gsm, gprs, and edge performance evolution towards 3gumts

Evolution Towards 3G/UMTS
Second Edition
GSM, GPRS
Performance
EDGE
AND
Edited by
Timo Halonen
Nokia
Javier Romero and Juan Melero
TarTec

GSM, GPRS
Performance
EDGE
AND

Evolution Towards 3G/UMTS
Second Edition
GSM, GPRS
Performance
EDGE
AND
Edited by
Timo Halonen
Nokia
Javier Romero and Juan Melero
TarTec
Copyright  2003 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,
West Sussex PO19 8SQ, England

Telephone (
+44) 1243 779777
Email (for orders and customer service enquiries):
Visit our Home Page on www.wileyeurope.com or www.wiley.com
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or
otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of
a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP,
UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed
to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West
Sussex PO19 8SQ, England, or emailed to , or faxed to (
+44) 1243 770620.
This publication is designed to provide accurate and authoritative information in regard to the subject
matter covered. It is sold on the understanding that the Publisher is not engaged in rendering
professional services. If professional advice or other expert assistance is required, the services of a
competent professional should be sought.
Other Wiley Editorial Offices
John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA
Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA
Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany
John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia
John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809
John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1
Wiley also publishes its books in a variety of electronic formats. Some content that appears
in print may not be available in electronic books.
Library of Congress Cataloging-in-Publication Data
GSM, GPRS, and edge performance : evolution towards 3G/UMTS / edited by Timo
Halonen, Javier Romero, Juan Melero.—2nd ed.
p. cm.
Includes bibliographical references and index.

ISBN 0-470-86694-2
1. Global system for mobile communications. I. Halonen, Timo, II. Romero, Javier
(Romero Garc
´
ıa) III. Melero, Juan.
TK5103.483.G753 2003
621.3845

6—dc22
2003057593
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN 0-470-86694-2
Typeset in 10/12pt Times by Laserwords Private Limited, Chennai, India
Printed and bound in Great Britain by TJ International, Padstow, Cornwall
This book is printed on acid-free paper responsibly manufactured from sustainable forestry
in which at least two trees are planted for each one used for paper production.
Contents
Acknowledgements xvii
Foreword xix
Introduction xxv
Abbreviations xxix
Part 1 GERAN Evolution 1
1 GSM/EDGE Standards Evolution (up to Rel’4) 3
Markus Hakaste, Eero Nikula and Shkumbin Hamiti
1.1 Standardisation of GSM—Phased Approach 4
1.1.1 GSM/TDMA Convergence through EDGE 5
1.1.2 GERAN Standardisation in 3GPP 6
1.2 Circuit-switched Services in GSM 8
1.2.1 Adaptive Multi-rate Codec (AMR) 9

1.2.2 High Speech Circuit-switched Data (HSCSD) 11
1.3 Location Services 13
1.3.1 LCS Standardisation Process 13
1.4 General Packet Radio System (GPRS) 14
1.4.1 Introduction of GPRS (Rel’97) 14
1.4.2 GPRS Network Architecture 15
1.4.3 GPRS Interfaces and Reference Points 20
1.4.4 GPRS Protocol Architecture 22
1.4.5 Mobility Management 25
1.4.6 PDP Context Functions and Addresses 27
1.4.7 Security 28
1.4.8 Location Management 28
1.4.9 GPRS Radio Interface 28
1.5 EDGE Rel’99 47
1.5.1 8-PSK Modulation in GSM/EDGE Standard 47
1.5.2 Enhanced General Packet Radio Service (EGPRS) 49
1.5.3 Enhanced Circuit-switched Data (ECSD) 51
vi Contents
1.5.4 Class A Dual Transfer Mode (DTM) 53
1.5.5 EDGE Compact 54
1.5.6 GPRS and EGPRS Enhancements in Rel’4 54
References 55
2 Evolution of GERAN Standardisation (Rel’5, Rel’6 and beyond) 57
Eero Nikula, Shkumbin Hamiti, Markus Hakaste and Benoist Sebire
2.1 GERAN Rel’5 Features 58
2.1.1 Iu Interface for GERAN and the New Functional Split 59
2.1.2 Header Adaptation of the IP Data Streams 61
2.1.3 Speech Capacity and Quality Enhancements 61
2.1.4 Location Service Enhancements for Gb and Iu Interfaces 62
2.1.5 Inter-BSC and BSC/RNC NACC (Network-assisted Cell Change) 63

