
electronic coin public telephone, 472
engineering complaints, 759
EPLANS Computer Program Service, 64243
fault-locating, for T1 (carrier) outstate,
380
HORIZON communications system, 492

Index

maintenance
customer switching systems, 582
network, 597
special-services circuits, 601
101E55 switch, 497
operations planning, 646-47
operator services, 590
postdialing delay, 280
service
administration, of customer switching
systems, 579
provisioning, 575
telephone numbers, inventory of, 620
see also Operations system(s)
Computer simulation
performance models, creation of, 671
see also Simulation techniques
Computer Subsystem (CSS), 633, 636, 637
diagram, 634
Computer System for Mainframe
Operations (COSMOS), 559, 620
Computer Technologies and Military
Systems area (Bell Laboratories),
24-25
Computer-to-computer data transfer, 66,
265,269,464,529
CCITT Signaling System No.6, 293
COMSAT, see Communications Satellite
Corporation
COMSTAR domestic satellites, 357, 360
Concentrated range extension with gain
(CREG), VF loop transmission,
338
Concen tration
digital carrier system(s), 373
local switching system(s), 397
space-division switching network(s), 24548
diagrams, 247, 248, 250
time-division switching network(s), 258,
259
Concentration network, 246
Concentrator, 164,259,373,438,505,506
see also Concentration
Concentrator technique in pair-gain
systems, 124
Conceptual operations system, 651
Concurrence, 697
Conduit, 86, 128, 327
C1, see Circuit (C1) system
Conference call, COM KEY key telephone
system, 490
see also Teleconferencing
Congestion
analysis of, 177-79
controls, 180
table, 181
detection, 179-80
graph, 178
switching system, 151, 153, 177

839

trunk, 177
see also Network management
Congestion theory, see Traffic theory
Connecting arrangement service for
network protection, 310, 311
Connection(s), 109, 179n, 198
basic service, 39
service evaluation, 664
as a distributing frame function, 550
in network, 397
signaling, in the establishment of, 265,
266-68
switching, 82, 241, 243, 259
Connections per circuit per hour (CCH),
179
Connector, 259, 260, 406
in step-by-step systems, 401, 402
Consent Decree (1956),689,694-95,702,
703
Console, attendant
crossbar PBX, 495
diagram, 496
DIMENSION PBX, 501
telephone answering system, 506, 507
Continuous presence, 529
Continuous Signaling, 286, 287
Contribution, in project evaluation, 723,
725-26
Control equipment, functioning of, in
typical telephone call, 94
Controlled environment vault (CEV), 384
Control mechanisms, switching, 242, 25864
common, 259-61
diagram, 261
direct progressive, 258-59
stored-program, 261-64
diagram, 262
see also Common control; Progressive
control; Stored-program control
Control of service and performance, 678-79
Control signaling, Digital Data System
interface, 315
Control terminal
BELLBOY radio paging set, 525
land-based radio system, 519
Conversion, analog-to-digital / digital-toanalog, 46, 97, 133,215,216
Converter
analog-to-digital/ digital-to-analog, 13233,251,258,413,429
dc-to-dc, 544, 545, 548
Coordinate network, 248-50,260,398
diagram, 252
electronic switching system, 418
Coordinate switch(es)
definition of, 248
diagrams, 249, 251, 252
rotary switch(es), equivalent, diagram,
250

840

Cord switchboard
dial PBX, 494
diagram, 495
manual PBX, 493
diagram, 494
telephone answering system, 505
CORNET,91
Corporate Engineering Division (Western
Electric), corporate functions,
predivestiture, 14-15
Corporate Headquarters (AT&T), 6
corporate functions, postdivestiture, 32
Corrective maintenance, 579, 597
diagram, 580
Costs, see Economics
Country code, 120-21
Credit authorization, 49
Transaction telephone, 475, 476
Credit status of customer, determination of,
620
Critical timing, dialing procedure, 117
Crossbar private branch exchange (crossbar
PBX), 494-98
diagram, 496
fields of use, graph, 497
756 PBX, 496
757 PBX, 496
770 PBX, 496
Crossbar switch, 244, 403, 406
coordinate switch, 248
diagram, 404
marker, 260
Crossbar switching system(s)
automatic call distributor, 504
billing, 451
billing data transmitter, 454
COEES, 639
load balancing, 164
No.1, 399, 402-6, 461
billing, 447
diagram, 405
No. 4,400
No. 4A, 400, 408-9, 462, 463
billing, 451
data collection, 627
diagram, 409
No.5, 399, 406-8, 461, 462, 463, 611
billing, 448, 451, 455, 456
diagram, 407
5XB COER, 625, 628, 629
numbers of, in Bell System, tables, 461,
463
operator systems, 438
progressive control, 248
tandem switching, 400, 438
traffic measurements, 183
Cross-connection, distributing frame, 550,
553
Crosspoint, 403, 437
switch,418

Index

Cross section (number of circuits)
digital carrier system, 381
long-haul network, 331
metropolitan area, 330
outstate area, 330
Crosstalk, 232-33, 488
carrier systems, 343
definition of, 232
digital carrier systems, 376, 377
index, 233,234
network interface, 315
network protection, 313
objectives, table, 234
overload, 236
paired cable, 202, 203
VOice-frequency transmission, 335,
339
Cumulative discounted cash flow (CDCF),
719,721-23
diagram, 722
Current planning
circuit provisioning, 606
diagram, 607
Facility and Equipment Planning System,
609
diagram, 609
network provisioning, 585-86, 588
Current taxes, 719,725
Custom Calling Services, 243, 420
Advanced Mobile Phone Service, 523
numbering, 121
Custom Calling Services I, 57-58
Customer access unit, 493
Customer Administration and Control
Center (electronic tandem
switching),70
Customer attitudes, 678
TELSAM, 680-8l
Customer Changeable Speed Calling, 58
Customer-detected trouble, 581-82, 598
Customer-line (station-loop) signaling, 265,
269,272-76,288,289,290
Customer Name and Address Bureau
(CNAB),653
Customer Network Control Center, EPSCS,
69,70
Customer opinion model(s), 671-73
Customer-premises equipment (CPE), 30,
32,84n
engineering, 576
FCC Computer Inquiry II ruling, 574n
installation, 576
power supply, 549
private-switched networks, 113
service administration, 579
Western Electric, 14
Customer-provided equipment
FCC interconnection rulings, 693-94
interconnection, 305-16, 701-2
Customer Record Center (CRC), 653

\

Index

Customer Records Information System
(CRIS),653
Customer-related operations, 573-85
PREMIS,616
Customer services
equipment, station, 465-89
data sets, 476-80
data terminals, 480-84
graphics terminals, 484-85
telephone sets, 465-76
human factors in design of, 485-88, 737,
739-40
switching systems, functions for, 243
systems, 489-540
data, 529-40
PSTN,507-16
switching, 489-507
automatic call distributors, 503-4
key telephone, 490-91
PBXs, 493-503
small communications, 491-93
telephone answering, 504-7
visual, 526-29
see also DA TAPHONE Select-a-station
service; Data set(s); Data
terminal(s); Digital Data System;
Graphics terminal(s); HORIZON
communications system; Mass
Announcement System; Mobile
telephone service(s); Packetswitching; Private branch
exchange; Stored-program control
network; Teleconferencing;
Telephone answering systems;
Telephone set(s)
Customer services operations, see Planning,
Bell System operations
Customer support services, 78-80
Customer switching services, 52-56, 243
Customer switching systems, 489-507
administration of, 578-79
central office, relationship with, diagram,
489
remote testing, 582
see also Administration; Maintenance;
Provisioning
Customized Call Routing, 63-64
Cutover
billing data transmitter, 413
call data accumulator, 455
call data transmitter, 456
definition of, 658n
5 ESS switch, 434
LAMA-C,455
No.5 Electronic Translator System, 455
lESS switch, 413
TSPS,443
Cutover committee, 658
Cut through (connected), Mass
Announcement System, 513, 515

841

Daily circuit provisioning, 606
diagram, 608
Data accuracy, on data networks, 188, 189
Data administration, 592-93
Data base
Automated Calling Card Service, 511, 655
Expanded 800 Service, 512, 513
operations systems, 603, 661
PREMIS, 616, 617, 619, 620,621
diagram, 618
signaling, 269
stored-program control, 508
Data Base Administration System, 445
Data-base development, mechanization in,
143
Data circuit-terminating equipment (DCE)
interface, 319-20, 322, 323
Data communications, 184
data sets, 43-45, 476-80
table, 481
data terminals, 45-47, 480, 483-84
digital data signals, impairments and
objectives, 236-40
EPSCS,113
interfaces and protocols, 316-23, 536, 537,
538
Transaction telephones, 476
Data communications systems, 529-40
DA TAPHONE Select-a-station service,
538-40
Digital Data System, 530-35
Packet-switching systems, 535-38
Data in the middle (DIM), 534
Data link, 265
CCIS, 272, 283, 284, 286,292,508
DIMENSION PBX, 501
No. lOA Remote Switching System, 423,
425
Data network(s), 61
private-line, 91
see also Data traffic
Data network services, 61, 70-73
Data on mastergroup (DOM), 534
Data over voice (DOV), 534
DATAPHONE data sets, 43, 44, 70, 478
DATAPHONE digital service, 26, 46, 70-71,
530,534
inherent network protection, 313
linear distortion, 238
DATAPHONE Select-a-station service, 70,
71, 530, 538-40
diagram, 539
DATAPHONE II data communications
service, 44, 70
picture, 45
DATAPHONE II data sets, 479, 481, 530
Dataport unit, 533
Data-processing services, 695
and telecommunications services, 702,
703

842

Index

Data products, 43-47
Dataservice unit (DSU)
interfaces, 320
table, 321
Data set(s), 43-45, 113,453,465,476-80,530
application, typical, diagram, 477
Bell System, characteristics, table, 481
design, 479-80
evolution, 478-79
functions, 476-77
interfaces, 320
table, 321
for satellite communications systems, 360
types, 477-78
Data signal, 196-97,336,337,476
impairments and objectives, 223, 236-40
long-haul microwave radio systems, 353
loops, 328
modulation, 217, 218
see also Digital signal
DATASPEED data terminal(s), 46,480,482
40 terminals, 480, 484, 610, 621
4540 terminals, 46,483
picture, 484
Data station selector (055), 538, 539, 540
Data store, DMS-10 switching system, 431
Data terminal(s), 20, 45-47, 465, 480, 483-84
for satellite communications systems, 360
see also DATASPEED data terminal(s);
Teleprinter; Transaction telephone
Data terminal equipment (DTE), interface,
319-20,322,323
Data traffic
engineering concepts, 190, 191
models, 186-88
nature of, 184-85
performance concerns, 188-90
Data transfer time, 190
Data transmission
private-line services, 66
satellite communications systems, 360
see also Data communications systems;
Data network services; Data set(s);
Data terminal(s)
Data under voice (DUV), 382, 396, 533,534
Davisson, C. J., 22
o channel bank(s), 377, 381
Deaf persons, see Hearing-impaired
persons, aids for
Debt ratio, Bell System, 721
Decoder, 408
Decoding, see Conversion
Dedicated plant assignment card (DP AC),
621
Defect, in quality assurance audit, 747, 749,
750,756
definition of, 746
graph,747
seriousness classifications, table, 748

Defective Equipment Replacement Program
(DERP),582
Delay, 233-34
answer, 270
definition of, 233
loaded cable, 334
postdialing, 270, 280
satellite communications systems, 360
see also Dial-tone delay
Delay distribution, see Grade of service
Delay probabilities, see Grade of service
Demand interrupt, 411
Demarcation point (OP), 306, 307, 308, 309,
310,311
Demodulation, 210, 211, 212, 213
data sets, 477
see also Modulation
Demultiplexing, 218, 219, 251, 392, 393
digital carrier system, 370
digital signaling, 291
see also Multiplexing
Depolarization, satellite communications
systems, 360
Depreciation
book,724,725
tax, 720, 734
DESIGN LINE decorator telephone(s), 41, 79
picture, 42
Dial, 85, 468, 470
see also Pushbutton keypad; Rotary dial
Dialing, 117,267-68,275
DTMF, 161,266,468
pulse, 160-61, 258, 401
see also Telephone call
Dialing time, definition of, 270
DIAL-IT network communications services,
9,61,65, 122, 426, 513
see also Media Stimulated Calling; Public
Announcement Service
Dial private branch exchange (dial PBX),
493-94
diagram, 495
fields of use, graph, 497
Dial pulse(s), 258, 275,284,468,470
diagram, 275
Dial-pulse (trunk) address signaling, 284,
289,290
Dial-pulse-delay distortion, 339
Dial-pulse receiver, 150, 160
Dial-repeating trunk(s), 284,285
Dial tone, 117, 149, 160, 267, 268, 401, 404,
406
coin service, 275
TOUCH-TONE service, 737
see also Telephone call
Dial-tone delay, 153, 163, 180, 183,270, 594,
672,676
stored-program control systems, 162
Dial-tone detector, 470

Index

Dial-tone-first
service,77,372,471,472,474,475
signaling, 275
telephone, 77
Dial-tone marker, 406
Dial-zero basis, in international calling, 439
Diamond State Telephone, 12, 36
Dielectric materials, in optical fibers, 205
Diesel-electric alternator, 543, 544
Differential phase-shift keying (DPSK), see
Modulation
Diffuser, 564
Digital Access and Cross-Connect System
(DACS), 391, 601, 605, 658, 733
Digital carrier module (DCM), 431, 432,433
Digital carrier transmission, 369-93
circuit-miles by type, tables, 395, 396
interoffice, 373-85
intercity, 381-85
metro, 373-79
outstate, 379-81
loop, 370-73
multiplex equipment, 386-93
Digital carrier trunk (DCT)
frame, 390
interface, 295,301-5
Digital channel, 189n, 199,200,272
Digital channel bank, see Channel bank
Digital communication, see Computer-tocomputer data transfer
Digital cross-connect (DSX), 370
Digital data signal, see Digital signal
Digital Data System (DDS), 46, 70-71,197,
530-35
Basic Packet-Switching Service, 72
characteristics, table, 322
D4 channel bank, 389
diagram, 532
interface, 313-15, 320-21
tables, 314, 321
network facilities, table, 534
lA-Radio Digital System, 382
point-to-point channel, diagram, 531
service objectives, 189,237
SLC-96 carrier system, 373
see also DATAPHONE digital service
Digital data transmission, see Data
communications systems
Digital encoding, Advance Calling, 740
Digital Equipment Corporation,
minicomputers, 453, 454, 455, 639
Digital facilities network, evolution, 132-34
Digitalfacility, 199
see also Channel(s)
Digital facility interface (DFI), 390
Digital interface controller (DIC), 390
Digital interface frame (DIF), 389, 390, 428,
429,430
Digital interface unit (DIU), 390

843

Digital line/trunk unit (DLTU), 390
5ESS switch, 437
Digital loop carrier systems, 370, 372-73
diagram, 371
see also SLC-96 carrier system
Digital Ordering and Planning System,
639n
Digital pair-gain system, 134
Digital services
characteristics, table, 322
inherent network protection, 313
Digital services unit, 436
Digital signal, 193, 194, 196-97,198, 199,
242,272,299,337,369,370,372,
476,477
data transmission impairments and
objectives, 236-40
DIMENSION PBX, 501
modulation, 211, 217-18
multiplex synchronization, 222-23
pulse stream, diagram, 197
regeneration, 308
U 600 channel, 366
see also Data signal; Repeater(s)
Digital Signaling, 287,291-92
paging systems, 525
techniques, 286
Digital signal (DSn) level, 221, 222, 299,
300, 301, 307, 312, 365, 366, 381,
382, 383, 386, 388, 389, 390, 391,
392, 393, 531, 533
Digital Data System, diagram, 532
DSX-1, diagram, 300
FDM plan, diagram, 363
PCM-TDM hierarchy, diagram, 386
Digital switching, private branch
exchanges,499-502
diagram, 502
see also DIMENSION private branch
exchange; 80/5 Call Management
System
Digital switching systems, 132, 133, 134,
336,394
see also Electronic switching system(s);
Lightwave digital system(s); T1
digital carrier system
Digital technology, at Bell Laboratories,
22
table, 21
Digital time-division multiplex hierarchy,
221,222,299,386-87
diagram, 386
Digital time-division switch, 253
Digital time-division switching network(s),
251
Digital-to-analog conversion, 46, 97, 133,
215,216
Digital-to-analog converter(s), 132-33, 251,
258,413

844

Index

Digital transmission, 43, 46, 133, 251, 428
advantages, 132
Bell Laboratories, 21, 25
electronic switching, 413, 414
impairments and objectives, 223, 236-40
metro area circuits, 330
network interface, 307
PICTUREPHONE meeting service rates,
527-28
synchronization, 221-23
see also Data communications systems;
Digital carrier transmission;
Modulation, pulse-code; Protocol
Digit receiver, 149-50
delay, 180
engineering of, 153-54, 160-61
Digit register, 268
Digit transmitter
switching congestion, 177
timeouts, 179
Digroup,305,386,389,429
Digroup terminal (DT), 389, 428, 429, 430
Digroup terminal controller (DTC), 389
Digroup terminal frame, 389
Digroup terminal unit (DTU), 389
DIMENSION private branch exchange
(DIMENSION PBX), 497, 499-503
automatic call distributor, 503
diagram, 500
feature package, list, 501
Diode-transformer gate, 419
Direct current (dc), 541
as carrier signal, 211
energy storage
for customer-premises equipment, 549
electrochemical cells, 543
engine-alternators, 543-44
power system, 544-47
dc-to-ac inverter, 544, 545, 548-49
dc-to-dc converter, 544, 545, 548
diagram, 545
power plant output voltages, table, 547
rectifiers, 546-48
see also Signaling
Direct current (dc) signaling, 286-89
range in vOice-frequency transmission, 339
Direct Department Calling, 501, 503
Direct distance dialing (DDD), 406
billing, 577
medium-size business customer, 54
numbering plan, 114-19
TSPS, 439
Direct inward dialing (DID), 55, 407, 501
dial PBX, 494
telephone answering systems, 506-7
diagram, 507
Directory assistance, 115
operators, 62, 111, 112, 584
public telephones, 77
toll-free service, 122

