Third generation of wireless systems

Third Generation Wireless Systems,

Volume 1

Post-Shannon Signal Architectures

For a complete listing of the Artech House Mobile Communications Library,

turn to the back of this book.

Third Generation Wireless Systems,

Volume 1

Post-Shannon Signal Architectures

George M. Calhoun

Artech House

Boston • London

www.artechhouse.com

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.

British Library Cataloguing in Publication Data

A catalog record for this book is available from the British Library.

Cover design by Igor Valdman

© 2003 ARTECH HOUSE, INC.

685 Canton Street

Norwood, MA 02062

All rights reserved. Printed and bound in the United States of America. No part of this

book may be reproduced or utilized in any form or by any means, electronic or

mechanical, including photocopying, recording, or by any information storage and

retrieval system, without permission in writing from the publisher.

All terms mentioned in this book that are known to be trademarks or service marks

have been appropriately capitalized. Artech House cannot attest to the accuracy of this

information. Use of a term in this book should not be regarded as affecting the validity

of any trademark or service mark.

10 9 8 7 6 5 4 3 2 1

Contents

Preface xiii

Reference xvii

1 The Golden Age of Wireless 1

1.1 The First Golden Age, 1890–1940 1

1.2 A Quiet Interregnum, 1940–1990 4

1.3 The Digital Radio Revolution 7

1.4 The Capacity Crisis, 1995–2001 10

1.5 The New Golden Age 19

References 21

2 Shannon 23

2.1 “Our Shannon” 23

2.2 Claude Elwood Shannon (1916–2001) 26

2.3 Shannon Theory 31

2.3.1 Nyquist 32

v

2.3.2 The Importance of Noise 35

2.3.3 Discrete Sources and Entropy 36

2.3.4 Channel Coding and Error Control 37

2.3.5 Quantization of Analog Sources 38

2.3.6 Power and Bandwidth 39

References 44

3 Wireless Systems Design: Problems

and Parameters 47

3.1 Three Unique Design Constraints 47

3.1.1 The Basic Communications Link and the

Nonengineerable Wireless Channel 48

3.1.2 The Nonclonable Wireless Channel and the

Challenges of Multiple Access 55

3.1.3 The Conundrums of Interference 59

3.2 The Basic Parameters: Channel, Signal, and Noise 61

3.2.1 The Primary Signal/Channel Dimensions: Space,

Time, and Frequency 62

3.2.2 Noise and Interference 65

3.2.3 Self-Interference 69

3.2.4 Orthogonality 72

3.2.5 Secondary Signal Dimensions: Amplitude and Angle 73

3.2.6 Tertiary Signal Dimensions: Signal Structure 77

3.2.7 Quaternary Signal Dimensions: The Structure of

the Payload 79

3.2.8 The Layered Signal and the Primacy of the Physical

Layer 81

3.3 First- and Second-Generation Wireless

Architectures 83

3.3.1 First-Generation Systems: Power Versus Noise 83

3.3.2 The Cellular Revolution: Reorganizing the

S-Dimension 84

3.3.3 Second-Generation Systems: Digital Extensions—

Reorganizing the T-Dimension 85

vi Third Generation Wireless Systems: Post-Shannon Signal Architectures

3.3.4 On the Threshold of 3G 86

References 87

4 Third Generation Systems: Physical Layer

Technology Strategies 89

4.1 From Interference Avoidance to Interference

Management 89

4.1.1 The Penalty for Orthogonality 90

4.1.2 Taming the Interference 98

4.2 Signal Hardening Techniques 100

4.2.1 Error Correction (Channel Coding) 101

4.2.2 Diversity Techniques 102

4.2.3 Convolutional Techniques 107

4.3 Signal Shaping Techniques 111

4.3.1 Compression: Source Coding 112

4.3.2 Baseband Shaping 113

4.3.3 Spectrum Shaping 114

4.3.4 Beam Forming: Smart Antennas 115

4.4 Signal Recovery 115

4.4.1 Cancellation of T-Interference (I): Equalization 116

4.4.2 Cancellation of T-Interference (II): Multipath

Combining 116

4.4.3 Cancellation of S-Interference: Spatially Selective

Receivers 118

4.5 Beyond Orthogonality: Convolved Wireless

Architectures and Design Principles 118

4.5.1 Signal Spreading 120

4.5.2 Interference Averaging 123

4.5.3 Signal Averaging: Noise-Like Signals 127