2.1.6 High Multi-slot Classes for Type 1 Mobiles 64
2.2 GERAN Architecture 64
2.2.1 General 64
2.2.2 Architecture and Interfaces 65
2.2.3 Radio Access Network Interfaces 69
2.3 GERAN Rel’6 Features 81
2.3.1 Flexible Layer One 82
2.3.2 Single Antenna Interference Cancellation (SAIC) 86
2.3.3 Multimedia Broadcast Multicast Service (MBMS) in GERAN 87
2.3.4 Enhancement of Streaming QoS Class Services in GERAN A/Gb
Mode 88
References 88
3 GERAN QoS Evolution Towards UMTS 91
Erkka Ala-Tauriala, Renaud Cuny, Gerardo G´omez and H´ector Montes
3.1 Mobile Network as a Data Transport Media for IP-based Services 92
3.2 Example of IP-based Applications Using Mobile Network as Data Bearer 95
3.2.1 WAP Browsing 96
3.2.2 Multimedia Messaging Service (MMS) 96
3.2.3 Audio/Video Streaming 96
3.2.4 IMS Services 98
3.3 End-to-end QoS in the 3GPP QoS Architecture 99
3.4 PDP-context QoS Parameter Negotiation 101
3.4.1 QoS Authorisation for IMS and Non-IMS Services with PDF 105
3.5 Negotiated PDP-context QoS Enforcement in GERAN (and UTRAN) 106
3.5.1 Control Plane QoS Mechanisms 107
3.5.2 User Plane QoS Mechanisms 112
3.6 End-to-end QoS Management 115
3.6.1 Example of Service Activation Procedure 116
References 117
Contents vii

4 Mobile Station Location 119
Mikko Weckstr¨om, Maurizio Spirito and Ville Ruutu
4.1 Applications 120
4.2 Location Architectures 121
4.3 Location Methods 123
4.3.1 Basic Service Level 123
4.3.2 Enhanced Service Level 126
4.3.3 Extended Service Level 131
4.4 LCS Performance 133
4.4.1 Basic Service Level Performance 133
4.4.2 Enhanced Service Level Performance 137
References 139
Part 2 GSM, GPRS and EDGE Performance 141
5 Basics of GSM Radio Communication and Spectral Efficiency 143
Juan Melero, Jeroen Wigard, Timo Halonen and Javier Romero
5.1 GSM Radio System Description 143
5.1.1 Basic Channel Structure 144
5.1.2 Transmitting and Receiving Chain 146
5.1.3 Propagation Effects 149
5.1.4 Basic TCH Link Performance with Frequency
Hopping 151
5.1.5 Discontinuous Transmission (DTX) 155
5.1.6 Power Control 158
5.2 Cellular Network Key Performance Indicators (KPIs) 159
5.2.1 Speech KPIs 159
5.2.2 Data KPIs 167
5.3 Spectral Efficiency 173
5.3.1 Effective Reuse 174
5.3.2 Fractional Load 175
5.3.3 Frequency Allocation Reuse 175

5.3.4 Frequency Load 176
5.3.5 Effective Frequency Load 176
5.3.6 EFL for Mixed Voice and Data Services 178
5.4 EFL Trial Methodology 178
5.4.1 Network Performance Characterisation 178
5.4.2 Trial Area Definition 179
5.4.3 Methodology Validation 181
5.5 Baseline Network Performance 182
References 184
viii Contents
6 GSM/AMR and SAIC Voice Performance 187
Juan Melero, Ruben Cruz, Timo Halonen, Jari Hulkkonen, Jeroen Wigard,
Angel-Luis Rivada, Martti Moisio, Tommy Bysted, Mark Austin, Laurie Bigler,
Ayman Mostafa Rich Kobylinski and Benoist Sebire
6.1 Basic GSM Performance 187
6.1.1 Frequency Hopping 188
6.1.2 Power Control 193
6.1.3 Discontinuous Transmission 194
6.2 Reuse Partitioning 196
6.2.1 Basic Operation 197
6.2.2 Reuse Partitioning and Frequency Hopping 198
6.3 Trunking Gain Functionality 200
6.3.1 Directed Retry (DR) 200
6.3.2 Traffic Reason Handover (TRHO) 200
6.4 Performance of GSM HR Speech Channels 201
6.5 Adaptive Multi-rate (AMR) 203
6.5.1 Introduction 203
6.5.2 GSM AMR Link Level Performance 204
6.5.3 GSM AMR System Level Performance 207
6.6 Source Adaptation 216