Directory organization, corporate functions,
predivestiture
AT&T,8,9
telephone company(ies), 14
Directory Service(s), 80,577,584-85
Direct progressive control, 258-59
Direct services dialing capability (DSDC),
61,64,511
Direct trunk(s), 96
alternate routing, 170
crosstalk objectives, table, 234
digital channel bank, 388
engineering, 137
local network, 104, 106
loss objectives, 225
table, 226
No. SA Remote ·Switching Module, 437
Direct trunk group
alternate routing, 170
engineering, 172, 173
local network, 104, 106
toll network, 110
Disabled persons, see Handicapped, aids for
the
Disconnection of service, 79
records, 619, 620
Discounted payback period (DPP), 723
Disk storage, 46-47, 484
4ESS switch, 427-28
Mass Announcement System, 513
3B20 computer, 513
3B20D computer, 536
Distant Early Warning (DEW) Line, 19
Distortion
amplitude / frequency, 234-35
diagram, 235
linear
data sets, 478
digital carrier system, 375
of digital data signals, 238-39
in loaded cable, 334, 339
nonlinear
analog carrier systems, 336
of digital data Signals, 239
voice-frequency transmission, 335
single-frequency Signaling, 289, 291
Distributed switching, 438
Distributing frame(s), 328, 541, 549-59
administration and engineering, 555-59
assignment methods, diagram, 557
termination layout mask, 557
functions, 550-51
diagrams, 550
hardware, 551-55
diagrams, 553, 554
pictures, 552, 555, 556, 558, 560
Distribution
diagrams, 247, 248, 250
facilities and services, 10
space-division network(s), 245-48

Index

Distribution areas, 122, 123
Distribution cables, 123, 124, 328
Distribution centers (Western Electric), 16,
17
District junctor, 404
Divestiture, 4, 29-32, 34, 571
see also Modification of Final Judgment
Dividend payments, 726
DMS-10 sWitching system, see Electronic
switching systems
DMS-100/200 switching systems, see
Electronic switching system(s)
Docket 19528,689,702
Dodge, H. F., 743n
Domestic satellite carrier, 306
Domestic satellite systems, 357
DIA channel bank, 387, 388
DIB channel bank, 387, 388
D1C channel bank, 387, 388
DID channel bank, 300, 388
D2 channel bank, 300, 388
D3 channel bank, 300, 380, 388
D4 channel bank, 300, 302, 380, 389, 390
Double-sideband amplitude modulation
(DSBAM),211-14
see also Modulation
DR11-40 digital carrier system, 383
Drop wire (buried wire service), 328
DR6-30 digital carrier system, 383, 534
Dry reed matrix switch, see Reed matrix
switch
DSn, see Digital signal level
DSX-l, see Interface(s), internal
D-type channel bank, 387, 388
tables, 394, 395
Dual polarization, 204-5, 346
Dual-tone multifrequency (DTMF), 468, 470
receiver, 150, 161
signaling, 286-87
symbols, 116
see also Signaling; TOUCH-TONE
telephone
Duplex central processor (DCP), 536, 537
Dynamic overload control (DOC), 181
E&M lead interface, 270
E&M lead Signaling, see Per-trunk signaling
Early Bird (lNTELSAT I) satellite, 358
Earth station(s)
High-Capacity Satellite Digital Service,
73
satellite communications systems, 133,
356,359,361,362
Echo,99,199n,223,225-30,244
carrier transmission, 335
No. 4A Crossbar System, 400
customer opinion models, 672-73
grade-of-service rating, 674-76
satellite communications systems, 360

845

speakerphone, 471
vOice-frequency transmission, 340
Echo canceler, 21, 228
diagram, 230
satellite communications systems, 360
Echo suppressor, 228, 310
Economic analysis in network planning,
140-41
Economic CCS, engineering, 172-73
Economic evaluation, of products and
services, 707, 713-29
calculating measures for, 718-27
accounting, 726
cash flow, 719-23
regulatory, 723-26
input data, 717-18
results, preparation of, 727-29
Economic Impact Study System (EISS), 718
Economics
alternate routing, 169-72
capacity expansion, 138-39
circuit routing, 138
financial measurements, 676
multiplexing, 139,219
network planning, 140-41
new technology, 730-31
cost reduction example, 731-33
PICS/DCPR,615
quality assurance, 750, 751, 752
service and performance objectives, 665,
675
TIRKS, 610-11
see also Economic analysis in network
planning; Economic evaluation of
products and services
Edison, Thomas, 700
Efficiency
operator services, 589
trunk group, 106, 137
EIA (Electronics Industries Association)
interface
RS-232-C, 319, 321
RS-449, 319, 321
800 Service, 63
80/5 Call Management System (CMS), 504
Expanded,63,511-13
numbering, 122
Elastic store, 73
E-Iead, see Per-trunk Signaling
Electrical hazards, network protection,
313
Electrical protection
distributing frame, 551·
telephone equipment building, 561
Electrical signaling, 242
Electric utility outage, see Power failure
Electroascoustic transmitter, 465-66
Electrochemical cell, 543, 546
Electromagnetic receiver, 465-66
Electromechanical switches, 30, 244

846

Index

Electromechanical switching system(s), 398409,410, 462
billing data transmitter, 454
ICAN,629
interoffice facilities network, 130
local facilities network, 124
traffic data acquisition, 623
see also Crossbar switching system(s);
Panel switching system(s); Stepby-step switching system(s)
Electromechanical telephone, 465-69
coin telephone, 471
Electronic blackboard, see GEMINI 100
Electronic Blackboard System
Electronic cash register, 480
Electronic coin public telephone (ECPT), see
Electronic telephone
Electronic components, manufacturing of,
table, 15
Electronic funds transfer, Transaction
telephones, 475
Electronic graphics, 48-49, 484-85
Electronic marker, 261
Electronic private branch exchange
(electronic PBX), 498-99
800A PBX, diagram, 498
Electronics
in loop carrier systems, 370
in residential telephones, 489
Electronics Industries Association (EIA),
310,319,321
Electronics Technology area (Bell
Laboratories), 24, 25
Electronic switches, 244, 245
gates, 253, 255
Electronic switching system(s), 409-38
auditing, 747
billing, 450
Call Forwarding, 56
Custom Calling Services I, 57
data processor, electronic, 410-11
distributing frame, 553
evolution of, 413-14, 735
exchange business services, 58-59
ESS-ACD, 59, 411, 420
ESSX-l, 59, 66-67
generic program, 411
interfaces, 301-5, 389-91
diagram, 303
table, 302
load balancing, 164
local facilities network, 124
maintenance, 599
memory, evolution of, 731-33
picture and table, 732
numbers of, in Bell System, 461-63
private network services
ESSX-l,66-67

quality assurance audit, 747

signaling, 280
space-division
No. lOA Remote Switching System,
130,422-25
diagram, 422
lESS switch, 113, 283, 413, 414, 415,
416,420,625
automatic call distributor, 504
equipment processor, 415-16
master control center, picture, 633
Switching Control Center, 632-34
TSPS, 439
lAESS switch, 414,731
mobile telephone service, 522-23
1/IAESS switches, 390, 416-20
billing, 450, 451,458
COEES, 639
interfaces, 302, 303, 305, 390
LAMA, diagram, 450
maintenance, 420
network structure, diagram, 417
NSPMP, 681-83
number of, in Bell System, 461-63
peripheral systems, 416-20
services, 420
2ESS switch, 413-14, 420,625
network,420
processor, 420
2BESS switch, 422
AMARC, 456
3A central control, 421
2/2BESS switches, 422
billing, 458
COEES, 639
interfaces, 302, 390
number of, in Bell System, 461
3ESS switch, 414, 422, 625
AMARC, 455-56
billing, 458
interfaces, 302, 390
number of, in Bell System, 461
stored-program control, 410
chronology of, table, 459
switching congestion, 180
Switching Control Center System, 637
time-division
DMS-I0 switch, 414, 430-34
AMA transmitter, 457
diagram, 432
network complex, 431, 433
processor complex, 431
software structure, 433-34
DMS-I00 switch, 459
DMS-200 switch, 459, 462n
4ESS switch, 133, 134,283,330,414,
425-30,462,627
administration and maintenance,
427-28
billing, 451, 458

Index

capacity, 426-27
data collection, 627
diagram, 428
features, 426
Expanded 800 Service, 513
interfaces, 302, 389-90
Mass Announcement System, 513,
515
number of, in Bell System, 462, 463
software structure, 427
switching network, 428-30
TSI-TMS complex, diagram, 430
5ESS switch, 133,414,434-38,458,462
administrative module, 434, 436
AMARC, 456
billing, 458
communications module, 436
diagram, 435
Digital Ordering and Planning
System, 639n
interface modules, 436-37
interfaces, 302, 390
multiprocessing, functional, 264
No. 5A Remote Switching Module,
437
software structure, 537-38
101ESS switch, 497
telephone call, typical, 412-13
traffic, 151
traffic data acquisition, 623
traffic measurements, 183
Electronic Switching System Programming
Language (EPL), 427
Electronic tandem switching (ETS), 68, 69-70
Electronic telephone, 469-70
diagram, 469
multibutton DIMENSION PBX, 501-2
32A, coin public, 472-73
Electronic Translator System(s), 458
No. 4A, 283,408, 627n
No. 5,455
Emergency service, see 911 Emergency
Service
Employees, designing for, 738, 740-42
diagram, 738
Encoding, see Conversion
End office (class 5 office; local switching
office), 108, 110,111,126,668
congestion, 180
service objective, 183
switching hierarchy, position in,
diagram, 108
talker echo, 226, 227
see also Central office
End-Office Connection Study, 668
diagram, 669
End office toll trunk, 110
End-of-study effect, in economic analysis,
140

847

End-to-end blocking, 184
End-to-end digital connectivity, 65, 396, 530
End-to-end service objective, 183
End-to-end signaling, 58
End-user, 31, 186,317,659,666,710,749
Energy, Department of, U.S., Sandia
Corporation, 20
Energy sources, 543
Energy storage, 542, 543-44, 546
Engine-alternator set, 543-44, 545, 546, 547,
549
picture, 544
Engineering
Bell Laboratories contributions to, table, 21
switching system(s), 160-65
telephone company function, 585, 586,
638-42
trunk groups, 165-68
Engineering and Administrative Data
Acquisition System (EADAS),

436n,623-28,629,659
Engineering and Administrative Data
Acquisition System / Network
Management (EADAS/NM), 595,
625,627,659
Engineering and operations support
(central services organization), 37
Engineering center, marketing support, 575
Engineering complaint(s), 758-59
Engineering periods, 151-55
Engineering Research Center (Western
Electric), 14-15, 17
Enhanced 911 (E911) service, see 911
Emergency Service
Enhanced Private Switched Communication
Service (EPSCS), 68, 69, 113
typical network, diagram, 114
Enhanced services, 316-17
Computer Inquiry II, 702-3
see also Service(s)
Entities, definition of, 316
Environmental support, telephone
equipment building, 561, 563
El, see Equipment (El) system
EPL, programming language, 427
EPLANS Computer Program Service, 642-43
EPLX, programming language, 427
Equal access, 30
Equalization, 308
adaptive transversal equalizer, 478
cable systems, 340, 351, 375, 377, 378, 381
data sets, 479
microwave radio, 356
Equipment
average service lifetime, 717
capacity expansion, 139
compatibility specifications, 27
competition in manufacture and
provision of, 6, 28, 703, 743

848

Index

Equipment (contd)
failures, 176
interface, definition of, 266
maintenance, administration, and control,
621-38
manufacturing of, table, 15
operator services, 88, 438-45
PICS, 611-15
sharing, 147
standards and procedures, 6
TIRKS, 605, 607-11
in traffic theory, 153
transition management, 595
utilization, 594, 610
see also Customer-premises equipment;
Station equipment; Terminal
equipment
Equipment and facility recovery, 603n
Equipment arrangements, design of, Bell
Laboratories Specific Development
and Design, 23
Equipment building system(s), 541, 559-65
electrical, 561
equipment areas, 559,561
mechanical, 561, 563
standards, 564-65
Equipment Compatibility Committee,
USITA,27
Equipment Design Standards, 564
Equipment frame, 559, 561
picture, 560
~quipment (E1) system, 607, 608, 609
Equipment-to-equipment (EEl) interface,
307,308
Equivalent random method, 168
Erasable programmable read-only memory
(EPROM),423
Erlang, A. K., 149
erlang (unit), 149
Erlang's Delay (Erlang C) Formula, 156-58,
160, 161, 162, 167, 589
Erlang's Loss (Erlang B) Formula, 158-59,
166, 167, 168, 172
Error-checking code, cqs, 294
Error-correcting code, 197,236
lESS switch, 415
Error-detecting code, 197, 294
Error detection
digital systems, 375, 382
packet switching, 293
Error objective
digital systems, 375, 378, 382, 384
packet switching, 189
Error rate, see Bit error rate; Block error rate
Error-seconds, 237
E6 repeater, 339
ESS-ACD, 59, 411, 420
ESSX-1, 59, 66-67
Exchange access services, see Service(s)

Exchange area, 56
definition of, predivestiture, 31
diagram, 105
paired cable, 202
trunks, 374n
Exchange area network, 103
see also Local network
Exchange business services, 58-59
Exchange Feeder Route Analysis Program
(EFRAP),144
Exchange lines, 56-57
Exchange service(s), 56-60
local switching system, 398
Executive control loop, 4ESS switch, 427
Expanded 800 Service, 63-64
Expansion
local switching system, 397
space-division switching networks, 24546,248
diagram, 248
time-division switching networks, 258
Expenses, 719,720,723,724,726,728
Experimental psychology, 738
Express terminal, 384
Extended Binary-Coded Decimal
Interchange Code (EBCDIC), 318
Extension telephones, 42, 43, 114n, 489, 494,
495,496,498,500,502
see also Telephone set(s)
Extreme value engineering, data networks,
190
Facilities
definition of, 305
for non-Bell System communications
firms, 40
non - Bell System standards and
procedures, 6
TIRKS, 605-10
Facilities Assignment and Control System
(FACS),501
Facilities network, 81, 84-88, 103,305
digital facilities, 132
interconnection, 306, 311
structure, 122-34
trunk requirements, impact on, 174
see also Interoffice facilities network;
Local facilities network
Facility, definition of, 305
Facility administration, 593
Facility and Equipment Planning System
(FEPS), 607, 609
diagram, 609
Facility Capacity (FCAP), 144
Facility-dependent signaling system, 286,
291,292
Facility-independent signaling system, 286,
289

Index

Facility interface processor (FIP), 536, 537
Facility junction point, 81
Facility network planning programs, 64142
Facility signaling system(s), see Signaling,
techniques
Facility (Fl) system, 607, 608,609
Fading,208,209,352,383,519
Fast busy tone, see Reorder tone
Feature package, DIMENSION PBX, 500, 501
Federal Communications Act, 689, 692
Federal Communications Commission
(FCC), 32, 41, 689, 692, 693
u Above-890" Ruling, 693
Bell PhoneCenters, 488n
cellular radio, 74,518,521,524
computer inquiries, 79, 689, 702-3
customer-premises equipment, 574n
data processing and telecommunications,
702-3
Docket 19528,689,702
frequency assignment, 207, 346, 358, 517,
693
interconnection(s), 306, 310, 311, 479, 499
interconnection rulings, 693-94, 701
interstate service, 68, 69, 699
mobile telephone systems, 517, 525
Part 68 Rules and Regulations, 312-;13,
487
plug-in units, 614
satellite communications systems, 358,
361
separations, 699
Specialized Common Carrier Decision,
693
tariff on new services, 708
terminal equipment
registration program, 307, 311-12, 689,
694,702
retail sales, 71
Federal Telecommunications Service (FTS)
network,91
Feeder cable (feeder route), 124,327,379
Feeder route area(s), local facilities network,
122-23
Ferreed switch, 418, 421, 442
Ferrod sensor, 418
Field effect transistor, insulated-gate, 421
Field performance tracking, 747, 759-60
Field representatives (Bell Laboratories
Quality Assurance Center), 27,
757-58,759
locations of, map, 758
File store, 416
Filter, 387
bandpass,213,341,364,365
channel separation, 390
high-pass, 364, 365
low-pass, 215, 341, 342

849

Final trunk(s), 106
Final trunk group(s), 166
cluster busy-hour technique, 173
definition, 106
engineering, 137, 166, 169, 175
measurement, 184
servicing, 176
see also Trunk Servicing System
Finance and General Services area (Bell
Laboratories),24, 25, 26
Financial measurements, 676-77
Fine-motion switch, 244
Firmware
DMS-I0 switching system, 433
packet-switching system, 537
First in, first out (FIFO), 149, 151, 158
5ESS switch, see Electronic switching
system(s)
500-type telephone
antisidetone network, diagram, 468
diagram, 467
picture, 466
signaling, 468-69
transmission, 465-68
Flat rate, 99, 445, 578, 697, 698
Flat-rate billing, 56
Floppy disk, 484
Flow-through technology, digital carrier
trunk,390
Focused overload, 180,596
definition, 179
Folded network, 421, 422
Fl, see Facility (Fl) system
Force management, 593
Forcing, operator services, 162,589
Forecasting, 585
equipment utilization, 594
in network planning, 142, 143
operator requirements, 593
project evaluation, 717
trunk, 174-75,580,630-31
Trunk Forecasting System, 625, 631
Foreign countries, calls to, see Service(s),
international telephone
Foreign exchange (FX) service, 40, 61, 63
data communications, 44
digital carrier system, 373
line, 55, 88-89
signaling, 265, 284, 290
Foreign numbering plan area (FNPA), 116
see also Numbering plan
Fortune, 14
43 teleprinter, see Teleprinter
4A call distributor, 504
4A speakerphone, see Speakerphone
411 directory assistance operator, 111,
112
see also Directory assistance; Directory
Service(s)

850

4ESS switch, see Electronic switching
system(s)
Four-wire analog carrier interface, 295-99,
301
Four-wire switching, see Switching
Four-wire transmission, see Transmission
Framing bit, 220, 375n
Frequency bands, assignment of, 207, 346,
358, 517,693
see also Analog carrier transmission;
Digital carrier transmission
Frequency diversity, microwave radio, 209,
346
Frequency-division multiplex (FDM), see
Multi plexing
Frequency frogging, short-haul microwave
radio, 346
Frequency modulation (FM), see
Modulation
Frequency-shift keying (FSK), see
Modulation
Frequency synthesizer, Advanced Mobile
Phone Service, 523
FT3 digital lightwave system, 378-79
FT3C digital lightwave system, 383-84, 534
diagram, 385
Fujitsu, 498
Full-detail bill format, 446
Full [trunk] group(s), 166n
Fundamental planning, 585, 590, 591
Future worth, 721
Gain (amplification)
electronic telephone, 470
4-wire analog carrier interface, 296, 297,
299
L-carrier systems, 350
satellite communications systems, 359
speakerphone, 470-71
VOice-frequency transmission, 334, 33839,340
Gain hits, 240
see also Incidental modulation
Gas turbine-alternator set, 544
Gate
time-division switch, 253
time-multiplexed switch, 255
diagram, 256
Gemini, Program, 19
GEMINI graphic communications system,
485
GEMINI 100 electronic blackboard system,
48,484-85
components, diagram, 486
picture, 49
structure, diagram, 487
General Departments (AT&T), 5, 11, 32
corporate functions, predivestiture, 6