4.5.4 Interference Cancellation: Unscrambling the

Omelet 128

4.5.5 Adaptive Signal Design 131

4.5.6 Convolutional Signal Structures 134

Contents vii

References 135

5 Signal Hardening Techniques 139

5.1 Coding: A Vast Philosophy 140

5.1.1 Coding: The Standard View 144

5.1.2 A Deeper Look 146

5.1.3 Quantization: Many-to-One Mapping 147

5.1.4 Nonlinearity and Threshold Effects 153

5.1.5 Coding as Redundancy Construction: One-to￾Many Mapping 159

5.1.6 Signal Expansion: Channel Coding as Noise

Averaging 163

5.1.7 Message Space Expansion: Channel Coding as

Signal Geometry 166

5.2 Basic Channel Coding Strategies 175

5.2.1 Block Codes 175

5.2.2 Convolutional Codes 177

5.2.3 Decoding Algorithms 182

5.2.4 Performance: Coding Gain 184

5.3 Advanced Coding Strategies 186

5.3.1 Interaction Between Channel Coding and Source

Compression Strategies 186

5.3.2 Channel Characteristics and the Choice of

Coding Schemes: Burst Errors 190

5.3.3 Soft Decision Techniques 194

5.3.4 Side Information 199

5.3.5 Pilot Signals 201

5.3.6 Trellis Coding 202

5.3.7 Hierarchical Coding Structures: Concatenated

Coding, Turbo Coding, and Parallel Coding 206

5.4 Diversity Techniques 213

5.4.1 Frequency Diversity 217

5.4.2 Time Diversity 221

viii Third Generation Wireless Systems: Post-Shannon Signal Architectures

5.4.3 Space Diversity 222

5.5 Convolutional Signals 223

References 227

6 Signal Shaping Techniques (Transmitter￾Oriented Strategies) 231

6.1 Concepts of Efficient Transmission: Compression

and Shaping 231

6.2 Signal Nonlinearities: A Conundrum 236

6.3 Compression: Post-Shannon Source Coding

Strategies 241

6.3.1 Lossless Compression 244

6.3.2 Lossy Compression 251

6.3.3 Perceptual Coding 254

6.3.4 Correlative Quantization 256

6.3.5 Source Modeling 265

6.4 Baseband Signal Shaping 267

6.5 RF Signal Shaping 272

6.5.1 Bandwidth-Efficient Modulation 272

6.5.2 Linearized RF Systems 276

6.6 Smart Antenna Technologies (Transmission) 281

6.7 Adaptive Link Technologies 284

6.7.1 Adaptive Power Control 285

6.7.2 Adaptive Time Alignment 289

6.7.3 Adaptive Modulation 292

References 295

7 Signal Recovery Techniques (Receiver￾Oriented Strategies) 299

7.1 Logical-Level Signal Recovery Strategies: The

Active Receiver 303

7.1.1 Reliability Assessment and Erasure Strategies 308

Contents ix

7.1.2 Residual Redundancy and Codeless Error

Detection 310

7.1.3 Error Concealment 312

7.2 The Transfer Function: Modeling the Channel 315

7.2.1 Transmitter-Assisted Acquisition of the Transfer

Function 321

7.2.2 Blind Acquisition of the Transfer Function 321

7.2.3 The Source Model as a Basis for Blind Acquisition 322

7.2.4 A Priori Knowledge of the Physical Channel as a

Basis for Blind Acquisition 323

7.2.5 “Multipath Is Your Friend” 323

7.3 One-Dimensional Signal Recovery Strategies:

Equalization and RAKE Receivers 342

7.3.1 Subtractive Techniques: Equalization 344

7.3.2 Constructive Techniques: RAKE Architectures 351

7.4 Spatial Techniques: Array Processing 355

7.5 Multidimensional Signal Recovery Strategies 360

7.5.1 Multidimensional Equalization 361

7.5.2 Multidimensional RAKE Receivers 361

7.6 Multiuser Detection and Interference Cancellation 362

References 367

8 Signal Expansion Strategies: Beyond

Orthogonality 377

8.1 An Analogy: Stereo and Beyond 380

8.2 Spreading Forced Through Multiplication in the

Time Domain: Direct-Sequence Spread Spectrum 393

8.3 Spreading Forced Through Multiplication in the

Frequency Domain: OFDM and Multicarrier

CDMA 406

8.3.1 Constructing the Complex F-Domain Signal 406

8.3.2 Tailoring the OFDM Signal to Fit the “Shape”

of the Channel Transfer Function 415

x Third Generation Wireless Systems: Post-Shannon Signal Architectures

8.3.4 Multicarrier CDMA 428

8.4 Forced Spatial Spreading: Creating Artificial

Multipath 435

8.4.1 Creating Multipath: The Physical Aspect of Spatial

Spreading—Transmitter Diversity and Multiple￾Input/Multiple-Output (MIMO) Channels 436