6.6.1 Introduction 216
6.6.2 System Level Performance 217
6.7 Rel’5 EDGE AMR Enhancements 219
6.7.1 Introduction 219
6.7.2 EDGE NB-AMR Performance 219
6.7.3 EPC Network Performance 222
6.7.4 EDGE Wideband AMR Codecs 222
6.8 Single Antenna Interference Cancellation (SAIC) 224
6.8.1 SAIC Techniques Overview 224
6.8.2 SAIC Link Performance and Conditioning Factors 225
6.8.3 SAIC Network Performance 227
6.9 Flexible Layer One 229
6.9.1 FLO for Circuit-switched Voice 229
6.9.2 FLO for VoIP 230
References 232
7 GPRS and EGPRS Performance 235
Javier Romero, Julia Martinez, Sami Nikkarinen and Martti Moisio
7.1 (E)GPRS Link Performance 236
7.1.1 Introduction 236
7.1.2 (E)GPRS Peak Throughputs 237
7.1.3 RF Impairments 237
7.1.4 Interference-limited Performance 238
7.2 (E)GPRS Radio Resource Management 245
7.2.1 Polling and Acknowledgement Strategy 245
Contents ix
7.2.2 Link Adaptation Algorithms for (E)GPRS 247
7.2.3 (E)GPRS Channel Allocation 252
7.2.4 (E)GPRS Scheduler 254
7.2.5 GPRS and EGPRS Multiplexing 255
7.2.6 Power Control 256

7.3 GPRS System Capacity 258
7.3.1 Introduction 258
7.3.2 Modeling Issues and Performance Measures 258
7.3.3 GPRS Performance in a Separate Non-hopping Band 261
7.3.4 GPRS Performance in a Separate Band with RF Hopping 268
7.3.5 GPRS Spectrum Efficiency with QoS Criterion 269
7.3.6 Reuse Partitioning Principle to Increase Spectral Efficiency
and QoS Provisioning 272
7.4 EGPRS System Capacity 272
7.4.1 Introduction 272
7.4.2 Modeling Issues and Performance Measures 272
7.4.3 EGPRS Performance with Link Adaptation in a Separate
Non-hopping Band 272
7.4.4 EGPRS Performance in a Separate Band with RF Hopping 276
7.4.5 Spectrum Efficiency with QoS Criterion 278
7.4.6 Throughput Distribution Analysis 281
7.4.7 Effect of Traffic Burstiness 281
7.4.8 (E)GPRS Deployment 284
7.4.9 Gradual EDGE Introduction 285
7.5 Mixed Voice and Data Traffic Capacity 287
7.5.1 Best-effort Data Traffic 288
7.5.2 Relative Priorities 289
7.5.3 Guaranteed Data Traffic 289
7.5.4 Erlang Translation Factors 289
7.6 (E)GPRS Performance Estimation Based on Real Network Measurements 292
7.7 Application Performance Over (E)GPRS 295
7.8 (E)GPRS Performance Measurements 297
7.8.1 TSL Capacity Measurements 297
7.8.2 EGPRS Performance Measurements 302
References 305

8 Packet Data Services and End-user Performance 307
Gerardo Gomez, Rafael Sanchez, Renaud Cuny, Pekka Kuure and Tapio Paavonen
8.1 Characterization of End-user Performance 307
8.1.1 Data Link Effects 308
8.1.2 Upper Layer Effects 309
8.1.3 Performance Characterization Example. HTTP Performance
in GPRS 319
8.2 Packet Data Services 319
8.2.1 Web Browsing 320
x Contents
8.2.2 WAP Browsing 322
8.2.3 Multimedia Messaging Service 323
8.2.4 Streaming 325
8.2.5 Gaming 327
8.2.6 Push to Talk over Cellular (PoC) 327
8.3 End-user Performance Analysis 333
8.3.1 Web Browsing Performance 334
8.3.2 WAP Browsing Performance 336
8.3.3 Multimedia Messaging Service Performance 338
8.3.4 Streaming Performance 339
8.3.5 On-line Gaming Performance 341
8.3.6 Push to Talk over Cellular Performance 342
8.4 Methods to Optimize End-user Performance 342
References 348
9 Dynamic Frequency and Channel Allocation 351
Matti Salmenkaita
9.1 Air Interface Synchronisation 352
9.1.1 GSM Synchronisation Basics 352
9.1.2 Implementation of Synchronisation 352
9.1.3 TDMA Frame Number Considerations 353