Index

General Services area (Bell Laboratories),
24,25,26
General Telephone and Electronics Satellite
Corporation (GSAT), 357
General-trade products
manufacturers, 6, 28
quality assurance, 743
Generic program(s)
stored-program control in electronic
switching, 411, 434
Total Network Data System, 632
Geostationary satellites, see Satellite(s)
Glass fiber, see Optical fiber(s)
Gold and Stock Telegraph Company, 700
Government and Commercial Sales
Division (Western Electric),
corporate functions,
pre divestiture, 19
Government regulation, see Legislation;
Regulation
Government systems
Bell Laboratories, 25
Western Electric, 19, 26
Grade of service (GOS), 149-51, 152, 153,
154,160
Advanced Mobile Phone Service, 523
criteria
delay distribution, 157, 158
graph,150
delay probabilities, 153
echo, 227-28
estimates, 183
loss-noise objectives, 230-32
graph,231
objectives, 224, 225
performance objectives, 670
ratings, 674-75
trunk groups, 166
Graham-Willis Act, 689, 691-92
Graphics terminal(s), 48-49,484-85
Gray, Elisha, 700
Gross-motion switch, 244
Ground-start signaling, 273,275,284
Half-duplex transmission, 477
in Bell System data sets, 481, 485
Handicapped, aids for the, 49-51
pictures, 50,51
Handoff, 524
Hands-free telephone equipment, 48
Hardwired DCPR subsystem, PICS/DCPR,
613,614
Hardwired equipment, 611, 612, 614
Harmonic interference, 276
Hearing-impaired persons, aids for, 49, 51
picture, 51
H88-loaded cable, 333-34,338
Heterodyne repeater, 352-53, 356

Index

Hewlett-Packard computers, 601
High-Capacity Digital Transport Services,
73
High-Capacity Satellite Digital Service
(HCSDS),73
High-Capacity Terrestrial Digital Service
(HCrDS),73
PICTUREPHONE meeting service, 75, 527
High-frequency line, 200-201
High-pass filter, 364, 365
High-seas maritime radiotelephone service,
74-75
High-Speed Switched Digital Service
(HSSDS),73
PICTUREPHONE meeting service, 75, 527
High-speed train telephone service, 74, 75
High-usage (HU) trunk(s), 166n
alternate routing, 170-72
in private switched networks, 113, 114
High-usage (HU) trunk group(s), 109, 110,
166
alternate routing, 170-73
cluster busy-hour technique, 173
diagram, 174
cost, 172-73
definition of, 106, 166
engineering, 137,166,171
minimum group size, 172
multi hour engineering, 173
nonrandom load, 168
optimal group size, 173
servicing, 176
special-services signaling, 284
Hi-lo N-carrier repeater, 343, 344
Hold feature, 57
business customer equipment, 53
key telephone systems, 490
2-line telephone sets, 43
Holding company, 691
Home numbering plan area (HNPA), 116
see also Numbering plan
Homing chain, in toll network, 109
HORIZON communications system, 491-93
diagram, 492
human factors in design of, 740
Horizontal group, 164
Horn reflector antenna, 352, 353-55
Hotel calls, guest-originated, 439
"Hot slide" installation, 422
Hour entry, LAMA, 449
Huinan factors, 737-42
Bell Laboratories, goal, 373-79
customer services, design of, 707n, 739-

40
employees, designing for, 740-42
diagram, 738
graphics terminal, 485
simulation, picture, 741
video teleconferencing, 528

851

Hundred call seconds (CCS) per hour, 149n,
171, 172, 182, 426
Hundred call seconds per hour (CCS)/main
station, 125
Hush-A-Phone Decision, 689, 694
HU trunk group, see High-usage trunk
group(s)
Hybrid, 200, 226,227,295
digital transmission, 370, 371, 387
Hybrid integrated circuitry,
A6 channel bank, 365
Hybrid signal, 196, 198
IBM Corporation
compatible computers, 451, 604, 610, 615
Information Management System, 615
Identifier, telephone answering systems,
506
diagram, 505
Idle, see Busy /idle status
Idle circuit noise, 225
Idle tone, mobile telephone systems, 519
Illinois Bell, 12, 36
Impairments, transmission, 223-40
analog speech signals, 224-36
am plitude / frequency distortion, 234- 35
carrier frequency shift, 235
crosstalk, 232-33
delay, 233-34
echo, 225-30
inductive interference, 235-36
message circuit noise, 230-32
overload, 236
received volume and loss, 224-25
digital data signals, 236-40
impulse noise, 237-38
incidental modulation, 239-40
linear distortion, 238-39
nonlinear distortion, 239
Impedance, 226, 336, 467
Impulse noise, 223, 230, 237-38
digital carrier system, 375, 376
Inband signaling, 286, 289-90
DA TAPHONE Select-a-station, 540
Incidental modulation, 239-40
Income statements, 726
Incoming calls, business customer services
for, 52, 53, 54, 55
Incoming digit-receiver delay, 183
Incoming register, 95, 96, 97
see also Register
Incremental study, in project evaluation,
717,718
Independent telephone companies, 6, 10,
11, 27, 40, 84, 571, 701
concurrence in tariffs, 697
distribution of offices in PSTN hierarchy,
109

852

Independent telephone companies (contd)
Kingsbury Commitment, 691
predivestiture relationship with Bell
System, 27-28
proliferation, 700-701
resources and volume of service (1982),
table, 29
settlements, 699
Index plan, see Measurement, in service
evaluation
Indiana Bell, 12, 36
Indices, see Measurement, in service
evaluation
Individual Circuit Analysis (ICAN)
program,624,625,627,629-30
Induced power hum, see Inductive
interference
Inductive coordination, 236
Inductive interference, 195, 230, 235-36
Inductive loading, VF facilities, 330
Information, signaling, 269
Information, technical, 27-28
Information Services directories, 80
Information systems development (central
services organization), 37
Information Systems organization (AT&T),
corporate functions,
predivestiture, 8, 9
Information theory, 21
Inherent network protection, 308, 313
Initial entry, LAMA, 448
Input/ output processor (lOP), 434, 435, 436
Insertion loss, 224n
Inside plant, 23n
Installation, 678
services, 79-80, 576
telephone sets, 488
Integrated circuit, 27, 730, 731,733,734
digital carrier transmission, 337
small communications systems, 491
Transaction telephone, 476
2BESS switch, 421
Integrated Planning and Analysis (IPLAN)
system, 718
Integrated services digital network (ISDN),
65,458,464,530
INTELSAT I (Early Bird), 358
Interactive television, see Television
Intercept call, 111, 112,443,444
Intercept file, 445
Intercept operators, 62, 112, 443, 445
Intercept trunk, Automatic Intercept
System, 443, 444
Intercity digital carrier systems, 381-85
Intercity Facility Relief Planning System
(IFRPS), 642
Intercity network, 130-32
coaxial cable, 203
multiplexing, 220

Index

Intercom calling, 52, 53
dial PBX, 493
HORIZON communications system, 492
key telephone system, 490
Interconnection, 30, 305-16
Bell System services only, 307-8
diagram, 307
definitions, 305-6
FCC rulings, 693-94, 701-2
interfaces, 306-7
evolution of, 310-11
network,313-16
network protection, 311-13
OCC and Bell System facilities, 308-10
diagram, 309
Interest on debt, 721,726
"lnrerexchange carrier (IC), 30, 31,703
Interexchange carrier services, 31
Interexchange services, 30-31
Interface(s)
AMARC, 454-56
CCIS, 272, 277, 280, 282, 283-84
definition,
in customer-line signaling, 272, 273
data communications, 319-23, 536, 537,
538
diagrams, 320, 322
table, 321
de signaling, 286-89
definitions, 266
digital carrier transmission, 370, 378
digital channel bank, 387, 388
Digital Data System, 531
digital time-division switching network,
251
DMS-10 switching system, 431
E&M lead signaling, 271, 273, 277, 27880,281,282,283,284,287,289,290
ESS switches, 389-91, 413,428,429,434
5ESS switch, 436-37
4-wire / 2-wire transmission, 226
inband signaling, 289
interconnection, 305-16
definitions, 305-7
environment, 307-10
diagrams, 307, 309
evolution, 310-11
network protection, 311-13
typical interfaces, 313-16
table, 314
internal, 294-305
digital carrier trunk, 295, 301-5, 390
versus traditional interface, diagram,
303
DSX-1, 295, 299-301
diagram, 300
specifications, table, 301
4-wire analog carrier, 295-99
diagram, 297

Index

interoffice trunk signaling, 276-84
loop carrier signaling, diagram, 274
loop-reverse battery, 273,277-78,280,
283,288,289,290
loop signaling, 271, 272, 273, 274, 275,
284,287,288
No. lOA Remote Switching System, 423
per-trunk signaling, 277-80
diagrams, 279, 281, 282, 283
PICS/DCPR,615
postdivestiture, 683
selector control unit, 540
serving area, 123,328
signaling techniques, 286-94
single-frequency signaling, 290
special-services signaling, 284-86
stored-program control, 263
3B20 computer, 513
TIRKS, 609-10, 615
TOUCH- TONE service with older
equipment, 736-37
TSPS, 439
2-wire /4-wire carrier-derived channels,
200,297
diagrams, 200, 297
see also Demarcation point; Equipment-toequipment interface; Network
interface; Protocol
Interface device, 266
Interface module (1M), 5ESS switch, 434,
435,436-37,438
Interface specifications, 266, 295
DSX-1 digital system cross-connect
interface, 301
table, 301
Interference
CCIS, 294
digital carrier system, 337, 375, 377, 378
inductive, 195,230,235-36
land mobile telephone systems, 519, 521,
524
network protection, 276, 313
overload, 236
satellite communications systems, 358, 361
diagram, 361
terrestrial radio systems, 209, 346
Interflow, in ACDs, 504
Inter-LATA telephone service, see Service(s)
Interlocation dialing, private network
services, 68
Intermediate distributing frame (IDF), 302,
553
Intermediate frequency (IF), 352-53, 356
Internal-combusion engine-alternator set,
543,544
picture, 544
Internal rate of return (IROR), 719, 723
International direct distance dialing
(IDDD), lIS, 120-21, 439

853

International numbering, 115, 120-21, 439
International Organization for
Standardization (ISO), 319
International record carrier(s), 40
International services, see Service(s),
international telephone
International Telecommunications Satellite
Consortium (INTELSAT), 357-58
International Telecommunication Union
(ITU),28
International Telephone and Telegraph
Corporation (ITT), 490n, 499
International telephone service, see
Service(s)
Interoffice call, 94, 397,437
Interoffice circuit(s), 328, 329, 330
Interoffice facilities network, 86, 122, 12532,327
diagrams, 127, 129
major Bell System transmission routes,
map, 131
planning, configuration, 134-35
see also Carrier system(s); Special-services
circuit(s); Trunk(s)
Interoffice plant, 394
Bell System investment, 332
digital carrier transmission, 373-84
long-haul network, 331-32
metropolitan area, 329-30
Interoffice trunk Signaling, 265, 268, 269,
273,276-84
Interregional trunk, 229
Interrupt-driven scheduler, electronic
switching systems, 427
Interruptions per minute (ipm), 94
Interrupt mechanism, electronic switching
systems, 411
Interstate Commerce Act, 688, 689
Interstate Commerce Commission (ICC),
688
Interstate private teletypewriter networks,
91
Interstate services, see Service(s)
Interstate telephone service rates, 698
Interstate WATS, see Wide Area
Telecommunications Services
Intersymbol interference, 378
digital data Signals, 238
Intertandem trunk, 107n
crosstalk objectives, 234
echo, 227
loss objectives, 229
Intertoll congestion, 596
Intertoll network, see Long-haul
transmission facilities
Intertoll trunk, 108, 331
CCIS,509
congestion, 179
crosstalk objectives, 234

854

Intertoll truck (contd)
digital carrier transmission, 382
digital channel bank, 388
echo, 226, 227
diagram, 227
loss objectives, 229,332
Via Net Loss plan, 227
Intraflow, in ACDs, 504
Intra-LATA telecommunications services,
see Service(s)
Intraoffice call, 94, 125, 128,412,419,422,
423
table, 125
Intrastate networks, 11
Intrastate telephone service rates, 699
Intrastate WATS, see Wide Area
Telecommunications Services
Inventory management
PICS, PICS/DCPR, 611-15
TIRKS,608
Transaction telephones, 475
Inventory manager, 611
Inverter, dc-to-ac, 544, 548-49
Investment tax credit, 720n, 725
INWATS, see 800 Service; Wide Area
Telecommunications Services
Jack-and-plug connector, network
interfaces, 314, 315
Jacking arrangements, 616, 619
Jitter, 239-40
digital carrier system, 378, 382
see also Incidental modulation
JMX jumbogroup multiplex, 368-69
J-number, 614
Journals, 28
J-specification, 614
Jumbogroup, frequency-division
multiplexing, 362, 363, 366, 367,
368-69
Jumper, 549, 550, 551, 553,555-59
diagrams, 553, 557
pictures, 552, 558
Junctor, 416, 417, 431, 448
Junctor circuit, 417n, 419, 421
Justice, Department of, U.S., 4, 689, 691,
692,694,703
Keyboard-display terminal, 480,483,484
Keyboard send-receive (KSR) data terminal,
483
Keyset signal, 117
Key telephone system (KTS), 489, 490-91,
701-3
COM KEY, 490
diagram, 491
1A KTS, 490

Index

1A1 KTS, 490
1A2 KTS, 490
Kingsbury Commitment, 689, 691
Kingsbury, Nathan c., 691
Land mobile telephone service and systems,
518-24
Advanced Mobile Phone Service, 52124
channel availability, 518
frequency band allocations, table, 518
system operation, conventional, 519-20
diagram, 520
transmission, 518-19
Large-scale integration (LSI)
DMS-10 switching system, 431
4ESS switch, 426
station equipment, 469, 479, 489
time-slot interchange, 733
picture, 734
Larynx, artificial, see Artificial larynx
Laser, 21, 22, 205
industrial applications, 15
Last in, first out (LIFO), 151
Last-number dialing, 470
Lateral cables, 123, 124
L-carrier analog carrier system(s), 347-50,
429,534
characteristics, table, 349
L5/L5E coaxial cable systems, 348-50,
368-69
diagram, 349
Lead-acid cell, 543, 546
Legal area (Bell Laboratories), 23,24,26
Legislation
Clayton Antitrust Act, 688, 689, 691
Federal Communications Act, 689, 692
Graham-Willis Act, 689, 691-92
Interstate Commerce Act, 688, 689
Mann-Elkins Act, 688, 689, 691
Sherman Antitrust Act, 688, 689, 691
License contract, 11-13,23,30
Light-emitting diode (LED), 21, 205
DIMENSION PBX, 501
small communications systems, 491
TRIMLlNE telephone, 488
Lightguide cable, 27, 132n, 201, 205-7, 217,
384,437
diagram, 205
Lightguide cable interconnection
equipment (LCIE), 384, 385
Lightning protection, 308, 315
Lightwave digital system(s), 133, 217, 37879,383-84
diagram, 385
Lightwave terminating frame (LTF), 384
Limited access network, 163-64
diagram, 164

Index

Line(s), 86, 333, 397,485
direct progressive control systems, 259
distributing frame, 549
extension telephones, 43
foreign exchange, 88-89
high-frequency, 200
interface, 271, 273
long-distance, 6
noise, 337
number of lines (1983), by equipment
tvoP.4';1

table,J 46i
private-line voice network, 90
facilities and services, diagram, 90
PSTN
facilities and services, diagram, 89
working lines, 125
signaling, 265, 271, 272-76, 284
stored-program control systems, table,
459
switching, 241
2-line telephone sets, 43
in a typical telephone call, 93, 97, 98
see also Loop(s)
Linear distortion, see Distortion
Linear phase-versus-frequency response,
238
Linear transmitter, 469, 470
Line build-out (LBO) network(s), 339, 378,
381
Line circuit, 1/ 1AESS switch, 419
Line fill, 642
Line finder, 164,401
Line group, 629
Line-haul costs
analog carrier systems, 336
DSBAM transmission, 213
Line-insulation testing, 598
Line link frame, 406, 407
Line link network (LLN), 417
Line link pulsing, 407
Line number, see Station number
Line of business, definition of, 712
Line-of-sight propagation path
definition of, 207 n
microwave radio, 207
satellite communications, 209
Line printers, data communications, 46
Line repeater station (LRS), digital carrier
transmission, 384, 385
Line repeater station (LRS), digital carrier
transmission, 384
Link, 169, 397
circuit routing, 138
marker control systems, 260
in network, 81
satellites, 234, 358, 359, 361, 362
Link network (switching network), TSPS,
442

855

List number, J-specification, 614
Litigation
Antitrust (1949), 689, 694
Antitrust (1974), 689, 703
Consent Decree (1956), 689, 694-95, 702,
703
Kingsbury Commitment, 689, 691
see also Justice, Department of, U.S.;
Modification of Final Judgment
Live load, 564
LMX group bank, 365
LMX supergroup bank, 366
Load
balancing, 163-65, 594
basic data, 175
estimate, 182
load-service curves, graph, 165
local automatic message accounting, 448
lost, 159
nonrandom load, 168-69
projection, 174
sharing, stored-program control, 263
see also Offered load
Load Balance Index, 164
Load Balance System (LBS), 624, 625, 628,
629
Load coil, voice-frequency transmission,
333,338
Loading, voice-frequency plant, 333, 334,
338
Local access and transport area (LATA), 31,
33,35
Local automatic message accounting
(LAMA), 399, 447, 448-51
diagram, 450
LAMA-A, 448,453, 454, 455, 456
diagram, 449
LAMA-C, 455, 458
Local calling, 56
Local facilities network, 122-25
planning, configuration, 134-45
Local Message Metering System (LMMS),
2900B,447-48
Local (exchange area) network, 103, 104-7
CClS, 294
class 5 offices, 108
minimum high-usage trunk group size,
172
time-division multiplexing, 220
topology, diagram, 105
typical configuration, diagram, 106
Localoffice(s)
busy season busy hour, 152
VMS-10 switching system in, 430
5ESS switch in, 434
Local service
billing, 31, 56, 445, 698
switching systems, 244
Local switching office, see End office