8.4.2 The Coding Aspect of Spatial Spreading: Space–

Time Codes 440

8.4.3 Space–Time Architectures: Ultimate RF? 443

References 446

9 Epilogue:

The Red Queen and the Kitten 453

9.1 Countertrends 455

9.1.1 Verdú 455

9.1.2 Abramson 457

9.1.3 Viterbi et al. 460

9.1.4 Chuang and Sollenberger 461

9.2 Predictions 463

9.3 Whither Shannon? 465

References 469

About the Author 471

Index 473

Contents xi

.

Preface

This is a book about the technological foundations of the next generation of

wireless networks, including the so-called third generation, or 3G, for the

mobile Internet. It is a vast subject, encompassing a variety of new services,

applications, and air interfaces. In this book I have focused on the lowest

layers of the system: how the wireless information-bearing signal is con￾structed, and the choice of physical-layer transmission techniques and cod￾ing strategies capable of withstanding the extreme environment of the

wireless channel.

I have tried to emphasize the general lines of technical development

and core principles, rather than focusing on specific air interface standards.

Underneath the wide variety of prospective technical solutions, there are a

few common engineering challenges. The third generation and other new

wireless standards must operate at much higher bit rates, and must deliver a

much cleaner signal. Conventional approaches to managing the physical

channel will not suffice. We are entering a new historical phase of intensive

innovation, in which I believe that wireless technology will again become the

leading edge for communications engineering in general. This book is

planned as the first of three or four related volumes on this next generation

wireless. Subsequent volumes will deal with higher-level architectures

designed to support multiple access in wireless networks, as well as the

emerging suite of wireless applications (beyond voice) that will soon trans￾form these networks into an infrastructure for true multimedia services.

xiii

This is also a book about an intellectual revolution, the outlines of

which are just now becoming visible. The source of illumination is the chal￾lenge of optimizing the wireless channel (although its effects will undoubt￾edly extend to all corners of communications and information science).

Today we are beginning to realize that many of our received ideas no longer

entirely fit the physical and engineering realities of the systems we are build￾ing. As has often happened in the history of science, engineering solutions

are outpacing the standard models. Our ways of thinking about communica￾tions problems are in flux, and our explanations are far less coherent than we

like to imagine.

The historical reference point—our ancien regime—is a body of

thought we may call the Shannon theory.1

For more than 50 years, the field

of communications engineering has oriented itself around the work and

insights of Claude Elwood Shannon (1916–2001). His seminal 1948 article

on the mathematical theory of communication packaged a set of powerful

ideas about information, coding, and the communications process that from

the very beginning has seemed both unusually cogent and nicely self￾contained, or complete, at least in principle. Although the Shannon theory

was for a long time something of an intellectual curiosity (rather than a prac￾tical engineering program), it held sway over us as a kind of ontological proof

(an “existence proof”) that laid down a metaphysical foundation for what has

become the core of our “information age,” a technological infrastructure

based on the manipulation of digital signals. It seemed to draw clean limits

and boundaries, to define the game board of communications—just as a

chess game is defined by 64 squares and a few simple rules of movement for

the chess pieces. The Shannon theory might not “solve” the engineering

equations of interest (any more than the basic rules of chess disclose the

winning strategy), but it created the right sort of finiteness for taming a com￾plex subject, and a deterministic framework to prove that a solution does

exist. A decade or so after Gödel had shown that arithmetic and other axio￾matic systems in mathematics were hopelessly open-ended, Shannon began

the perpetual search for the grail of perfect computability all over again.

There is often a mundane problem at the fulcrum of the paradigm

shift. Copernican astronomy was motivated (it is said) by the problems of

calendar construction. The practical spur for Newtonian mechanics is

xiv Third Generation Wireless Systems: Post-Shannon Signal Architectures

1. “Shannon theory” is admittedly a loose shorthand for a set of ideas and techniques

that encompass a variety of disciplines and may be referred to variously as informa￾tion theory, communications theory, or coding theory, although all of these terms

have been defined with different scopes as well.