9.1.4 Synchronisation Accuracy 353
9.2 DFCA Concept 358
9.2.1 CIR Estimation 358
9.2.2 Combination with Frequency Hopping 359
9.2.3 Radio Channel Selection 359
9.2.4 Information Exchange 361
9.2.5 DFCA Frequency Hopping Modes 362
9.3 Application of DFCA for Circuit-switched (CS) Services 363
9.3.1 Multitude of CS Services 363
9.3.2 The DFCA Way 364
9.4 DFCA Simulations with CS Services 364
9.4.1 Performance in Ideal Network Layout 364
9.4.2 Performance in Typical Network Layout 371
9.4.3 Summary of the Simulation Results 375
9.5 DFCA for Packet-switched (PS) Services 375
9.6 Simulations of DFCA in Mixed CS and PS Services
Environment 377
9.7 Summary 378
References 379
10 Narrowband Deployment 381
Angel-Luis Rivada, Timo Halonen, Jari Hulkkonen and Juan Melero
10.1 What is a Narrowband Network? 381
10.1.1 Frequency Spectrum Re-farming. Technology Migration 382
Contents xi
10.1.2 Narrow Licensed Frequency Spectrum 382
10.1.3 Microcell Deployment 383
10.2 Performance of Narrowband Networks 384
10.3 Allocation of BCCH and Hopping Bands 385
10.3.1 BCCH Reuse for Narrowband Scenarios 385
10.3.2 Narrowband BCCH and Hopping Deployment Strategies 386

10.3.3 Need of Guardband 387
10.4 BCCH Underlay 389
10.4.1 Description of BCCH Underlay Concept 390
10.4.2 BCCH Underlay Simulation and Trial Results 390
10.5 Transmit Diversity Gains 394
10.6 Common BCCH 396
10.7 Other Strategies to Tighten BCCH Reuse 396
References 396
11 Link Performance Enhancements 397
Riku Pirhonen, Matti Salmenkaita, Timo K¨ahk¨onen and Mikko S¨aily
11.1 Basics of Radio Link Performance 397
11.1.1 Minimum Performance Requirements 398
11.1.2 Radio Link Power Budget 400
11.2 Overview of Radio Link Enhancements 403
11.2.1 Uplink Diversity Reception 403
11.2.2 Uplink Interference Rejection 404
11.2.3 Mast Head Amplifier 405
11.2.4 Downlink Transmit Diversity 406
11.2.5 Macrodiversity 410
11.3 Coverage Improvements 413
11.3.1 Coverage Efficiency 413
11.3.2 Field Measurements 413
11.4 Capacity Improvements 416
11.4.1 Uplink Diversity Reception 416
11.4.2 Downlink Transmit Diversity 417
11.4.3 Macrodiversity 420
References 423
12 Control Channels Performance and Dimensioning 425
Timo K ¨ahk¨onen and Jorge Navarro
12.1 Introduction to Control Channels 425

12.1.1 Physical and Logical Channels 425
12.1.2 Control Channel Configurations 426
12.1.3 Usage of Control Channels 428
12.1.4 Channel Coding and Interleaving 430
12.2 Physical Layer Reliability 431
12.2.1 Simulation Model 431
12.2.2 Comparison of Channels 432
xii Contents
12.3 Signalling Reliability and Delays 434
12.3.1 Probabilistic Models 435
12.3.2 SCH Information Broadcast 436
12.3.3 System Information Broadcast 437
12.3.4 RR Connection Establishment 437
12.3.5 L2 Link Establishment 439
12.3.6 L2 Link Failure 440
12.3.7 Call Establishment and Location Update 440
12.3.8 Handover and Channel Transfer 441
12.3.9 Measurements and Power Control 442
12.3.10 Radio Link Failure 444
12.3.11 Conclusions 444
12.4 Control Channels versus AMR TCH 445
12.4.1 Physical Layer Comparison 445
12.4.2 System Level Comparison 446
12.4.3 Conclusions 449
12.5 Signalling Capacity 450
12.5.1 Signalling Capacity Criterion 450
12.5.2 Signalling Capacity for GSM Voice 450
12.5.3 Signalling Capacity for (E)GPRS 458
12.5.4 (E)GPRS Traffic Assumptions 458
12.5.5 Conclusions 465

References 465
13 Automation and Optimisation 467
Volker Wille, Sagar Patel, Raquel Barco, Antti Kuurne, Salvador Pedraza, Matias
Toril and Martti Partanen
13.1 Introduction to Radio Network Optimisation 467
13.1.1 Operational Efficiency 468
13.1.2 Characteristics of Automation 469
13.1.3 Areas of Automation 470
13.2 Mobile Measurement-based Frequency Planning 472
13.2.1 Outline of the Problem 472
13.2.2 Traditional Frequency Planning 474
13.2.3 The New Frequency Planning Concept 477
13.2.4 Signal-level Reporting in GERAN 478
13.2.5 Review of Interference Matrix Types 482
13.2.6 MMFP Trial Results 483
13.3 Automated Measurement-based Adjacency Planning 484
13.3.1 Maintaining Adjacencies 485
13.3.2 Solution Description 486
13.3.3 A Description of the Adjacency Management Process 486
13.3.4 Network Test Results 488
13.4 Automated Parameter Optimisation 491
13.4.1 Outline of Problem 491
Contents xiii
13.4.2 Control Engineering for Automatic Parameter Optimisation in
Mobile Networks 492
13.4.3 Applications of Radio Network Parameter Optimisation 495
13.5 Automated Troubleshooting of Cellular Network Based on Bayesian
Networks 501
13.5.1 Introduction 501
13.5.2 Troubleshooting Process 502