856

Local Switching Replacement Planning
(LSRP) System, 144
Local switching system(s), 85, 87, 104, 110,
122,397,398-99
basic service, 39
call data accumulator, 454
digital switching technology in, 133
evolution of, 461, 734-35
interoffice facilities network, 125, 128
loops, 328
number identification, 451
number of systems (1983), 461-62
table, 461
1A processor, 414
see also Central office; Crossbar switching
system(s), Panel switching
system(s); Step-by-step
switching system(s)
Local tandem office, 397
interoffice facilities network, 128
switching congestion, 180
Local tandem switching system, 87
Local telephone companies, see Telephone
company(ies)
Logic circuit, 242, 733
Lo-hi N-carrier repeater, 343, 344
Long-distance calling, 11
delay, 234
echo suppressor, 229
market, 710
Long-distance network, see Long Lines
Department (AT&T)
Long-distance telephone service, rate
setting, 699
Long-haul analog microwave radio
system(s),352-56
characteristics, tables, 353, 355
Long-haul interoffice analog carrier
transmission, 213, 347-62
noise objectives, 232
graph,'232
Long-haul interoffice plant, 331-32
Bell System investment, 332
economy of scale, graph, 332
Long-haul intertoll trunk, 428
Long-haul transmission facilities, 130-32,
133,203
Long-haul trunks, 109
Long Lines Department (AT&T), 5, 8, 10, 11,
32,571
Automated Calling Card Service, 655
corporate functions, predivestiture, 6
Economic Impact Study System, 718
facilities, 130
Network Operations Center, 6,627
order negotiation, 574
regions (1982), map, 7
trunk group size, optimal high-usage, 173
Long Range Switching Studies, 144

Index

Long Route Analysis Program, 144
Long-route design, VF loop transmission,
338
Loop(s), 86, 87, 89, 90, 93, 271,272, 393,
465
carrier application, 86, 272, 274, 329
crosstalk objectives, table, 234
equalization, 466
impairments, 234
impedances, 226
linear distortion, 238
local facilities network, 104, 122, 124
loss objectives, 225
maintenance, 598-99
noise objectives, 232
off-hook, condition, 211
paired cable, 124, 202
plant, 123,327-29,393-94,543
cable size, 288, 329
loop length, 288, 328
remote switching system, 87
signaling range extension, 288
station equipment compatibility, 487
time-division switching, 254
transmission systems, 327, 328, 329
analog carrier, 340-41
digital carrier, 370-73
voice-frequency, 337-39
2-wire transmission, 199
voice-frequency system design methods,
338
table, 338
see also Line(s); Signaling, customer-line;
Signaling, interoffice trunk
Loop Assignment Center (LAC), 616, 618,
619n, 621
Loopoack, for testing, 308, 315
Loop Maintenance Operations System
(LMOS),655
Loop-reverse-battery Signaling, see Pertrunk signaling
Loop signaling, 271, 272, 273, 275, 284, 287,
288
diagram, 274
Loop-start signaling, 273, 275, 284,307
network interface, 314
Loop transmission systems (subscriber loop
systems), 86, 543
Loss (insertion loss), 97, 329
analog carrier systems, 336
calculation, diagram, 224n
design methods and, 338
digital data signals, 236
electronic telephone, 470
4-wire analog carrier interface, 296, 297
L-carrier systems, 350
long-haul microwave radio systems, 352
Long Range Switching Studies, 144
Long Route Analysis Program, 144

Index

loop length and, 329
metro area circuits, 330, 339
performance objectives, 225
customer opinion models, 272-73
echo, control of, table, 229
grade-of-service ratings, 674-75
received volume, table, 226
voice-frequency transmission, 333
satellite communications systems, 358,
359
speakerphone, 470
TOUCH-TONE signaling, 276
in transmission path, 227, 329
voice-frequency transmission, 340
Loudness loss, 230, 231
graph,231
Loudspeaker, speakerphone, 48, 470
Low-capacitance (LOCAP) cable, 381
Low-cost telephone service, 698
Low-pass filter
analog carrier transmission, 341, 342
pulse-amplitude modulation, 215
LIT connector, 428, 429
LT-l connector, 365-66
Macro measurements of service and
performance, 676
Magnetic core devices, 236
Magnetic tape(s)
automatic message accounting, 451, 453
4ESS switch, 427
Main radio repeater station (main station),
356
Main station, 114, 115, 125, 312
Maintainability objective, Digital Data
System, 237
Maintenance, 572
customer-service, 79-80, 576, 579-82
diagram, 580
data sets, 479
digital carrier transmission, 378,380,
382
Digital Data System, 533
digital multiplexers, 392
electronic coin public telephone, 472
ESS switches, 420, 422, 427-28,632-34,
637
network, 597-602
carrier system, 600-601
diagram, 598
loop, 598-99
Network Service Center, 597-98
special-services circuit, 601
switching system, 599
trunk, 599-600
new network services, 655
new products, 714
No.1 PSS packet switch, 537

857

operator services, equipment and
facilities, 593
TSPS, 442
PICS, 611, 613
public telephone service, 77, 583-84
station equipment, 488
Switching Control Center, 635, 637
Maintenance span, digital carrier
transmission, 377,382
Maintenance terminal, 420, 442
Mann-Elkins Act, 688, 689, 691
Manually-operated switch, 244
Manual private branch exchange (manual
PBX),493
diagram, 494
Manuals, technical, 28
Manufacturing divisions (Western Electric),
predivestiture, table, 15
Manufacturing information, preparation of,
Bell Laboratories Specific
Development and Design, 23
Marine radiotelephone services, 74-75
Marker(s}, 260-61, 403, 404, 405, 406, 407,
408,409
Market
application of new technology, 729-30
introduction of products and services
into, 708
Marketing, 574, 709-13
concepts, 709-11
definition of, 709
in mature product management, 708
in product development, 707
Marketing mix, 709, 711
definition, 710
Marketing organizations, 574, 575, 712-13
Market management organization, 712
Market penetration, 710
Market research organization, 713
Market segmentation, 711
definition of, 709-10
Market trial, PICTUREPHONE meeting
service, 527
Mass Announcement System (MAS), 513-16
configuration, diagram, 514
event flow, diagram, 516
nodes and islands, map, 517
Master control center (MCC), 632, 633, 635,
636
picture, 633
Mastergroup, frequency-division multiplex,
362,363,364,366,367,368
translator, 356, 366-68
Material and Account Management
Division (Western Electric),
corporate functions,
predivestiture, 16
Material management centers (Western
Electric), 16, 17

858

Materials manager, 611, 612
McDonnell Douglas Automation, 604
Measured call, 445, 446, 447
Measured component, 681-83
Measured-rate billing (measured service),
56-57
Measurement
in service evaluation, 676-84
administration of measurement results,
679-80
future trends, 683-84
plan(s), 658, 677, 679, 680-83
see also Traffic measurements
Mechanization, telephone company
operations, 603, 616
table, 604
see also Computerization
Mechanized Engineering and Layout for
Distributing Frames (MELD), 558
Mechanized network planning, 143-45
planning tools, table, 144
Media promotion service, 65
Media Stimulated Calling (MSC), 9,65
Medium-scale integration, data sets, 479
"Meet me" conferencing, 69
Memory, 542
Automatic Intercept System, 444
call store, 641
DMS-10 switching system, 431, 433
electronic switching systems, 411, 415,
416,419,421,422,426,427,428,
429,431
evolution, 730, 731-33
graphics terminal, 485
No.1 AMARC, 453
No.1 PSS Packet Switch, 537
No. lOA Remote Switching System, 423
repertory dialer, 470
signaling, 286
stored-program control, 262
Transaction telephone, 476
TSPS,441,442
Mercury, Project, 19
Message circuit noise, 230-32
definition of, 230
graphs, 231, 232
objectives, table, 233
Message-rate service, 578
Message register, 447
Messages, as traffic, 81
Message service, short-haul radio systems,
345
Message storage, 47
Message switch, 5ESS switch, 435, 436
Message trunk, 607
Message-waiting light, key telephone
system, 490
Metallic facility signaling systems, 286-89
Metallic facility terminal (MFT), 428, 429

Index

diagram, 341
VOice-frequency transmission, 340
Metallic loop signaling, 272
Metals, recycling, 20
Metro (metropolitan) interoffice plant, 32930
Bell System investment, 332
voice-frequency transmission, 339
Metroliner trains, telephone service, 75
Metropolitan areas, see Urban areas
Metropolitan Area Transmission Facility
Analysis Program (MATFAP), 144,
641-42
Metropolitan interoffice analog carrier
transmission, 341, 343
Metropolitan interoffice digital carrier
transmission, 373-79
see also Coaxial cable; Lightwave digital
system(s); Paired cable
Metropolitan interoffice voice-frequency
transmission, 339-40
Metropolitan junction office, 132
MGT-B mastergroup translator, 356,366-68
diagram, 368
Michigan Bell, 12, 36
Microcomputer, 21, 669
WE 4000, 472
Microelectronics, 22, 730, 733
Bell Laboratories contributions to, table, 21
Micro measurements of service and
performance, 676, 677, 678, 681
Micl'Oprocessor,21
data sets, 479
data terminals, 483, 484
long-haul microwave radio system, 356
No.1 PSS packet switch, 537
No. lOA Remote Switching System, 423
stored-program control, 264
telephone sets, 470, 489
WE 8000, 305, 423, 479
Microwave radio, terrestrial, 6, 133,201,
207-9,345-47,352-56
as high-frequency line, 200-201
station, picture, 208
see also Satellite(s); Waveguide(s)
Microwave station, remote, power systems,
543,547
Microwave systems, private,
interconnections, 311, 693
Military services, automatic voice network,
91,399
Military systems
Bell Laboratories, 24-25, 26
Western Electric, 19
Military Systems Division (Bell
Laboratories),25
Minicomputer maintenance center, 623
Minimum-cost routing, 138,642
Miscellaneous common carrier, 306

Index

MJ mobile telephone system, 519, 520, 521,
523
MK mobile telephone system, 519, 520, 521,
523
M-Iead, see Per-trunk signaling
MMGT-C multimastergroup translator, 356,
368-69
MMX mastergroup multiplex, 366-67
diagram, 367
Mobile telecommunications switching office
(MTSO), 522, 523, 524
Mobile telephone service(s), 26, 40, 73-75,
516-25
land, 74, 518-24
paging, 524-25
see also Advanced Mobile Phone Service;
Paging
Modal dispersion, optical fibers, 206
Model
alternate-route networks, 169-72
area and company, in Bell System
operations plan, 652-53
customer opinion models, 671-73
data traffic, 186-88
economic CCS cost, 173-74
economic impact of a project, 718-26
network planning, 141-45
performance models, 670-71
traffic theory, 148, 155, 156, 158, 161, 168
Modem, 290n, 477,479
Advanced Mobile Phone Service, 524
CCIS,272
electronic cash register, 480
frequency-division multiplex, 363
interfaces, diagram, 320
Modification of Final Judgment (MFJ), 4, 34,
646,666,687,695,703
Bell System marketing structure, effect
on, 712
provisions
major, table, 30
summary, 29-32
Modular distributing frame, 553-55, 558
picture, 555
Modular engineering, 172-73
Modular jack, 616, 619
Modular telephone set(s), 43, 488, 576, 582
Modulation, 197, 210-18,219, 296, 476
definition of, 210, 478
differential phase-shift keying, 218
double-sideband amplitude, 211-14, 218
analog carrier, 341
frequency, 194, 211
long-haul microwave radio, 353, 355
mobile telephone, 519, 521
satellite communications, 359
frequency-shift keying, 197,218
pulse-amplitude, 214-16, 251, 253
DIMENSION PBX, 500, 502

859

pulse-code, 211, 214-17, 220, 251, 253,
425,428
digital carrier system, 337, 370
DSX-1 interface, 299
PBX, 502
PCM-TDM hierarchy, 222-23, 386
signaling bits, 291
quadrature amplitude, 218, 383, 478
single-sideband amplitude, 211, 213-14
analog carrier, 347, 362, 364
long-haul microwave radio, 356
undersea coaxial cable, 350
vestigial sideband, 478
M1C multiplexer, 300, 377, 386, 391, 392
M12 multiplexer, 300, 386, 392
M13 multiplexer, 300, 386, 392-93
Monopoly(ies), 688, 691, 692, 700,701,703
Motion compensation, 528-29
Mountain States Telephone and Telegraph
Company (Mountain Bell), 11, 12,
36
M34 multiplexer, 392, 393
Muldem, 391, 392, 393
Multibutton electronic telephone (MET), see
Electronic telephone
Multichannel analog carrier system, 340-41
diagram, 342
Multichannel carrier system, 201
Multifrequency (MF) signaling, 286, 287,
289,290,426
Multihour engineering, 173
Multimastergroup, frequency-division
multiplex, 356, 362, 367, 369
translator, 356, 368-69
Multimode optical fiber, 206-7
Multipair cable, 84n, 86
analog carrier systems, 343
carrier systems, 337
metropolitan area, 339
VOice-frequency transmission, 333
see also Paired cable
Multiparty service, 57, 274
charge calls, 451
digital carrier system, 372
ringing, 57, 274
Multipath interference (fading), 209, 358,
519,524
Multiplex/ demultiplex terminal, 200
Multiplexers, 222
digital, 370, 371, 392-93
Multiplexing, 46, 136, 139-40, 199,200,211,
218-23,254,255,291,296,335,642
analog signal, 299
channel group, 173
cost tradeoff, diagram, 219
definition of, 218
digital carrier system, 370, 373, 379
Digital Data System, 531, 533
digital data transmission, 529

860

Index

Multiplexing (contd)
electronic switching systems, 413, 426,
429,431
equipment, digital carrier, 386-93
frequency-division, 219-20, 272
analog carrier system terminals, 362-69
crosstalk, 233
hierarchy, 221
diagram, 363
incidental modulation, 240
long-haul microwave radio systems, 353
satellite communications systems, 359
synchronization plan, 221-22
signaling, 283
SSBAM transmission, 213
time-division, 219-20, 251, 253, 272, 373
diagram, 220
digital multiplex equipment, 386-93
DSX-l interface, 299
hierarchy, 221,269-70, 386-87
PCM-TOM diagram, 386
synchronization, 222-23, 375
visual systems, 529
Multiplex (MUX) loop, 431, 432, 433
Multiplex synchronization, 221-23
Multiplex terminal, 200, 239, 362-69
Multipoint networks, 285
data, 91
voice, 90
Multiprocessing
functional, stored-program control, 264
independent, stored-program control, 263
Music transmission, private-line services,
66,67
MX3 multiplexer, 300, 379, 386, 392, 393
MX3C lightwave terminating frame, 384
Narrowband channel, 198
data communications, 44
Narrowband (low-speed) data sets, 478
Nassau Recycle Corporation, 5, 20
National Aeronautics and Space
Administration (NASA), 19
private data network, 91
National CSS, 604
National number, 120
National Television Standards Committee
(NTSC),526
television standards, table, 527
Nationwide paging, 525
Nationwide rate averaging, 697, 698-99
Nationwide telephone network, 305
NATO, see North Atlantic Treaty
Organization
Natural monopoly, 688, 692, 701, 703
Navy, U.S., 19
N-carrier analog carrier system, 330, 343-45,
423,642

Neal-Wilkinson 13 tables, 167, 168, 169
Negative exponential distribution, 156, 157,
158, 160
Net income, 723,726
Net present value (NPV), 723, 725
Network(s), 3,4,9,81,103
access, 11, 30, 31, 39, 666
administration operations, telephone
company, 592-96
capacity, 138-39,586-87
characterization, 664, 665-69
component compatibility, 487
data, 184-91
engineering, 576, 585, 586
facilities, 84-88, 122-34,641-42
implementation, 585, 586-87
interconnection, 305-6, 693-94,701-2
interfaces, 313-16, 666
inter-LATA,31
intraexchange, 666
intrastate, 11
maintenance operations, telephone
company,597-602
diagram, 598
management, 176-80,595-96
controls, table, 181
measurement plans, 677, 680, 681-83
numbering plan, 114-22
operations centers, 647, 650-51
operations planning, 645-62
operations systems, 660-661
performance objectives, 669-676
planning, configuration, 134-45
protection, 310, 311-13, 694, 702
provisioning operations, telephone
company, 585-92
restructuring, postdivestiture, 29-31, 666
SCCS, 632-38
services, 60-76
data, 70-73,529-40
introduction of, 654-56, 705-7
signaling, 193,265-66,282
structures, 103-34
switching, 241
telecommunications, 81-99, 103-45
definition of, 81, 103
TNDS, 621-32
Total Network Operations Plan, 647-48,
651
traffic, 88-92, 103-14,169-76
transmission and switching, 82
diagram, 83
see also Circuit-switching network(s);
Digital facilities network; Facilities
network; Interoffice facilities
network; Local network;
Operations center(s); Operations
system(s); Packet switching;
Private-line data network; Private-

Index

line voice network; Private
switched network; Program
network; Public switched
telephone network; Toll network;
Traffic network(s)
Network Administration Center (NAC),
629, 630, 631
Network channel-terminating equipment
(NCTE), 307, 308, 309, 314, 315
inherent network protection, 313
Network characterization, 665-69
diagram, 667
mechanization of, 668-69
Network control point (NCP), 508, 510, 511,
512,513
Network Data Collection Center (NDCC),
623,627,631
Network delay, data traffic, 188, 189-90
Network interface (NI), 306, 307, 309, 310
FCC registration program, 312
typical interfaces, 313-16
table, 314
Network management, 176-80, 243, 59596
controls, table, 181
signals, 269, 282
Network Management Center (NMC), 179,
595,627
Network Operations Center (NOC), 595,
606,627
Network Operations Center System
(NOCS), 595, 627
Network Operations Report Generator
(NORGEN), 627
Network Operations Trouble Information
System (NOTIS), 597n
Network organization(s), corporate
functions, predivestiture
AT&T, 8, 9-10
telephone company(ies), 14
Network planning, 27-28
Bell Laboratories, 22-23, 25, 26
central services organization, 37
configuration planning, 134-45
components, diagram, 135
telephone company(ies), 585
Network Planning, Subcommittee on
(USITA),27-28
Network Planning and Design
organizations (AT&T), corporate
functions, predivestiture, 8, 9-10
Network protection, 309, 310-13
see also Network channel-terminating
equipment
Network-related operations, 573, 585-602
Network Service Center (NSC), 597-98
Network Service Center System (NSCS),
597n
Network services, 60-76

861

Network Services organization (AT&T),
corporate functions,
predivestiture, 8, 10
Network Switching Performance
Measurement Plan (NSPMP), 680,
681-83
table, 682
Network Systems and Product Planning
Division (Western Electric),
corporate functions,
predivestiture,16
Nevada Bell, 12, 36
New England Telephone, 12,36
New Equipment-Building Systems (NEBS)
standards, 551, 564-65, 567
equipment buildings, pictures, 566
New Jersey Bell, 12, 36
New York Telephone, 12, 36, 669
Nickel-cadmium cell, 359, 543
Nike-Ajax missile defense system, 19
Nike-Hercules missile defense system, 19
911 Emergency Service, 56, 60
Basic 911 (B911), 60
Enhanced 911 (E911), 60
diagram, 61
public telephone access, 77
Nippon Electric, 498
Nobel Prizes, Bell Laboratories, 20, 21, 22
Node(s), 169
circuit routing, 138
Mass Announcement System, 516
map, 517
in network, 81, 84, 86, 88
principal city concept, 120
stored-program control network, 508, 509,
510
test, ASPEN system, 668, 669
Noise, 97, 216, 236, 276, 487
analog carrier transmission, 336, 345
digital carrier transmission, 337, 375, 376,
388
digital data signals, 237
E&M lead interface, 280
impulse noise, 223, 237-38, 239, 375, 376
inductive interference, 235n
land mobile telephone systems, 519
message circuit noise, 225, 230-32, 237
grade of service, graph, 231
objectives
graph,232
table, 233
performance objectives, 235n, 672, 673,
674,675,676
quantization noise, 216, 337, 388
satellite communications systems, 359,
360,361
single-frequency signaling, 289
thermal noise, 21, 230, 360
N oise meter, 230

862

Index

Nonconformance, in quality assurance, 75556
N ondial trunk, 284, 285
Nl carrier, 341
Nonlinear distortion, see Distortion
Nonrandom load, 168-69
Non-sent-paid call(s), 77, 78, 471
Nonswitched private networks
data, 91
voice, 90
North Atlantic Treaty Organization, 19
Northern Telecom, 414
Northwestern Bell, 12,36
N + 1 redundancy, 431
Number group, 405, 406, 407
Numbering plan, 103, 114-22
history and evolution, 117-19
international, 120-21
mobile telephone system, 520
national, postdivestiture, 31
terminology, 115-16
Numbering plan area (NPA), 408n
Automated Calling Card Service, 512
definition of, 116
splitting, 118n
No.1 Crossbar System, see Crossbar
switching system(s)
No.1 PSS packet switch, 535-38
architecture, diagram, 536
No. lilA Automatic Message Accounting
Recording Center, see Automatic
Message Accounting Recording
Center
No. 4A Electronic Translator System, see
Electronic Translator System(s)
No. 4/4A Crossbar System, see Crossbar
switching system(s)
No.5 Crossbar Central Office Equipment
Reports (5XB COER), 625, 628, 629
No.5 Crossbar System, see Crossbar
switching system(s)
No.5 Electronic Translator System, see
Electronic Translator System(s)
No. 5A Remote Switching Module, see
Remote Switching Module
No. lOA Remote Switching System, see
Remote Switching System(s)
Number service(s), 443-45
Number-service operators, 62
NYNEX Corporation, 36
Nyquist interval, 215
Occupancy
definition of, 159
space-division switching network, 246
Off-axis energy, 209
Offered load, 149, 154, 159, 160, 166, 168
cluster busy-hour technique, 173
definition of, 148

management of, 177-79
graph,176
measurement, 182-83
reporting process, 630
Off-hook, 85, 92, 97, 211, 259, 401
electronic coin public telephone, 472
electronic switching systems, 412, 419
mobile telephone service, 524
No. lOA Remote Switching System, 423
Signaling, 265, 267, 268, 271, 272, 278,
289, 290, 291
E&M lead, 278
Office carrier module (OCM), 432, 433
Office channel unit (OCU), 531
Off-premises station (OPS), 307
Ohio Bell, 12, 36
OKI Electric, 498
lAESS switch, see Eledronic switching
system(s)
lA-Radio Digital System (IA-RDS), 382,
533
lESS Switch, see Electronic switching
system(s)
195 Broadway Corporation (AT&T), 5,10
101ESS switch, see Electronic switching
system(s)
One-way trunk, 95, 277-78
On-hook, 85, 92, 99
electronic switching system, 412, 419
No. lOA Remote Switching System, 423
Signaling, 265,267,268,271,272,278,
280,289,291
E&M lead state, 278
On-line system(s)
PREMIS, 621
quality control, 742
TIRKS, 610
TNDS, 625
Open Systems Interconnection (051),
Reference Model for, see Reference
Model for Open Systems
Interconnection
Open-wire line(s), 128, 201
Operations, telephone company, 571-602
customer-related, 573-85
administration, 577 -79
directory service, 584-85
maintenance, 579-82
provisioning, 573-77
public telephone service, 582-84
network-related, 585-602
administration, 592-96
maintenance, 597-602
provisioning, 585-92
Operations center(s), 572, 573, 607, 648,
650-51, 652, 656, 657, 658, 659,
660
billing, 577-78
definition of, 647
packet-switching systems, 538

Index

see also Network Administration Center;
Network Data Collection Center;
Network Management Center;
Network Operations Center;
Network Service Center
Operations planning, 23, 26, 573, 64562
Operations plans
elements, 648-54
model areas and companies, 652-54
operations centers, 650-51
operations processes, 652, 653
operations systems, 651-52
initial efforts, 646-47
operations systems, impact on
development of, 659-60
operations systems network, 646, 660-61
telephone company(ies), impact on, 65459
improving operations, 654
operating new network services, 65456
operations planning in telephone
companies, 656-59
Operations process, 649, 650, 652, 654, 656
definition of, 648
Operations system(s), 26, 243, 572, 573, 584,
603-43,645,646,647,648,650,
651-52,654,655,656,657,658,
677n, 707, 715
coin collecting and counting, 583
development of, 23, 642-43, 659-60
engineering, 639-42
equipment administration, 622-32
equipment maintenance, 269, 538, 632-38
order-processing, 616-21
planning, 641-42
recordkeeping,605-15
support, 642-43
see also Automated Trouble Reporting
System; Bell Administrative
Network Communications System;
Centralized Automatic Reporting
on Trunks; Centralized Message
Data System; Centralized Results
System; Central Office Equipment
Engineering System; Computer
System for Mainframe Operations;
Engineering and Administrative
Data Acquisition System;
Engineering and Administrative
Data Acquisition System/Network
Management; Facilities
Assignment and Control System;
Local Switching Replacement
Planning System; Long Range
Switching Studies; Long Route
Analysis Program; Loop
Maintenance Operations System;
Metropolitan Area Transmission

863

Facility Analysis Program;
Network Operations Trouble
Information System; Network
Service Center System; Outstate
Facility Network Planning System;
Plug-In Inventory Control System;
Premises Information System;
Switched Access Remote Test
System; Switched Maintenance
Access System; Switching Control
Center System; T-Carrier
Administration System;
Telecommunications Alarm
Surveillance and Control System;
Total Network Data System; Trunk
Servicing System; Trunks
Integrated Records Keeping
System
Operations Systems and Network Planning
area (Bell Laboratories), 24, 25
Operations systems network, 265, 646, 654
diagram, 661
interfaces, 320
planning, 660-61
Operator
force engineering, 161-62
functions, 111-13
see also Operator service(s)
Operator number identification (ONI)
Automatic Intercept System, 444
CAMA-ONI, 62,111,112,451
Operator service(s), 10,88,243,452
administration, 593
Automatic Intercept System, 111, 112,
443-45
dialing, 117
directory assistance, 62, 111,584
forcing, 161-62
provisioning, 589-90
PSTN,62,111-13
public telephones, 77
stored-program control network, 508, 509,
511
toll services, 438, 443
TSPS,77,111-12,440-43,508-9,511-12,
513
Operator systems, 438-45
number services, 443-45
toll service, 438-43
see also Automatic Intercept System;
Traffic Service Position System
Optical fiber(s), 6, 132,205-7
diagram, 206
5ESS switch, 434
Optimum cost standard, 750
Orbit, satellite communications systems,
209,356,357,359
diagram, 357
satellite eclipse, 359-60
sun transit outage, 359, 360

864

Order completion, 576-77
Order generation, 575-76
Order negotiation, 574-75
Order processing, 605-21
PICS, 611-15
PREMIS, 616-21
TIRKS,605-11
Order turret, 504
Ordinary service(s), definition of, 40
Organization of the Bell System, see Bell
System
Originating register, 94, 95, 406, 407
see also Register
Other common carrier (Ocq, 40
interconnection, 305-6,308-11
diagram, 309
interface (demarcation point), 306-7,
311
Outage
data circuits, 237
FTC3 digital lightwave system, 384
sun transit, satellite communications
systems, 359, 360
Outgoing calls, business customer services
for, 52, 53, 54, 55
Out-of-band signaling, 286, 290-91
Outpulsing, 96, 97, 408, 419
Outpulsing register, 94, 95, 96, 97
see also Register
Outside plant, 23n, 26, 272n, 328
dedicated plant assignment card, "621
energy storage, 543
human factors in design of, 741
Outstate Facility Network Planning System
(OFNPS), 144, 642
Outstate (rural) interoffice analog carrier
transmission, 343-47
Outstate interoffice digital carrier
transmission, 379-81
Outstate interoffice plant, 330-31
Bell System investment, 332
distribution of system length, graph, 331
Outstate interoffice voice-frequency
transmission, 340
OUTWATS, see Wide Area
Telecommunications Services
Overflow, 168, 171, 172, 173, 174, 176, 182,
623
Overload, 155, 164
control of, 163, 596
network congestion, graph, 178
network interface, ~16
reattempt effect, 168
transmission impairment, 236
VOice-frequency transmission, 339
Pacific Bell, 5n, 12,36,655
Pacific Northwest Bell, 12,36

Index

Pacific Telesis Group, 36
Packet, 71, 185, 187, 188, 293, 535
Packet switch, 535
No.1 PSS, 72, 530, 535-38
diagram, 536
Packet switching, 21, 188, 293
Basic Packet-Switching Service, 71-72,
535
networks, 72, 185, 187-88,457,530,535,
666
data performance concerns, 188-90
interfaces, 320
traffic engineering concepts, 190-91
systems, 187-88,530,535-38
Paging
BELLBOY radio paging system, 74, 524-25
diagram, 525
HORIZON communications system, 492
Paired cable, 84, 124, 201-3
digital carrier systems, 374-77
digital facilities network, 132
interoffice facilities network, 128
picture, 202
voice-frequency transmission, 330
see also Multipair cable
Pair gain, analog carrier transmission, 340
Pair-gain system(s), 370
digital facilities network, 134
local facilities network, 124
Panel switch, 244
Panel switching system(s), 399, 400, 735n
billing, 447
Paper tape, automatic message accounting,
448,449,450,451,452,453,454
Parallel transmission, 477-78
data sets, 478, 480
Partial-dial abandon, 161
Patents
A. G. Bell's, 687, 689, 700
Bell Laboratories, 20
royalties and licensing, 695
Patron, 308
Peakedness, 168, 169, 172, 182
Peak load, 151,153
Peak-to-average ratio (P / AR) measurement,
239
Peg count, 182,623
Peg count and overflow (PCO), 182
Pending-order file, 575-76
Penzias, Arno, 22
Percent break, 275
Per-channel signaling, 291
Performance indicator(s), 679
l/lAESS NSPMP, 681-83
Performance measurement(s), 665, 668, 67678
Performance model(s), 670-71
Performance objective(s), 665, 669-76, 684
Performance planner, 666-67

Index

Peripheral bus computer (PBC), 408, 627n
Peripheral unit controller, 305
Per-line signaling, see Per-trunk signaling
Permanent magnet twistor, 415
Permanent virtual circuit capability, 72
Personal identification number (PIN), 476,
512
Personnel and Public Relations area (Bell
Laboratories), 24, 25, 26
Per-trunk signaling, 271, 277-80, 282, 294
E&M lead, 271, 273, 277, 278-80, 281, 282,
283,284,287,289,290
loop-reverse-battery, 273, 277-78, 280,
283, 288, 289, 290
Phase hits, 240
see also Incidental modulation
Phase-shift keying (PSK), data sets, 478, 481
Phone Center Store(s), 616
see also Bell PhoneCenters
Photonics, Bell Laboratories contributions
to, 21, 22
PICTUREPHONE meeting service, 75-76,
527-28
human factors in design of, 740
picture, 76
PICTUREPHONE visual telephone service,
75,526-27
television standards, table, 527
Pilot tone
analog carrier system, 600n
A-type channel bank, 365
coaxial carrier system, 350
long-haul microwave radio system, 356
Planning
Bell System operations, 645-66
circuit provisioning, 607, 609-10
network characterization studies, use of,
665-69
network configuration, 134-45
mechanized tools, 143-45
performance, 665-69
telephone company function, 585-86,
638-42
Planning tools, design of, mechanization in,
143-44
Plant, 26, 619
definition of, 23n
transmission systems, 327-32
see also Outside plant
Plug-In Administrator (PIA), 612, 614
Plug-In Inventory Control System (PICS),
605, 611, 643, 714
Plug-In Inventory Control System with
Detailed Continuing Property
Records (PICS/DCPR), 611,612-15
benefits, 615
implementation, 612-15
interfaces, 615
subsystems, 613-15

865

Plug-in unit, 340, 343, 375, 548, 601
PICS/DCPR,611-15
Point-to-point circuit, 138, 140
Point-to-point forecasting method, 588
Point-to-point network, 90
Poisson Blocking Formula, 166, 168
Poisson distribution, 155, 156, 157
data traffic, 186
Poisson process, 156, 186
POLARIS, data base management system,
669
Polarity guard, 469
Polar Signal, 374, 377
Police emergency service, see 911
Emergency Service
Position engineering, 589
Postdialing delay, 270, 280, 672, 676
Post Office Department, U.S., 692
Postpay coin service, 473
Posttax cash flow, 719, 720-21
Power control circuits, 66
Power failure, 542, 546, 547, 548, 549
Power hum, induced, 230, 240
see also Inductive interference
Power-line hum, 223
Power system(s), 541-49
customer-premises equipment, 549
dc-to-ac inverter, 548-49
dc-to-dc converters, 548
diagram, 545
energy sources, 543
energy storage, 543-44
operation, 544-47
power plant.
output voltage, table, 547
picture, 546
rectifiers, 547-48
Preferential assignment, 556, 558-59
picture, 558
Prefix, address, 115, 117
definition of, 116
international numbering, 121
Premises Information System (PREMIS),
605,616-21
applications, 617-20, 655
characteristics, 621
data bases, diagram, 618
picture, 618
PREMIS/LAC, 616, 621
Premium telephone sets, 41
Prepay coin service, 472-73
Prep lanning, 596
Present-status inventory, in operations
planning, 657
Present worth, 140, 591, 640, 721, 722
Pretax funds, 719-20
Pretrip, 313
Preventive maintenance, 579, 597
Primary center, 108, 109

866

Primary frequency supplies, 222
Primitives, switching, 64, 510
PRINCESS telephone, 41, 582, 713, 714
picture, 466
Principal city concept, 120
Printed circuit, DMS-10 switching system,
431
Printer
data terminals, 482, 483, 484
graphics terminal, 485
Privacy
DATAPHONE Select-a-station service, 538
telephone answering systems, 505, 506
Private branch exchange (PBX), 27, 40, 54,
55,58,87,161,307,407,489,701-2
automatic call distributor, 503, 504
crossbar, 494-98
diagram, 496
756 PBX, 496
757 PBX, 496
770 PBX, 496
dial, 493-494
dialing procedure, 117
digital, diagram, 502
digital carrier transmission, 373
DIMENSION PBX, 497, 499-503
diagram, 500
features, 501
electronic, wired-logic, 498-99
800 PBX, 498
800A PBX, diagram, 498
801 PBX, 498
805 PBX, 498
812 PBX, 498
fields of use, graph, 497
manual,493
diagram, 494
power system, 549
private network services, 67-68, 69, 70
private switched networks, 113
special-services signaling, 284
step-by-step dial
diagram, 495
701-type, 494, 611
740-type, 494
supervision, 85
vintage PBX, 493-99
Private branch exchange (PBX) tie trunk, 90
Private branch exchange (PBX) trunk, 88
Private line(s), 44, 45
data sets, 477
data terminal, 483
data transmission, 530
Private-line circuit, 89-90
Private-line data network, 91
Private-line data service, signaling, 285
Private-line service(s), 40, 66, 285
data communications, 44-45
DA TAPHONE Select-a-station service,
538-40

Index

digital carrier transmission, 373
FCC Part 68 Rules and Regulations, 312
inherent network protection, 313
marketing, 710
other common carriers, 306
signaling, 265, 285
table, 67
see also Data network services
Private-line voice network, 89-91, 92
Private microwave systems, 693
Private network service(s), 40, 61, 66-70
Private switched network, 90
data transmission, 530
structure, 113-14
Prizes and awards, Bell Laboratories, 20, 21
table, 22
Process cooling, 563
Processor
Automatic Intercept System, 444
electronic switching systems, 410-11
DMS-10 switching system, 431
lESS switch, 415-16
2ESS switch, 414, 421
2BESS switch, 414, 421
3ESS switch, 422
4ESS switch, 426
5ESS switch, 434-36, 438
No.1 PSS packet switch, 536
lA, 416,426
stored-program control, 261, 262, 263,
264
stored-program control switching
systems, table, 459
TSPS, 439, 441,442,443
Product development, 710
see also Service evaluation
Product development organization, 713
Product evolution, see Service evolution
Product life cycle, 709, 710-11
diagram, 711
Product line, 712
Product management organization, 713
Product mix, 712
Product modulation, 212, 215
definition of, 212
Product Performance Survey (PPS), 759-60
diagram, 759
Product portfolio management, 712
Product positioning, 712
Program channel, 198
Program for Arrangement of Cables and
Equipment (PACE), 558
Program network, 92
Program signal, 194-95
Program store(s), 262, 415, 416, 421, 431
Program store bus, 415, 416
Progressive control, 398
control mechanisms, 258-59
space-division switching network, 246-48
step-by-step switching systems, 402

Index

Project, in product development, 713-16
classification, diagram, 715
economic evaluation
calculating economic measures, 718-26
diagram, 716
preparing input data, 717-18
presenting and interpreting results,
727-29
Propane-fueled thermoelectric generator,
543
Protection channel, 237
Protection switching, 240, 548
digital carrier transmission, 373, 378, 379,
380,381,391
JMX terminal, 369
long-haul microwave radio systems, 353,
355,356
MMGT-C multimastergroup translator
terminal, 369
outstate networks, 330
Protective coupling device, 694, 702
Protector frame, 554
Protector unit
distributing frame, 551
picture, 552
Protocol,272
data communications, 316-23
diagram, 318
definition of, 316
packet-switching system, 536, 537, 538
Provisioning, 571, 573
circuit, TIRKS, 605-9, 610
diagram, 606
customer service, 573-77
diagram, 574
equipment engineering, 576
installation, 576
order completion, 576-77
order generation, 575-76
order negotiation, 574-75
definition of, 571
network, 585-92
common systems, 590-92
diagram, 591
diagram, 586
operator-services, 589-90
switching equipment, 588-89
transmission facility, 588-89
trunk, 587-88
public telephone service, 582-84
Public Announcement Service (PAS), 9, 65,
513
Publications, 28, 667
Public Communications Services, 76-78
see also Public telephone(s); Service(s),
public
Public Relations area (Bell Laboratories), 24,
25,26
Public safety answering point (PSAP), 60
diagram, 61