13.5.3 Decision Support Systems 503
13.5.4 Bayesian Network Models 505
13.5.5 Knowledge Acquisition 507
13.5.6 Troubleshooting Sequence 508
13.5.7 Advanced Features 509
13.5.8 Interaction with the Network Management System 509
References 510
Part 3 3G Evolution Paths 513
14 IMT-2000 3G Radio Access Technologies 515
Juan Melero, Antti Toskala, Petteri Hakalin and Antti Tolli
14.1 IMT-2000 3G Technologies and Evolution Paths 515
14.2 3G Technology Support of Licensed Frequency Bands 517
14.3 3G Radio Access Technologies—Introduction 518
14.3.1 WCDMA Basics 518
14.3.2 Multi-carrier CDMA (cdma2000) Fundamentals 525
14.4 3G Radio Access Technology (RAT) Performance Benchmark 528
14.4.1 Voice Performance 528
14.4.2 Data Performance 531
14.4.3 Conclusions 535
14.5 UMTS Multi-radio Integration 536
14.5.1 Introduction 536
14.5.2 UMTS Multi-radio Evolution 536
14.5.3 Mechanisms for UMTS Multi-radio Integration 538
14.5.4 Trunking Efficiency Benefits from Multi-radio Integration 538
14.5.5 QoS-based Multi-radio Integration 539
14.5.6 Architecture Integration 541
References 541
15 3G Technology Strategy and Evolution Paths 543
Juan Melero
15.1 3G Multimedia Services 543

15.1.1 Operators’ Business Impact 543
15.1.2 3G Technologies—Requirements 545
15.2 Globalisation 546
15.2.1 Technology Globalisation 546
15.2.2 Economies of Scale 547
xiv Contents
15.3 3G Technology Evolution Paths. UMTS Multi-radio and cdma2000 549
15.3.1 From 2G to 3G 550
References 553
Appendixes 555
Appendix A MAIO Management Limitations 555
A.1 MAIO Management Limitations and Planning 555
A.2 MAIO Management Limitations for Different Effective Reuses and
Antenna Beamwidth 557
Appendix B Hardware Dimensioning Studies 559
B.1 Blocking Probability for Half- and Full-rate Speech Modes 559
B.1.1 The Erlang-B Formula 559
B.1.2 Blocking Probability for HR/FR Scenario 560
B.1.3 Effective Factor 564
B.2 (E)GPRS HW Dimensioning Analysis 568
B.2.1 Dedicated PS Resources 569
B.2.2 Shared PS and CS Resources 571
References 578
Appendix C Mapping Link Gain to Network Capacity Gain 579
C.1 Introduction 579
C.2 Theoretical Analysis 579
C.3 Simulations 581
C.3.1 BCCH Layer Performance 582
C.3.2 Hopping Layer 583
C.3.3 Effect of Power Control 583

C.4 Final Results and Conclusions 585
References 586
Appendix D Interference between GSM/EDGE and Other Cellular Radio
Technologies 587
D.1 Introduction 587
D.2 Interference Mechanisms 589
D.2.1 Adjacent Channel Power 589
D.2.2 Intermodulation Distortion (IMD) 590
D.3 Coverage Effects 590
D.3.1 Downlink 591
D.3.2 Uplink 591
D.4 The Interference from WCDMA to GSM 593
D.5 Monte-Carlo Simulation Study (GSM/EDGE and IS-95) 594
D.6 Summary 596
References 597
Contents xv
Appendix E Simulation Tools 599
E.1 Introduction 599
E.2 Static Simulations 599
E.3 Basic Principles of Dynamic Simulation 600
E.4 Description of the Simulator and Basic Simulation Models Used in this
Book 602
E.4.1 Software and Programming Issues 602
E.4.2 Basic Functionality of the Simulator 602
E.4.3 Link-Level Interface 604
E.5 Description of the Basic Cellular Models 608
References 608
Appendix F Trial Partners 609
Index 611