867

Public Services organization, corporate
functions, predivestiture
AT&T, 8, 9
telephone company(ies), 14
Public switched data networks, interface,
308
Public switched telephone network (PSTN),
3,87,88-89, 198,226,241,398,
606
air / ground service, 74
construction program, 622
data transmission, 184,482,530
data sets, 44, 477, 479, 480, 482
data terminals, 45, 483
data traffic, 190
equipment utilization, 89
GEMINI 100 electronic blackboard
system, 484
high-speed train telephone service, 75
linear-distortion objectives, 238-39
numbering plan, 115-22
performance model, 671
private networks, facilities shared with,
89-91,113,329-30
services, 61-66
service systems, 507-16
structure, 103-10
local network, 104-7
toll network, 107-10,397
switching hierarchy
diagram, 108
offices, distribution, table, 109
telephone call, typical, 92-99
traffic, 89, 147, 151-52
transmission objectives, 224
see also Automated Calling Card Service;
Circuit-switched digital capability;
DIAL-IT network communications
services; Direct services dialing
capability; Final trunk group(s);
Foreign exchange service; Highusage trunk(s); High-usage trunk
group(s); Interface(s), internal;
Mass Announcement System;
Operator service(s); Storedprogram control network; Wide
Area Telecommunications
Services; Wire center
Public telephone(s), 76-78, 243, 471-75
A-type, 471, 472
coin collecting and counting, 583,646
coin-handling functions, 474-75
C-type, 471, 472, 474
digital carrier transmission, 372
D-type,472
maintenance, 583-84
picture, 78
provisioning, 582-83
supervisory Signaling, 275
system operations, 472-74

868

Public telephone(s) (contd)
TSPS,440,443
operator functions, table, 440
see also Automated Calling Card Service;
Automated Coin Toll Service;
Charge-a-Call; Electronic
telephone
Public telephone service, see Service(s)
Public utilities commission (PUC), 692
tariff on new services, 708
Pulse-amplitude-modulated switching
network, 250-51
Pulse-amplitude modulation (PAM), see
Modulation
Pulse-code modulation (PCM), see
Modulation
Pulse correction, 290
Pulse dispersion, digital data signals, 238
diagram, 238
Pulse distributor
1/IAESS switch, 416,418,419
TSPS,442
Pulse rate, 197, 199
see also Bit rate
Pulse stuffing, 222
Purchased Products Engineering (Western
Electric), corporate functions,
predivestiture, 17, 19
Purchased Products Inspection (Western
Electric), corporate functions,
predivestiture, 17, 19
Purchasing and Transportation Division
(Western Electric), corporate
functions, predivestiture, 16, 19
Pushbutton keypad, 85, 117,275,468,470,
736
dialing frequency groups, diagram, 276
see also TOUCH-TONE dialing
Quadrature-amplitude modulation (QAM),
see Modulation
Quadrature distortion, 213
Quality assurance, 26, 742-60
AT&T,743
auditing, 745-56
definition of, 742
evolution of products and services, role
in,742-45
monitoring, 756-60
telephone company(ies), 743, 749, 755,
756,757,758,759,760
Western Electric, 14,25,742,743,745,
757,758
Quality assurance audit, 743, 744, 745-56
defects, table, 748
nonconformance, 755-56
structure, 745-55

Index

Quality Assurance Center (Bell
Laboratories), 25, 743,745,747,
749,755,757,758,759
Quality assurance monitoring, 743, 744,
756-60
engineering complaints, 758-59
field performance tracking, 759-60
field representatives, 757-58
surveillance, 756-57
surveys, 757
Quality assurance system, 742, 743
diagram, 744
Quality control, 742, 755, 757
definition of, 742
Quality Measurement Plan (QMP), 752
Quality standard(s), 749
Quality Surveillance System (QSS), 743, 745
Quantization noise, 216, 337, 388
Quantizing, pulse-code modulation, 215
Queuing theory, see Traffic theory
Radio
digital radio systems, 382-83
high-frequency, 28
long-haul microwave radio systems, 35256
mobile telephone systems, 516-25
modulation, 211
paging systems, 524-25
satellite communications systems, 209-10,
356-62
short-haul microwave radio systems, 34547
terrestrial microwave radio, 207-9
waveguide, 204-5
Radio broadcasting, 92,194-95
Radio-frequency (RF) channel
long-haul microwave radio systems, 352,
353,355,356
short-haul microwave radio systems, 345,
346,347
Radio-frequency interference, 223
Rain attenuation, 354
satellite communications systems, 210,
360
terrestrial microwave radio, 209, 383
Random-access memory (RAM), 731, 732
Random traffic, 156
Range extender, loop plant, 289, 338
public telephone systems, 475
Range extension, loop plant, 338
Rate-and-route operators, 62, 111, 112-13
Rate base, 699
Rate setting, 697-99
RCA Global Communications, 40
Read-only memory (ROM), 422
Ready signal, 96

Index

Real-estate management services, 10
Reattempts, 167-68
Received-tone detector, 289
Received volume, transmission objectives,
224-25
table, 226
Receive-only (RO), data terminal, 483
Receiver
land mobile telephone systems, 519
lightwave systems, 378
long-haul microwave radio systems, 352,
353,354,356
paging systems, 524-25
satellite communications systems, 359
telephone set, 85, 97, 465, 466, 467, 468,
469,470,471
Recordkeeping, 605-21
PICS, 611-15
PREMIS, 616-21
TIRKS, 605-11
Rectifier, 545, 546, 547-48
customer-premises equipment, 549
Recycling of metals, 20
Redundancy
interoffice facilities network, 132
power system, 542, 548, 549
Reed matrix switch, 244, 248, 498
Reference file subsystem, PICS/DCPR, 613,
614-15
Reference Model for Open Systems
Interconnection (051), 316-19
diagrams, 317, 318, 320, 322
Reference signals, 221
Refraction, index of, definition, 206
Regenerator, digital carrier system, 375, 377,
378,381,382,383,384
see also Repeater(s)
Regional Bell operating companies
(RBOCs), 33, 34, 35, 37
resources, table, 36
territories, map, 36
Regional center, 108
Regional frequency supplies, 221
Regional Services Company, 35
Register, 117,259,260,261,262,268,406,
407,410, 577, 641
TOUCH-TONE dialing, 736, 737
see also Incoming register; Originating
register; Outpulsing register;
Ringing register; Sender
Registration program, see Regulation
Regulation, 687-95
"Above-890" Ruling, 689, 693
Computer Inquiry 1,689,702
Computer Inquiry II, 10,32, 488n, 574n,
689,702-3
Federal Communications Act, 689,
692

869

interconnection
Carterfone Decision, 310, 311, 479, 689,
694,701
Hush-A-Phone Decision, 689, 694
Interstate Commerce Act, 688, 689
Interstate Commerce Commission, 688,
689
Mann-Elkins Act, 688, 689, 691
Part 68 Rules and Regulations, FCC, 31213,487
public utilities commissions, 692, 708
registration program, FCC, 307, 311-12,
689, 694, 702
Specialized Common Carrier Decision,
689,693
see also Federal Communications
Commission; Tariff(s)
Regulatory measures, in project evaluation,
719,723-26
Relay(s), 261, 419,549
Reliability, 597
performance objectives, 676
undersea coaxial cable systems, 351
signaling, 294
workmanship and, 747
Remote carrier module (RCM), 432, 433
Remote Data Entry System (RDES), 642
Remote digital switching system, 133
Remote equipment module (REM), 432, 433
Remote Memory Administration System
(RMAS),643
Remote metering, telegraph channels, 91
Remote switching module (RSM), No. SA
RSM,130,434,437
Remote switching system (RSS), 87
No. lOA RSS, 130,422-25
diagram, 424
Remote testing, 582
digital carrier transmission, 373, 375,378
4ESS switch, 428
Remote test module (RTM), 668, 669
Remote trunk arrangement (RTA), TSPS,
443,473
Remreed switch, 418, 421, 422
Reorder tone (fast busy tone), 94
switching congestion, 177, 268
trunk congestion, 106, 109, 149, 166, 177
Repair, see Maintenance
Repeater(s)
analog carrier systems (amplification), 336
L-carrier, 348-50
diagram, 349
multichannel,341
diagram, 341
N -carrier, 343
diagram, 344
undersea coaxial cable systems, 350-51
VOice-frequency cable systems, 334, 339

Index

870

Repeater(s) (contd)
digital carrier systems (regeneration),

270n,337
cable systems, 377, 381-82
FT3lightwave system, 207, 378-79
radio systems, 383
SLC-40 carrier, 372
T-carrier systems, 372, 375, 376, 377,
378,379-80
pulse-code modulation, 217
satellite communications systems, 210,
356,358,359
short-haul microwave radio systems, 34547
Repeatered line
analog carrier systems
L5/L5E coaxial carriers, 348-50
multichannel, 341
diagram, 342
N-carrier, 343-45
diagram, 344
digital carrier systems, 370, 371
T1 carrier, 376, 377
T1/0S, 378, 380
T4M coaxial cable system, 377
Repeater station, long-haul microwave
radio,208,352-56
alarm, 600
diagram, 354
power, 541, 547
Repertory dialer, 47-48, 470
Transaction telephone, 49, 476
Required earnings, 724
Rerouting, 181
Research and development, Bell
Laboratories, 25
Research and Development Planning area
(Bell Laboratories), 23, 24
Research and Systems Engineering (R&SE),
Bell Laboratories, 22-23
Research area (Bell Laboratories), 23, 24
Reserve energy storage, 542, 543-44,545,
546
Residence Account Service Center (RASC),
653
Residence customer, service and
performance expectations, 677
Residence Customer Billing Inquiry
Process, 578, 652
functional diagram, 653
Residence installation questionnaire,
TELSAM,680-81
Residence Marketing Sales and Service
organization (AT&T), corporate
functions, predivestiture, 8, 9
Residence organization(s), corporate
functions, predivestiture
AT&T, 8, 9
telephone company(ies), 14

Residence products, manufacturing of,
table, 15
Residence service, 39, 40
Resistance design, VF loop transmission,
338
Resistor, 467
Retail sales, 43, 79
Retained earnings, 726
Revenue Accounting Office (RAO), 447, 448,
449,450,451,453,577-78
Revenue requirements, 723-25
Revenue requirement taxes, 724-25
Revenues, 719
Revised resistance design, VF loop
transmission, 339
Ring conductor, see Ring lead
Ringdown, non dial trunk, 285
Ringer, telephone, 465, 469, 488
Ringing detector, 469
Ringing key, 285
Ringing register, 97, 98
Ringing signal, 97, 267, 268, 274, 402, 412,
419,425
see also Signaling
Ringing tone, see Audible ring
Ring(R) lead,274n, 275,278,296n, 307,315
see also Tip lead
Ring tripping, 274
Roosevelt, Theodore, 691
Ross, Ian M., 729
Rotary dial, 39,41, 85, 117, 150,328,468
Call Forwarding, 121
compatibility requirements, 487
pulsing times, 160
Rotary switch(es), 244,246
diagrams, 247
and equivalent coordinate switches,
diagrams, 250
Route, definition of, 330
see also Circuit(s); Span
Royalties, patent, 695
Rural (outstate) areas, 338
DMS-10 switching system, 414, 431
interoffice facilities network, 128-30
diagram, 129
Outstate Facility Network Planning
System, 642
outstate interoffice networks, 330-31
analog carrier transmission, 343-47
digital carrier transmission, 379-81
VOice-frequency transmission, 340
table, 125
wire center parameters, 124-25
Safeguard antiballistic missile system, 19
Safety, station equipment design
considerations, 488
Sales force, 713

Index

Sales support, and service provisioning,
574,575
Sampling, in quality assurance, 745, 751
universal sampling curve, 753
Sampling gate, channel bank, 387
Sandia Corporation, 5, 20
SATCOM domestic satellites, 357
Satellite(s), 6, 28,201, 209-10
communications systems, 133,356-62
COMSTAR, 357, 360
delay, 234
guidance equipment, missile-borne, 19
High-Capacity Satellite Digital Service,
73
INTELSAT I (Early Bird), 358
Telstar 3, 357, 358, 360
Satellite eclipse, 359-60
SA undersea coaxial cable system, 350
SBS-l domestic satellite, 357
SB undersea coaxial cable system, 350, 351
Scanner
Electronic Translator System, 455
1/IAESS switch, 418, 419
stored-program control, 262
TSPS, 441, 442
Schawlow, Arthur, 22
Science, Bell Laboratories contributions to,
table, 21
Scoring dass, in quality assurance, 746
"S" Data Analysis System, 669
SD undersea coaxial cable system, 351
Sectional center, 108
Security, 71, 583,615, 638
Seizure, 94-98,268,272,275,282,284,289,
406, 412, 439
see also Off-hook
Selective dynamic overload controls
(SDOC),181
Selector control unit (SCU), 540
Selector switch, in SXS system, 401, 402
Self-timing, Tl lines, 376
Semiconductor memory, 421, 731-33
Semipublic telephone, 76-77
Sender,399,404,406,408
Sensitivity analysis, 143, 640
Sensor(s), automatic message accounting,
453,454
Sent-paid call(s), 77, 471
Separate ownership, interconnection and,
306
Separations, in rate setting, 697, 699
Serial transmission, 477
data sets, 478, 480
Service(s),39-80
administration of, 577-79
basic, 39, 702
customer support, 78-80
customer switching, 52-56
enhanced,316-17,702-3

871

exchange,30-31,56-60
exchange access, 30-31, 35
handicapped, aids for, 49-51
interexchange, 30-31
interexchange carrier, 31
inter-LATA telephone, 31
international telephone; 28, 31, 115, 234
No.5 Crossbar System, 407
signaling, 270, 286, 293
TSPS, 439
interstate, 11,68,69,699
intra-LATA telecommunications, 35
intrastate, 699
local telephone, 31
network,60-76
new, growth of, 666
ordinary and special, 40
provisioning, 573-77
public, 9, 31, 40, 76-78,582-84
rates, 697-99
vertical, 40, 41-43
see also Service evaluation; Special
service(s); Terminal equipment
Service access unit, 493
Service administration, 572, 577-79
Service centers (Western Electric), 16, 17
Service circtiit(s)
Automatic Intercept System, 444
DMS-10 switching system, 431
1/IAESS switch, 416, 417
function, 419-20
TSPS, 441
Service code, 119
definition of, 116
Service development organization, 713
Service evaluation, 663-84
concept, 663-65
diagram, 664
measurement and control, 676-83
network characterization, 665-69
performance objectives, 669-76
Service Evaluation System, No.2, 683
Service evolution, 705-60
diagram, 706
economic evaluation, 713-29
human factors, role of, 737-42
marketing, 707,708,709-13
new systems, integration of, 734-37
new technology, application of, 729-34
quality assurance, 742-60
Service management organization, 713
Service measurement(s), 676-78
Service objectives(s), 180, 183-84, 611, 665,
666,669-76
Digital Data System, 237
formulation of, 675-76
Service order, 575, 579, 582, 584, 616, 617,
620
Service-problem analysis and correction, 594

872

Service representative, 574,578,616,617,
619,620,621
picture, 618
Services of the Bell System, see Bell System
Service verification, 581
Serving area interface, 123, 328
Settlements, in rate setting, 697, 699
SF undersea coaxial cable system, 351
SG undersea coaxial cable system, 350, 351
Shadowing, in radio transmission, 519
Share Owners Newsletter, AT&T, 726

Sheaths
lightguide cable, 205
paired cable, 202-3
Sherman Antitrust Act, 688, 689, 691
Shewhart, W. A., 743n
Shielding systems, telephone equipment
building, 561
Shockley, William, 22
Short-haul analog carrier system(s), 201,
341-43
noise objectives, 232
graph, 232
Short-haul analog microwave radio
system(s), 345-47
characteristics, table, 346
Short-haul digital transmission, 374
Short-haul intertoll trunk, 428
Short-haul toll call, billing, 447
Short-haul transmission facilities, 132,
133
Sideband, 364-65, 367, 368
double-sideband amplitude modulation,
211-13,214,218
single-sideband amplitude modulation,
211,213-14
Sidetone, 225, 466-68
Signal(s), 198,299
analog carrier systems, 336
N-carrier,345
degradation, causes of, 223
digital carrier systems, 337
T1 carrier, 374
durability, in digital transmission, 132
interface, 313-16
lightwave, 383
loops, 328
modulation, 210-11
switching system, 242
types and characteristics, 193-98
typical telephone call, 92-99, 265, 267-68
see also Analog signal(s); Carrier Signal,
modulation of; Digital Signal;
Signaling
Signal conversion unit, 280
diagram, 282
Signal distributor
l/lAESS switch, 418-19
stored-program control, 262
TSPS,442

Index

Signaling, 84, 242, 265-323
addressing, 85, 269
customer-line, 265, 269, 272-76,288, 289,
296
definition of, 242, 265
digital channel bank, 388
functions, telephone set, 85, 92-99, 26669,468-69
fundamental considerations, 269-70
implementation, table, 287
in terfaces( s)
data communications, 316-23
definition of, 266
interconnection, 305-16
internal, 294-305
loop carrier, diagram, 274
Signaling between, diagrams, 279, 281,
282,283
systems applications and, 270-86
table, 27'3
interoffice trunk, 276-84
common-channel 280-84
diagram, 277
per-trunk, 277-80
metallic facility terminal, 340
modes
compelled, 286, 287
continuous, 286, 287
spurt, 286, 287
performance objectives, 676
private-line voice network, 90
special-services, 40, 284-86
supervision, 85, 269, 328
techniques, 286-94
CCIS, 292-94
dc, 286-89
inband, 289-90
out-of-band, 290-91
over digital facilities, 291-92
Signaling range extension, 288
SIGNALMAN relay switch, 51
Signal polarization, 204
Signal processor, 389, 415-16, 428
Signal-to-noise ratio
digital carrier transmission, 337
L-carrier system, 348
long-haul microwave radio system, 355
message circuit noise, 231
N -carrier system, 345
pulse-code modulation, 216
satellite communications systems, 360
undersea coaxial cable systems, 351
Signal transfer point (STP), 283, 292, 293,
294,508,509,511,512,596
Simulation techniques
congestion analysis, 178-79
stored-program control systems, 163
Singing, 199
speakerphone, 471
voice-frequency transmission, 334, 339