Acknowledgements
We would like to thank all contributors for the “world-class” material produced to support
this book. This book has been the result of a collaborative effort of a truly motivated and
committed team.
We would also like to thank our listed colleagues for their valuable input and com-
ments: Heikki Annala, Harri Jokinen, Mattias Wahlqvist, Timo Rantalainen, Juha Kasinen,
Jussi Reunanen, Jari Ryynanen, Janne Vainio, Kari Niemel
¨
a, Simon Browne, Heikki
Heliste, Petri Gromberg, Oscar Salonaho, Harri Holma, Fabricio Velez, Jyrki Mattila,
Mikko S
¨
aily, Gilles Charbit, Jukka Kivij
¨
arvi, Martti Tuulos, Lauri Oksanen, Ashley Col-
ley, Mika K
¨
ahk
¨
ol
¨
a, Per Henrik Michaelson, Rauli Jarvela, Rauli Parkkali, Kiran Kuchi,
Pekka Ranta, Brian Roan, James Harper, Raul Carral, Joe Barret, Timothy Paul, Javier
Munoz, Poul Larsen, Jacobo Gallango, Regina Rodr
´
ıguez, Manuel Mart
´
ınez, Sheldon
Yau, and Art Brisebois. Special thanks to Jussi Sipola for the link performance material
provided in Chapter 7.

Thanks to Mark Keenan for the support, guidance and encouragement provided during
the last years. We hope he feels as proud as we do about the result of this book.
Thanks to the authors of the forewords (Mike Bamburak, Mark Austin and Chris Pear-
son) and the companies/institutions they represent (AT&T, Cingular and 3G Americas)
for the support provided.
Many thanks to the operators (CSL, Radiolinja, Sonofon, AT&T, Cingular, Telefonica
and many others) we have been closely working with since only through this tight collab-
oration the understanding of the technology capabilities contained in this book has been
possible. Special thanks to Optus, for their collaboration, support and excellent technical
team (Andrew Smith, Carolyn Coronno, Bradley Smith and the rest of the team).
Part of the studies presented in this book have been performed as the result of the
cooperation agreement between Nokia and the University of Malaga. This agreement is
partially supported by the “Programa de Fomento de la Investigaci
´
on T
´
ecnica” (PROFIT)
of the Spanish Ministry of Science and Technology. We want to thank Malaga University
and the Ministry of Science and Technology for the provided support.
Thanks to the publishing team from Wiley (Mark Hammond, Sarah Hinton and Geoff
Farrell), which has given an excellent support throughout this project, which has been
very important to accomplish the demanding time schedule.
Finally we would like to express our loving thanks to our wives Cristina, Laila and
Susanna for the patience and understanding during the holidays, weekends and late nights
devoted to the writing of this book.
xviii Acknowledgements
The authors welcome any comments and suggestions for improvements or changes
that could be implemented in possible new editions of this book. The e-mail address for
gathering such input is geran

Malaga, Spain
The editors of the “GSM, GPRS & EDGE Performance” book
Forewords
(Taken from GSM, GPRS and EDGE Performance, 1st Edition).
I have worked in the mobile communications industry longer than I would like to admit.
In the early 1970s, I started my career as a radio engineer for Motorola. At that time,
Motorola designed and manufactured low-, mid- and high-tier private land mobile radios.
Motorola had few competitors for the mid- and high-tier product lines (50- to 100-W
radios). However, in the low tier, less than 25-W radio category, there were numerous con-
tenders, mostly from European manufacturers with a ‘Nordic Mobile Telephone’ heritage.
But times were changing. In the late 1970s, the American public got their first taste
of mobile communications when Citizen Band (CB) radio became popular (‘10–4, good
buddy’). It was an unlicensed, short-range, ‘party-line’ experience. Those skilled in the
art knew that something better was needed. And the American communications industry
responded. The Federal Communications Commission and major industry players, like
AT&T and Motorola, specified America’s first public mobile radio telephone system,
AMPS (Advanced Mobile Telephone System). By the mid-1980s, AMPS was a proven
technology and cellular subscriber growth was constantly exceeding forecasts.
By the early 1990s, cellular technology had become so popular that the first-generation
analog systems could not keep up with the demand. New second-generation digital sys-
tems were developed to address the capacity shortfall. In the United States, three digital
technologies were standardized and deployed: IS-136 (a TDMA technology utilizing the
AMPS 30-kHz structure), IS-95 (a 1.25-MHz CDMA carrier scheme) and GSM (the
European 200-kHz TDMA standard). This multi-standard wireless environment provided
a unique proving ground for the three technologies. While IS-136 and IS-95 engaged
in ‘standards wars,’ GSM gained a foothold in America. At the same time, GSM was
achieving global acceptance because it offered a rich selection of capabilities and fea-
tures that provided real incremental revenues for operators. As more and more countries
adopted the technology, GSM experienced tremendous economies of scale for everything
from chipsets to handsets, infrastructure and applications.

While the industry continued to experience stellar growth, American manufacturer
dominance was challenged by Nordic companies, especially for the GSM technology.
They brought to the United States, innovative, competitively priced products, backed by
talented communications professionals with years of experience in designing, manufac-
turing, engineering and installing cellular equipment and systems throughout the world.
By the late 1990s, the Internet was pervasive and the wireless industry looked to mobile
data as the growth opportunity. Once again, the industry undertook the task of defining
new wireless systems—this third generation, 3G, was to be based on packet data. Three
new wireless standards emerged; CDMA2000 (evolution of IS-95), EDGE (evolution of
xx Forewords
GSM for existing spectrum) and WCDMA (evolution of GSM for new spectrum using a
5-MHz WCDMA carrier).
The evolution of GSM to 3G is about gradually adding more functionality, possibilities
and value to the existing GSM network and business. The evolution begins with an
upgrade of the GSM network to 2.5G by introducing GPRS technology. GPRS provides
GSM with a packet data air interface and an IP-based core network. EDGE is a further
evolutionary step of GSM packet data. EDGE can handle about three times more data
subscribers than GPRS, or triple the data rate for one end-user. EDGE can be achieved
through a very fast and cost-effective implementation. The only requirement is to add
EDGE-capable transceivers and software.
With the continuation of EDGE standardisation towards GERAN (GSM/EDGE Radio
access network), EDGE will achieve a full alignment with WCDMA. The goal for EDGE
is to boost system capacity, both for real-time and best-effort services, and to become
perfectly competitive with other 3G technologies.
What emerges with these evolutionary steps from GSM to GPRS, EDGE and WCDMA
is a seamless 3G UMTS (Universal Mobile Telecommunications System) Multi-Radio
network, one that maximizes the investments in GSM and GPRS.
It stands to reason that both EDGE and WCDMA will be mainstream 3G UMTS
products from Nordic companies. This book, written by engineers from one of these
Nordic companies, is an authoritative treatise on GSM evolution to 3G. The book provides

an in-depth performance analysis of current and future GSM speech and GPRS/EDGE
packet data functionality. Furthermore, the concept of a 3G UMTS Multi-Radio network
(GSM/EDGE/WCDMA) is presented in depth as the best solution for wireless operators
to evolve their networks towards 3G.
Times change, but some things do not. Nordic companies have been at the forefront
of wireless communications for more than a half of a century. They have earned their
pre-eminent position in the industry. I encourage you to listen to what this book has
to say.
Mike Bamburak
VP Technology Architecture & Standards
AT& T
Over the years, scientists and dreamers have revolutionised the way we work and live
through great inventions. Almost as quickly as news of the inventions spread, soothsayers
rose to highlight real or imagined barriers to the success, popularity or the long-term use
of these products and services. As occurred with electricity, the automobile and the televi-
sion, soothsayers often misread the long-term impact of these inventions, underestimating
qualities that led to their long-term success and adoption by the masses. Ultimately, all
three inventions had tremendous social and economic impact on global societies, and the
soothsayers were proven to have undervalued the importance of these great inventions
over time.
In a slightly different way, the future of EDGE has been understated, underestimated,
and undervalued by the latest pundits. Over the last few years, several global wireless orga-
nizations, including the GSA, the UWCC and now 3G Americas have stood their ground
Forewords xxi
as advocates for EDGE because of the merits of the technology and its value to operators
and customers as a spectrally efficient and cost-effective solution for third-generation (3G)
wireless services. 3G Americas is firm in their belief that a comparative review of how
EDGE meets three key criteria, performance, cost and the ease of transformation to 3G,
will show that EDGE is indeed a superior technology choice.
The Reality of EDGE