Index

Single-channel analog carrier systems, 340
Single-frequency (SF) inband tone
signaling, 284, 286, 287, 289-90
Single-frequency (SF) signaling unit, 296,
297
Single-mode optical fiber, 206
Single-Number Service, 63
Single-party service, 57
loop-start signaling, 284
ringing, 274
SLC-96 carrier system, 372
Single-sideband amplitude modulation
(SSBAM), see Modulation
SKYNET 1.5 Service, 73n
see also High-Capacity Satellite Digital
Service
SLC-40 carrier system, 372
SLC-96 carrier system, 372-73, 379, 434, 437
Sleeve, 274n
SL-1, programming language, 433
Small Office Network Data System
(SaNDS), 625, 629, 630
Social policies, telephone rates as a
reflection of, 698
Software, 642-43
at Bell Laboratories, 21, 22
operations systems enhancement, 654
in capital-saving projects, 714
DIMENSION PBX, 500-501
in economic evaluation, 718
No. 1 PSS packet switch, 537
Solar cell, 21, 359, 543
Solid-state circuitry, 340, 352, 498
Sonar, submarine, 19-20
South Central Bell, 12, 36, 621
Southern Bell, 12,36
Southern New England Telephone, 5n, 12,29
Southwestern Bell, 12,36
Southwestern Bell Corporation, 36
Space diversity
satellite transmission, 210
terrestrial microwave radio transmission,
209
Space-division switching network(s), 244,
245-50
coordinate, 248-50
marker control, 260
progressively controlled, 246-48
topology, 245-46
TSPS,442
Space-division switching system(s), see
Electronic switching system(s);
Swit<:hing system(s)
Space program, Western Electric, major
contributor to, 19
Span, 330,376
Span line, 376
Spares, 608, 611, 612, 615
SPC No. 1A processor, TSPS, 442, 443
SPC No. 1B processor, TSPS, 443

873

Speaker
key telephone system, 490
small communications systems, 491
Speakerphone, 4A, 48, 470-71
picture, 48
Special final [trunk] group(s), 166n
Specialized common carrier, 306
Specialized Common Carrier Decision, 689,
693
Special-purpose terminals, 47-49
see also GEMINI 100 electronic blackboard
system; Repertory dialer;
Speakerphone; Transaction
telephone
Special service(s), 103n
definition of, 40
1 /lAESS switch, 420
SLC-96 carrier system, 373
see also Service(s), network
Special-services circuit(s), 103n, 199-200,
329,330,339,642
circuit order control system, 607
linear distortion objectives, 238
maintenance, 601
PBX, 87-88
private-line voice network, 89
provisioning, 606-7
Special-services signaling, 265, 273, 284-86,
287,290
Specific Development and Design (SO&O),
Bell Laboratories, 22, 23
Speech signal(s), 194,211
analog carrier system, 336, 345
average signal energy, graph, 195
crosstalk, 232-33
delay, 233
digital carrier system, 337
DMS-10 switching system, 431
echo canceler, 228
echo suppressor, 228
impairments and objectives, 223, 224
loops, 328
low-pass filter, 215
modulation of, 211
pulse-code modulation, 216-17
signal-to-noise ratio, 216
talker echo, 225-30
time-division switching network, 253
Speed Calling, 58, 501
Sports-phone, numbering, 122
Spurt signaling, 286
Staff organization (AT&T), corporate
functions, predivestiture, 8, 9
Standard Metropolitan Statistical Area
(SMSA),31
Standard Oil Trust, 688
Standard supply contract, 16, 19,30
State regulation(s), 695, 697
electronic tandem switching, 69
intrastate services, 699

874

Index

State regulation(s) (contd)
public telephones, 77
see also Public utilities commission
State regulatory commissions, 692
Stationary process, 151n
Station equipment, 84, 85, 103, 122, 465-89
carrier system, 272
see also Data set(s); Data terminal(s);
Graphics terminal(s); Telephone
set(s)
Station-loop signaling, see Signaling,
customer-line
Station (line) number, 115, 117
definition of, 116
see also Numbering plan; Telephone call
Station Signaling and Announcement
Subsystem, 443
Station-to-station call, custom.er-dialed, 439
Status display, in network management, 595
picture, 596
Step-by-step switch, 244
terminal bank, picture, 401
Step-by-step switching system(s), 94n, 398,
400-402,461,625,735
address format, 119
automatic call distributor, 504
automatic ticketing, 447
billing, 451, 452
billing data transmitter, 454
call data accumulator, 454
diagram, 458
COEES,639
coin telephones, 473
common control, 260
dial PBX, 493
diagram, 495
load balancing, 164
number of, in Bell System (1983), tables,
461,463
No. 355A SXS, 422
1000-line system, diagram, 402
progressive control, 246, 258, 260
tandem switching, 400
TNDS,630
toll switching, 399
TOUCH-TONE dialing, 736
Store and forward, see Billing system(s)
Stored-program control (SPC), 162-63, 241,
242,261-64,398,731
Automatic Intercept System, 444
centralized system, diagram, 262
chronology of switching systems, 458
table, 459
digital signaling, 292
growth, projected, 461-62
No. 4A Crossbar system, 408
processor arrangements
centralized processing, 263
functional multiprocessing, 264
independent multiprocessing, 263

signaling, 283
SPC No. lA processor, 442
SPC No. IB processor, 443
systems, engineering, 162-63
TSPS,439-43
see also Electronic switching system(s);
Stored-Program Control System
Stored-program control (SPC) network, 61,
507-13,655
Automated Calling Card Service,
diagram, 511
Expanded 800 Service, diagram, 512
generic architecture, diagram, 510
network schematic, 508
Stored-Program Control System (SPCS),
632,634
Switching Control Center System, 635-38
Stored-Program Control Systems, Central
Office Equipment Reports (SPCS
COER), 624, 625, 628, 629
Street Address Guide (SAG), 619
Stromberg-Carlson, 490n, 494
Strowger switching system, 398
Structure of the Bell System, see Bell
System, structure and activities
Submarine cable, see Undersea cable
Submarine carrier system, see Undersea
cable
Subnetwork planning, 142
Sub path, see Span
Subrate multiplexer, 531
capability, table, 533
Subscriber loop systems, see Loop
transmission systems
Subscriber main distributing frame (SMDF),
551,555
diagrams, 554, 557
Subscriber sender, 404, 405
Suburban area(s)
electronic switching system, 414
feeder route area, 122-24
wire center parameters, 124-25
table, 125
Suffix, address, 115, 117
definition of, 116
Suitability of service, 678
Summary bill format, 446-47
Sun transit outage, satellite communications
systems, 359
diagram, 360
Supergroup, frequency-division multiplex,
362, 363, 365, 366
Supervision, definition of, 85
see also Signaling
Support systems, 642-43
Surge protector, 469
Surveillance, in quality assurance
monitoring, 756-57
Surveys, in quality assurance monitoring,
757

Index

Switch
coordinate, diagrams, 249, 250, 251
packet, 190
rotary, diagrams, 247, 250
terminal bank, diagram, 401
time-division, 253
Switch access line port, engineering of, 188
Switchboard(s), toll, 111,438
picture, 439
Switched Access Remote Test System
(SARTS), 391n, 601
Switched Maintenance Access System
(SMAS),391
Switched message network
interconnection of terminal equipment,
412
interfaces for interconnection, 315
see also Network(s)
Switchhook, 85, 92
signaling, 267, 271, 468
Switching, 82, 241-64
congestion, 177, 178, 180
control mechanisms, 258-64
digital multiplexer, 392
4-wire, 244, 397-98,400,408,425,459,
462
functions, 241-43
incidental modulation, 240
local network, 104
in network planning, 137
networks, 243-58
special services, 40
2-wire, 397, 400, 442, 459
in 4-wire analog carrier interface, 395,
396
in local network, 244
see also Switching system(s)
Switching Control Center (SCC), 599, 632,
650
diagram, 634
picture, 651
Switching Control Center System (SCCS),
422, 436n, 632-38
enhancements, 636-37
evolution, 637-38
future plans, 638
operation, 634-36
Switching equipment
COEES, 639-41
CU/EQ,628
distributing frame, 549-50, 554, 557
line equipment, 164
manufacturing of, table, 15
provisioning, 588-89
service objectives, 182
switching networks, 1/ 1AESS switch,
416-18
telephone equipment building, 559-61
picture, 560
usage and peg count measurements, 183

875

Switching hierarchy, 184,241
distribution of offices, table, 109
toll, 107-9
2-level, 104, 106, 107
Switching network(s), 163, 243-58, 398, 622
control of, 242
definition of, 242
DMS-10 switch, 431, 433
electronic switching system, 413
No.5 Crossbar, 406, 407
l/lAESS switch, 416-18
TSPS,442
see also Circuit-switching network(s)
Switching office(s), 81
private-line voice network, 90
public switched telephone network, 8889
status evaluation, 594
Switching Control Center System,
connection to, 635
Switching primitives, 64, 510
Switching system(s), 6, 10,27,84,87-88,
103,122,397-464
administration and maintenance, 243
capacity expansion, 138-39
congestion, 151, 153, 177
control mechanisms, 242, 258-64
digital technology, impact of, 133
electromechanical, 398-409
evolution of, 413-14, 459, 734-35
local, 398-99
toll and tandem, 399-400
No.1 Crossbar, 402-6
No. 4A Crossbar, 408-9
No.5 Crossbar, 406-8
panel, 399
step-by-step, 400-402
electronic, 162-63, 409-38
concepts, 409-13
evolution of, 413-14, 459, 734-735
space-division, 292, 415-25
No. lOA Remote Switching System,
422-25
l/lAESS switch, 415-20
2/2BESS switch, 420-22
3ESS switch, 422
time-division, 425-38
DMS-10 switch, 430-34
4ESS switch, 425-30
5ESS switch, 434-38
engineering of, 160-65
functioning of, in typical telephone call,
92-95
load balancing, 164
maintenance, 599
number of, in Bell System, 458-63,622
performance model, 671
PICS,611
principal city concept, 120
private-line networks, 90, 91

876

Index

Switching system(s) (contd)
remote, 87
signaling, 242, 265
switching functions, distribution of,
diagram, 110
traffic engineering, 137
see also Billing system(s); Crossbar
switching system(s); Customer
switching services;
Electromechanical switching
system(s); Electronic switching
system(s); Interoffice facilities
network; Local facilities network;
Local switching system(s);
Operator systems; Packetswitching; Panel switching
system(s); Step-by-step switching
system(s); Stored-program control;
Supervision; Switching
Switching Systems area (Bell Laboratories),
24,25
Sykes Datatronics, Inc., 484
Synchronizing signal, 214, 221-22,240
Synchronous transmission, 477
data sets, 479, 480, 483
data terminals, 482
System code, 119
definition of, 116
Systems engineering, 23
Talker echo, 225-30
path, diagram, 227
versus grade of service, graph, 228
Talkoff, 289
Tandem Cross Section Program (TCSP), 144
Tandem office, 93, 94, 104, 128, 169,398
Tandem switching system(s), 286, 397, 399400
billing, 451, 454
definition of, 87
electronic, 425, 426
interoffice facilities network, 125, 127, 128
local network, 104
operator systems, 438, 439
Tandem Tie-Line Service, 67
Tandem trunk(s)
crosstalk objectives, table, 234
digital channel bank, 388
engineering, 137
local network, 104, 107
loss objectives, 225
table, 226
Tandem trunk group(s), local network, 104,
106
Tariff(s), 87, 667,694,695-97,708,734
basic services, 56, 702-3
developm~nt, 697
interconnection, 701

sample, 696
TOUCH-TONE service, 736
Tax depreciation, 720,725
Taxes,719,720,724-25,726
T-Carrier Administration System (TCAS),
600n, 638
T-Carrier Fault-Locating Applications
Program (TFLAP), 380
T-carrier transmission facilities, 423, 437,
611
interface, diagram, 302, 303
see also Tl digital carrier system
TD long-haul analog microwave radio
system(s), 352-56, 366, 533, 534
diagram, 354
table, 353
TD-2, 352, 353, 355
Technical advisories, 27-28
Technical descriptions, 28
Technical information, 27-28, 30
Technical references, 28
Technical standards, 28, 30
Technology, application of new technology,
666,729-33
Technology systems (central services
organization), 37
Telecommunications, and data-processing
services, 702, 703
Telecommunications Alarm Surveillance
and Control (TASC) System, 600n
Telecommunications network, see Facilities
network; Network(s)
Teleconferencing, 48,666
graphics terminal, 484-85
PICTUREPHONE meeting service, 527
video service, 75-76
Telegraph channel(s), 198
private-line data networks, 91
Telegraph signals, 193
Telemetry, 66,67,310,529
DA TAPHONE Select-a-station, 71
digital carrier transmission, 379
master control center, 633
satellite communications systems, 359
Telephone answering systems (bureaus), 56,
504-7
bridged connection, 504-5
diagram, 505
Call Forwarding, 506-7
diagrams, 506, 507
concentrator/identifier, 505-6
diagram, 505
direct inward dialing, 506-7
diagram, 507
Telephone call
electronic switching system processing
of,412-13
functional description, 92-99
signaling, 266-67

Index

Telephone Cases, 126 U.S., 700n
Telephone company(ies), 3, 4, 5, 6, 10, 16,
701
Bell Laboratories
field representatives, 27
projects, funding of, 23
cellular radio, 524
COEES, 641
Computer Subsystem, 634
concurrence in tariffs, 697
customer service centers, 43,79,582
customer support services, 78-80
distribution of offices in PSTN hierarchy,
table, 109
divestiture of, 29-32, 666, 703
economic evaluation, 714, 715, 717
matching new technology to market
needs, 730
mature product management, 708
network, 241, 245
administration, 592-96
maintenance, 597-601
provisioning, 585-92
number of calls per day, 445
operations, 571-602
customer-related, 573-85
functions, 571-73
network-related, 585-602
operations planning, 573, 643, 647, 648,
652,654,656-59,662
diagram, 657
impact of, 654-59
paging systems, 525
performance objectives, 671
PIes, 612
public telephone service
administration, 583
directory service, 584-85
maintenance, 583-84
provisioning, 582-83
quality assurance, 758
regulation, 691, 694
resources and volume of service,
predivestiture (1982), table, 29
service
administration, 577-79
evaluation, 665
evolution, 705, 706, 707, 708
maintenance, 579-82
measurement, 679, 680
provisioning, 573-77, 575
settlements, 699
standard supply contract, 19,30
structure and activities
postdivestiture, 34-37
predivestiture, 11-14
functional organizations, diagram, 13
territories (1982), map, 12
TNDS, 623, 631, 632

877

toll center code, 116
tracking studies for quality assurance, 756
traffic engineering methods, 148n
see also Operations system(s); Telephone
equipment building
Telephone equipment building
alarm systems, 567
building system standards, 564-65
cable distribution systems, 565, 567
cable entrance facility, 565, 567
electrical systems, 561
equipment area, 559-61
diagram, 559
mechancial systems, 561, 563
pictures, 560, 562, 566
special construction, 563-64
Telephone number, 87
international numbering, 120-21
numbering plan, 114-20
special numbering, 121-22
see also Address
Telephone Service Attitude Measurement
(TELSAM) Program, 678, 680-81
Telephone set(s), 41-43, 85, 465-76
electronic,469-70
500-type, 465-69
functioning of, in typical call, 92
installation, 576
maintenance, 582
modular, 43
number in service, 28,687
public, 471-75
sales, 43, 79
special features, 470-71
Transaction, 475-76
see also Handicapped, aids for the;
Repertory dialer; Speakerphone
Telephone voting service, 65
Teleprinter, 43 teleprinter, 483, 484
picture, 483
Telethon service, 65
Teletype Corporation, 5,20,483,610
data terminals, 46, 480
Teletypewriter (TTY), 66, 67, 480, 610
telegraph channels, 91
Television
channels, 198
early work in, 526
private-line services, 66, 67
program networks, 92
program Signals, 194
scanning process, 195
diagram, 196
video signals, 195-96
see also PICTUREPHONE visual telephone
service
Telstar 3 satellite, 357, 358, 360
Terminal, 369
carrier, 373, 383, 384, 389, 394, 395

878

Terminal (contd)
control, 519, 525
express, 384
intelligent, 46, 480, 483, 661
multiplex/demultiplex, 200, 239, 362-69
special-purpose, 47-49
synchronization loss, 375
Terminal box, 619
Terminal equipment, 41-51
American Bell Inc., 703
carrier systems, 336, 337
data communications, 319
definition of, 306
design considerations, 488
FCC registration, 311, 312
incidental modulation, 240
installation, 576
interconnection, 305-6, 307-10,311-13,
693-94,701
interfaces, 306-7, 310, 313, 315
network protection, 311-13
retail sales, 79
settlements, 699
signaling over digital facilities, 291
see also Station equipment
Terminal-to-terminal message service, 529
Termination layout mask, 557
Terrestrial microwave radio, see Microwave
radio
Test access, 391, 551
Test centers, DA TAPHONE digital service,
534
Test specifications, preparation of,
Bell Laboratories Specific Development
and Design, 23
Textbooks, 28
T4M digital repeatered line, 377-78, 534
Thermal noise, 21
definition of, 230
satellite communications systems, 360
Third-number billing, from public
telephones, 77
TH long-haul analog microwave radio
system(s), 352, 356, 366, 533
table, 355
TH-l,355
TH-3,355
3A central control (3ACC), 421-22
3A-Radio Digital System (3A-RDS), 382-83
3A-radio digital terminal (3A-RDT), 383
3B20 computer, 434, 513
3B20D computer, 536, 537
3ESS switch, see Electronic switching
system(s)
Three-Way Calling, 57
Throughput, 188, 189,237
Tie cable, 551, 554
Tie-line service, 67-68, 69
Tie trunk, see Trunk(s)

Index

Time Assignment Speech Interpolation
(TASI), 194n
Timed coin service, 473-74
Time-division multiplex (TOM), see
Multiplexing
Time-division switching network(s), 244,
245,250-58
architecture, 255-58
Automatic Intercept System, 443-44
operation, 253-55
Time-division switching system(s),
see Electronic switching system(s);
Switching system(s)
Time-multiplexed switch (TMS), 255-57,
428,429
analog eqUivalent, diagram, 256
configuration, diagram, 256
5ESS switching system, 434, 435, 436
TSI-TMS complex, diagram, 430
Timeouts, 178-79,681-82
Time sharing
COEES,639
data traffic, 185
digital multiplex equipment, 391
No. 4A Crossbar System, 410
signaling, 286
stored-program control switching
systems, 162, 262
Time-slot interchange (TSI), 253, 254, 255,
257-58,389,428,434,436,733
diagram, 254
Mass Announcement System, 513
picture, 734
TSI-TMS complex, 429-30
diagram, 430
Time-space-space-time (TSST) switching
network, 257
Time-space-time (TST) switching network, 257
diagram, 257
5ESS switch, 434
Time value of money, 720-23
Timing, in marketing, 711
Timing information, T1 carrier, 376
Tip and ring conductors, see Ring lead; Tip
lead
Tip (T) lead, 274, 275, 278, 284, 296n, 297,
307,315
see also Ring lead
TL-A2 short-haul analog microwave radio
system, 346-47
table, 347
TM-2/TM-2A short-haul analog microwave
radio system(s), 345-46
table, 346
TNDS Performance Measurement Plan
(TPMP),631
TN-l short-haul analog microwave radio
system, 346-47, 383
table, 347

Index

Toll-and-assistance operator(s), 62, 77, 111
Toll calls, 61, 62, 290
address format, 119
billing, 445, 446, 698
coin telephones, 77
routing, 109
small-business customers, 53
Toll center, 108, 109, 128
distributing frame system, 550
picture, 566
Toll center code, 119
definition of, 116
Toll charge(s), 63, 77, 578
Toll connecting trunk(s), 108, 388
crosstalk objectives, table, 234
loss objectives, 225
table, 226
talker echo objectives, 226, 227
diagram, 227
table, 229
Toll network, 103, 107-10,397, 671n
address format, 117n
IFRPS, 642
minimum high-usage trunk group size,
172
stored-program control network, 508,
509
switching hierarchy, diagram, 508
Toll office, 128, 180,226,227,293,397,408,
563
Toll operation, switching networks, 244
Toll service(s), 61
CCIS,293,294
No. 4A Crossbar System, 408
operator systems, 62, 414, 438-43
Toll switchboard, 111
Toll switching system(s), 87, 130, 398, 459,
627
census (1983), 462
table, 463
evolution, 399-400
4ESS switch, 133,425-26
see also Crossbar switching system(s);
Electronic switching system(s)
Toll traffic, 87
T1 digital carrier system, 132, 173, 330, 341,
373,374-77,379,380,437,534
Digital Access and Cross-Connect System,
391
efficiency, 337
as a high-frequency line, 201
see also T1 Outstate (T1/0S) digital
system
Tone-off,289
Tone-on, 289
T1 Outstate (T1/0S) digital system, 300,
379-81,534
T1 repeater, 372, 373
Tone ringer, 469

879

Total Network Data System (TNDS), 62232,643
component systems, table, 625
data flow, diagram, 624
evolution, 632
functions, 623, 627-32
central office reporting, 628-30
data acquiSition, 623, 627-28
system performance measurement,
631-32
trunks network reporting, 630-31
modules, 623
Total Network Operations Plan (TNOP),
632,647,648,651
TOUCH-A-MATIC S repertory dialer, 47-48,
470
picture, 47
TOUCH-TONE calling
dial PBX, 494
rates, 698
TOUCH-TONE dialing, 42, 285, 328,468,736
Call Forwarding, 121
Transaction telephone, 476
TOUCH-TONE service, 42, 58, 276, 286
circuits, 419
crossbar PBX, 496
interfacing with older equipment, 73637
paging systems, 525
TOUCH-TONE telephone, 41,150,469,481,
736,737
address input, 117
business and credit transactions, 49
dialing frequency groups, diagram, 276
dialing times, 161
high-speed train telephone service, 75
keypad,470
picture, 466
signaling, 285, 290
Townes, C. H., 22
Traffic,54,147-91
blocking criteria, 31
business customers, 54, 55
data networks, 184-91
definition of, 81
measurements, 180-83, 243, 589, 622,628,
631-32
in rural areas, 125, 128
service objectives, 183-84
theory, 148-59
TNDS, 622, 623, 625, 627, 628, 629, 630,
631,632,659
see also Network management; Switching
system(s), engineering of; Traffic
engineering; Traffic network(s);
Trunk group(s), engineering of
Traffic Data Administration System (TDAS),
624,627,628,629,630
table, 625

880

Traffic engineering, 104, 107,109, 136, 137
methods, definition of, 148
studies, 575
Traffic network(s), 81, 88-92, 103
design, 169-76
management, 176-80
structure, 103-14
see also Private switched network;
Program network; Public switched
telephone network
Traffic service position (TSP), 438
Traffic Service Position System (TSPS), 77,
111-12,283,414,438-43,590
Automated Coin Toll Service, 472
digital channel bank, 388
human factors in design of, 742
operator functions, table, 440
public telephone, trouble detection, 583
remote trunk arrangement, 443, 473
SCCS, 637
in stored-program control network, 508,
509, 511, 512
Traffic theory, 148-59
demand, measure of, 148-49
engineering periods, 151-55
Erlang's models, 156-59
grade of service, 149-51
techniques, 155-56
Traffic usage recorder, 182, 623
Train telephone service, 75
Transaction length, data traffic, 185, 186,
187,188
Transaction rate, data traffic, 185, 186, 187
Transaction telephone, 49,475-76
picture, 50, 475
printer, 476
Transatlantic coaxial cable system(s), 350-51
Transformer, digital channel banks, 387
Transistor, 21, 22
A-type channel bank(s), 365
Transition management, 595
see also Network(s), administration
operations
Translation(s), 94, 95, 98, 400, 408, 448
Translator, 259-60, 408
Transmission, 193-240
channels, 198-201
customer opinion model, 672
graphs, 673
4-wire, 274n, 289, 308, 315, 531
4-wire analog carrier interface, 295,
296,297,301
multiplexing, 387, 391, 397
No. 4A Crossbar System, 400, 408
talker echo, 226-27
voice-frequency, 199-201,225,335,340,
341,540
impairments, 223-40
land mobile telephone systems, 518-19
local network, 104

Index

media, 201-10
modulation, 210-18
multiplexing, 218-23
in network planning, 137
objectives, 223-40
performance characterization studies, 668
performance model, 670-71
performance objectives, 676
signal types, 193-98
special services, 40
telephone set, 465-68
2-wire, 86, 274n, 397,400,441
analog carrier, 345
digital carrier, 372
4-wire analog carrier interface, 295,
296,297
multiplexing, 362, 387, 391
talker echo, 226-27
VOice-frequency, 199-201,225,333,334,
339, 340, 341, 540
typical telephone call, 95, 96, 97
see also Loss
Transmission equipment
manufacturing of, table, 15
in telephone equipment building, 559
Transmission facilities, 10, 81, 84, 85-86,
103, 122, 241
capacity expansion, 138-39
circuit routing, 138
diagram, use of, 89
engineering data networks, 190
inventory and assignment records, 608
planning, 607, 609
provisioning, 588-89
PSTN and private networks, 91
signaling, 270, 280
see also Channel(s); Digital Data System;
Digital facilities network; Digital
transmission; Interoffice facilities
network; Transmission system(s)
Transmission interface, 389, 390
Transmission level point (TLP), 296-99, 607
diagram, 297
Transmission media, 84, 139,201-10
Transmission mode, waveguide, 204
Transmission performance objective, Digital
Data System, 237
Transmission system(s)
analog carrier, 336-37
maintenance, 600, 601
signaling, 289, 290
applications, 327-32
digital carrier systems, 128, 337
maintenance, 600, 601
MATFAP, 641
signaling, 269, 272
voice-frequency, 333-35
see also Analog carrier transmission;
Digital carrier transmission; Voicefrequency transmission

Index

Transmission System Optimum Relief Tool
(TSORT), 144
Transmission Systems area (Bell
Laboratories), 24, 25
Transmitter
FT3 lightwave system, 378
land-based radio systems, 519
long-haul microwave radio systems, 352,
352n, 353,354,356
paging systems, 525
speakerphone, 48,470-71
telephone set, 85, 97, 211, 465, 470
Transponder, satellite communications
systems, 358
Transverse electric (TE) mode, 204n
Transverse magnetic (TM) mode, 204n
Traveling wave tube (TWT), 21, 356
TRIMLINE telephone, 41, 488
picture, 466
Trouble
detection, 579, 594, 580
location, 580, 581
notification, 579,580
repair, 581
verification, 579, 580
Trouble report, 581-82, 598, 678
Trouble ticket, 637
Trunk(s), 87, 397
alternate routing, 170-72
automatic call distributor, 503
Automatic Intercept System, 443, 444
Basic Packet-Switching Service, 72
common-channel interoffice signaling,
509
congestion, 177, 178
cost, 172-73
crosstalk objectives, table, 234
customer switching system load, 579
definition of, 86
digital channel bank, 388
distributing frame, 549
DMS-I0 switching system, 431
echo, 226-27,229
engineering, 137, 175-76, 190
forecasting, 174-75, 588, 630-31
functioning of, in typical telephone call,
92-99
holding times, 166n
interoffice facilities network, 126, 128,
130
local facilities network, 124, 125
local network, 104
loss objective(s), 225, 226, 232
table, 229
maintenance, 282, 599-600
message, 607
metro area (exchange area), 374
metropolitan interoffice plant, 329
number of, in Bell System, 622
outstate interoffice plant, 330

881

PBX, 88
tie, 90
private switched networks, 90, 113
provisioning, 587-88
PSTN model, 671
remote office, introduction of, 130
servicing, 176, 588, 630-31
signaling
interoffice trunk, 265, 268, 273, 276-84
per-trunk signaling, 271, 277-78, 280
supervision, 419
stored-program control systems, table,
459
tie
dial-repeating, 284, 285
nondial, 284, 285
toll network, 108, 109
TSPS,439,441
2-wire transmission, 199
see also Busy season busy hour;
Centralized Automatic Reporting
on Trunks; Trunks Integrated
Records Keeping System
Trunk Addition Process, 648, 650
Trunk-based forecasting method, 588
Trunk circuit, 406, 407, 408, 431, 441, 651
D4 channel bank, functions, 302-5
1/IAESS switch, 416, 417
functions, 419
Trunk Forecasting System (TFS), 624, 625,
630-32
Trunk group(s), 94
alternate routing, 170-72
CCIS signaling, 282, 292
congestion, 179
efficiency, 106, 137
engineering of, 137, 165-69
interoffice facilities network, 126, 128, 130
local facilities network, 125
local network, 104
measurements, 153, 175, 182
network management control, 181
number of, in Bell System, 622
toll network, 109
utilization, 631
see also High-usage trunk group(s); Trunk
Servicing System
Trunk link, 407
Trunk link frame, 406
Trunk link network (TLN), 390, 417
Trunk main distributing frame (TMDF),
551,554,555
Trunk scanner, 160
Trunk Servicing System (TSS), 176, 624,
625,630-31
Trunks Integrated Records Keeping System
(TIRKS), 559, 605-11, 613, 643, 660
benefits to BOCs, 610-11
components, 607-9
diagram, 609

882

Index

Trunks Integrated Records Keeping System
(TIRKS) (contd)
human factors in design of, 740
implementation, 610
interfaces, 609-10, 615
role in circuit provisioning, 605-7
diagrams, 606,608
Trust, 688, 691
T2 digital carrier system, 381-82, 534
Tube, coaxial cable, 203, 204
Turbine, 544
2BESS switch, see Electronic switching
system(s)
2ESS switch, see Electronic switching
system(s)
Two-line telephone set, 43
Two-point networks, interstate private
teletypewriter, 91
Two-tandem-office local network, diagram, 107
Two-way radio, 516, 518
see also Mobile telephone service(s)
Two-way trunk, 94-95, 278, 282
Two-wire interface, 307, 312, 314, 315
Tymshare,604
Undersea cable, 28, 201
coaxial cable systems, 203, 350-51
succession of, graph, 351
delay, 234
Underwater surveillance projects, 19-20
Uniform Call Distribution, 501, 503
U nigauge design, VF loop transmission, 338
United Nations, 28
United States Independent Telephone
Association (USITA), 6, 27-28
UNIVAC computers, 604, 618, 621
Universal Cable Circuit Analysis Program,
380
Universal Sampling Plan (USP), 751
Universal Telephone Service, 3, 39, 692,
698,700
UNIX operating system, 668
Urban areas, 524, 525
Advanced Mobile Phone Service, 522, 523
busy season busy hour, 152
focused overload, 179
metro interoffice plant, 329-30, 339-40,
341,343,373-79
metropolitan interoffice facilities
networks, 126-28
diagram, 127
lESS switch, 413
wire center parameters, 124-25
table, 125
Usage-sensitive rates, 697, 698
User-premises communications system, 306,
308-10
U 600 channel, 366

U S West, Inc., 36
Utilization of equipment, 594, 610
Vacuum-tube technology, A-type channel
banks, 365
Vail, Theodore, 700
Value-of-service pricing, 697, 698
Variable Call Routing, 64
Variance, in common systems provisioning,
592
Varistor, 467
Velocity of propagation
carrier transmission, 335
reduction of, with loading, 334
Vertical integration, 3-5,695
Vertical service(s), see Service(s)
Very high frequency (VHF), maritime
[radiotelephone] service, 74-75
Very large-scale integrated chip, picture,
734
Vestigial sideband (VSB) modulation, see
Modulation
Via Net Loss (VNL) Plan, 227, 229, 340
loss component versus length, table, 229
Vidar,447
Video display terminal, see Cathode-ray
tube
Video program network, 92
Video signal, 195-96
long-haul microwave radio systems, 353
short-haul microwave radio systems, 345
Video teleconferencing, 527-28
Video teleconferencing service, 75-76
Video telephony, see Television
Video transmission
cable facility, 310
channel(s), 198
High-Speed Switched Digital Service, 73
private-line circuits, 66, 67
Violation-removal circuit, 382
Virtual call capability, 72
Virtual circuit, 185
Visual display, paging receiver with, 74,
525
Visual systems, 526-29
PICTUREPHONE visual telephone
service, 256-57
research, 528-29
television, early, 526
video teleconferencing, 527-28
see also High-Capacity Terrestrial Digital
Service; High-Speed Switched
Digital Service; PICTUREPHONE
meeting service
Vitel,447
Voiceband (voice-frequency [VF]) channel
(circuit), 194, 198, 199-201,230,

295,305n,431

Index

amplitude / frequency distortion, 234-35
data communications, 73,132,197
interoffice, 394
lightwave systems, 207, 383
long-haul microwave radio systems, 352,
353,355,356
loops, 393
metropolitan interoffice plant, 128, 330
modulation, 211, 217-18
DSBAM transmission, 213
pulse-code, 217
SSBAM transmission, 213
stacking, diagram, 214
multiplexing, 219-221
frequency-division, 362-69
outstate interoffice plant, 330
private-line services, 67, 71, 113, 538, 540
radio digital systems, 382
satellite communications systems, 357
short-haul microwave radio systems, 34547
signaling, 266, 276, 289
2-wire /4-wire, 199-200
diagram, 199
undersea coaxial cable systems, 350
Voiceband data sets, 44, 477, 478, 479
tables, 321, 481
Voiceband interface unit (VIU), 428, 429,
430
Voiceband private lines, 44, 45, 479
Voiceband signal, 251, 299
Voice bandwidth, telegraph channels, 91
Voice camera switching, PICTUREPHONE
meeting service, 528
Voice channel, see Voiceband channel
Voice-frequency (VF), loss objectives, 225
VOice-frequency repeater (amplifier), 239,
330,334
Voice-frequency (VF) transmission, 290,
337-40,601
loop, 337 -39
design methods, 338-39
metropolitan interoffice, 330, 339-40
outstate interoffice, 340
Voice paging, 525
Voice path test, 293
Voice switching, speakerphone, 471
Volatile memory, 542
Voltage drop, 548
V.35 interface, 321
Waveguide(s), 201, 204-5
long-haul microwave radio system, 352
picture, 208
WE 8000 microprocessor, 305, 423, 479
WE 4000 microcomputer, 472
WESCOM, Inc., 449, 504
WESTAR domestic satellites, 357

883

Western Electric Company, Incorporated, 4,
5,30, 33, 37, 689,700
antitrust (1949), 694
antitrust (1974), 703
AUTOPLEX cellular radio system, 522,
523
Bell Laboratories
branch laboratories, 25, 27
Specific Development and Design,
funding for, 23
Business Information Systems products,
642-43
computers and processors, 305, 423, 434,
455,472,513,536,537
economic evaluation of projects, 715, 717,
727,728,729
EPLANS Computer Program Service, 642
government systems, 19
litigation, 694, 703
locations and regions (1982), map, 17-18
manufacturing, 14, 15, 16, 17, 241, 379,
419,454,496,523,641,733,745
military systems, 19-20
postdivestiture, 33
product evolution, 708
quality assurance, 742, 743, 745, 757, 758
Remote Data Entry System, 642
service evolution, 707
step-by-step switching system, 398, 735
structure and activities, predivestiture,
14-20
subsidiary companies, pre divestiture, 20
Western Union Telegraph Company, 40,
689,691,700
White pages, 9, 39, 77, 80, 577, 584
Wide Area Telecommunications Services
(WATS), 40,61,63-64
billing, 577
for business customers, 54-55
data communications, 44
INWATS, 63, 426, 512
AMA,455
marketing, 710
OUTWATS, (800 Service), 63
signaling, 284
Wide band data set, 321
Wilson, Robert, 22
Wilson, Woodrow, 691
Wind-driven generator, 543
Wink, 268
Wire, 27
open-wire lines, 201
paired cable, 201-3, 333, 347
Wire center, 104, 105
distributing frame systems, 550, 551
interoffice facilities network, 126, 127,128
local facilities network, 122, 123, 124, 125
average parameters for PSTN, table,
125

884

Wire center (contd)
picture, 566
planning, 589
Wire center area, 104, 105, 122
Wired-logic electronic private branch
exchange,498-99
Wired music system program signals, 194
Wire-line carrier, 517, 518
Wire pair(s), 373, 393
Wire products, manufacturing of, table, 15
Wiring plan, 490
Wisconsin Telephone, 12

Index

Work package, 606, 607
World zone, 120-21
table, 121
X.25 interface, 320, 322, 323, 536, 537, 538
X.75 interface, 320, 322
Yellow Pages, 9, 77, 80, 577, 585
Zone registration, 447

ENGINEERING AND OPERATIONS IN THE BELL SYSTEM

Text set at AT&T Bell Laboratories in Palatino Roman
on an AUTOLOGIC, Inc. APS-5 phototypesetter
using the TROFF formatting and phototypesetting program
under the UNIX operating system.
Display lines set on a Varityper Comp/Edit 5810 in Korinna Bold
Index set by University Graphics, Inc.
Printed by the Maple-Vail Book Manufacturing Group
on Perkins & Squier Offset Smooth text stock
Bound by the Maple-Vail Book Manufacturing Group
in James River's Kivar 9 Mulberry cover
with Lindenmeyr's Multicolor Antique Muscatel endsheets



Source Exif Data:
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37
Producer                        : Adobe Acrobat 9.12 Paper Capture Plug-in
Modify Date                     : 2009:07:22 22:31:03-07:00
Create Date                     : 2009:07:22 22:31:03-07:00
Metadata Date                   : 2009:07:22 22:31:03-07:00
Format                          : application/pdf
Document ID                     : uuid:79fddf39-a221-46f6-b1dc-087f3efbf6d6
Instance ID                     : uuid:b420cbb1-da1e-495d-837c-dc0886a4e9f1
Page Layout                     : SinglePage
Page Mode                       : UseOutlines
Page Count                      : 902
EXIF Metadata provided by EXIF.tools

Navigation menu