On October 30, 2001, Cingular Wireless with its vendor partners announced its commit-
ment to become the first operator in the world to deploy EDGE at 850 and 1900 MHz.
With over 22-million wireless customers, Cingular is a major player in the global wire-
less marketplace. The reasons cited by Cingular for its EDGE selection included capacity
and spectral efficiency competitive with any other technology choice (performance), the
ability to deploy 3G EDGE in existing spectrum including 850 MHz (cost), a total capital
cost of about $18 to $19 per Point of Presence (POP) in TDMA markets with plenty of
go-forward capacity (cost), ridiculously low cost to upgrade the GSM network (only 10 to
15 percent of the network’s cost), the enormous economies of scale and scope offered by
the family of GSM technologies, ensuring the availability of equipment and applications
at the lowest possible cost, and a transition path through GSM and GPRS achievable
seamlessly through the use of the GAIT terminal (GSM-TDMA phone) that will ease
transformation and result in customer retention.
Similarly, almost a year before Cingular’s announcement, AT&T Wireless Services
announced its commitment to EDGE. As of November 2001, reported operator com-
mitments to EDGE in the Americas encompassed hundreds of millions of potential
customers. These commitments validate the future of this third-generation technology.
Cingular’s commitment to EDGE in the 850-MHz band sets the stage for an accelerated
uptake by operators throughout the Western Hemisphere. Regional US operators and many
Latin American operators will find the opportunity to deploy EDGE in 850 MHz espe-
cially appealing. Furthermore, these commitments increase the possibility that Europe
will recognize that EDGE’s capacity and cost qualities make it an important comple-
mentary technology to WCDMA. As spectrum shortages inevitability occur in Europe,
EDGE will provide an excellent solution for GSM operators as a complement to their
WCDMA networks.
Benefits of EDGE
EDGE will benefit operators and customers because it is a cost-effective solution for 3G
services. Cost efficiency is enabled by the economies of scope and scale demonstrated
by the GSM family of technologies, including both TDMA and GSM, which represented
nearly 80% of the world’s digital cellular subscribers in 2001. More than half a billion

GSM phones existed by mid-year 2001, and within a mere four months that number
had risen by 100 000 000 phones to 600 000 000. Bill Clift, Chief Technology Officer of
Cingular, noted that the cost differential between GSM and CDMA devices gets fairly
significant at $15 to $20 per handset times millions of handsets each year. The economies
of scale played a key role in the Cingular decision.
Another major benefit of EDGE cited by operators is that it enables TDMA and
GSM carriers to offer 3G services while still realizing lower costs due to higher spec-
tral efficiency and higher data rates. With the implementation of Adaptive Multi-Rate
xxii Forewords
(AMR) Vocoders, and Frequency Hopping, GSM is competitive with CDMA on spectral
efficiency, which translates into higher capacity and faster data rates. EDGE offers trans-
mission speeds of 384 kbps—fast enough to support full motion video—and throughput
capacity 3 to 4 times higher than GPRS. Thus, EDGE is fast, EDGE is efficient and
EDGE performs.
Additionally, the opportunity for international roaming with the GSM family of tech-
nologies offers yet another major incentive for operators to provide their customers with
seamless communications services. Since EDGE and WCDMA will operate on the same
GPRS core network, the EDGE/WCDMA customer will be able to travel the world
seamlessly, staying connected with one number and one device.
Conclusion
EDGE will contribute to a bright future for 3G services, a vision shared by major analysts
and industry groups. The Strategist Group predicts that revenue from wireless data will
reach $33.5 billion globally by 2004. Frost & Sullivan expects that the proportion of
operator revenues derived from non-voice services will be in excess of 45% by 2006. A
UMTS Forum survey has estimated that non-voice revenues may take over voice revenues
by 2004, while simple voice revenues will remain a critical revenue component comprising
34% of annual revenues in 2010. The UMTS Forum study also predicts that 3G revenues
of $37.4 billion in 2004 will increase to $107 billion by 2006. All in all, predictions may
vary but the consensus is clear that results will be positive for third-generation services.
This work offers the reader more than an evolutionary technical strategy of GSM’s

transition to 3G. It also provides a set of benchmarks for a core evaluation of the merits
of EDGE as a central component of the wireless industry’s fulfilment of its promise for
higher degrees of service and convenience. This process is already being established, as
evidenced by the first live EDGE data call completed by Nokia and AT&T Wireless on
November 1, 2001. The connection of an EDGE handset with a laptop to the Internet for
web browsing and streaming applications marked the first successful completion using
EDGE 8-PSK modulation in both directions in the air interface. Indeed, it is another
sign that EDGE will flourish in this new billion-dollar marketplace as a leading 3G
technology in the Americas owing to its performance, cost, interoperability and ease of
transformation. EDGE will outlast the neigh-sayers and in the long term, EDGE will far
exceed expectations. And just as electricity, the automobile and the television changed
our lives, EDGE will change our lives by providing 3G services for the masses.
Chris Pearson
Executive Vice President
3G Americas
I am honored to have been asked to provide a foreword and a few thoughts for the
second edition of this book, which, although I am sure has been useful to the tried and
true GSM operators, vendors, and researchers worldwide, has been particularly invaluable
to those operators, like ourselves at Cingular Wireless, who have been intimately involved
in actually deploying the latest new GSM networks. At Cingular Wireless, we have been