TURBO CODES

Principles and Applications

 THE KLUWER INTERNATIONAL SERIES
IN ENGINEERING AND COMPUTER SCIENCE
 TURBOCODES
Principles and Applications

Branka Vucetic
The University ofSydney
Sydney, Australia

Jinhong Yuan
The University ofNew South Wales
Sydney, Australia

"
~.

Springer Science+Business Media, LLC

Library of Congress Cataloging-in-Publication

Vucetic, Branka.
Turbo codes : principles and applications / Branka Vucetic, Jinhong Yuan.
p. cm. -- (The Kluwer international series in engineering and computer science
; SECS 559.)
lncludes bibliographical references and index.
ISBN 978-1-4613-7013-0 ISBN 978-1-4615-4469-2 (eBook)
DOI 10.1007/978-1-4615-4469-2
1. Coding theory. 2. Signal theory (Telecommunication) I. Yuan, Jinhong, 1969- II.
Title. III. Series.

TK5102.92. V83 2000

003'.54--dc21
OO-Q33104

Copyright@ 2000 by Springer Science+Business Media New York

Originally published by Kluwer Academic Publishers, New York in 2000
Softcover reprint of the hardcover 1st edition 2000
AII rights reserved. No part of this publicatlOn may be reproduced, stored in a retrievat
system or transmitted in any form or by any means, mechanical, photo-copying, recording,
or otherwise, without the prior written permission of the publisher,
Springer Science+Business Media, LLC.

Printed an acid-free paper.

Contents

List of Acronyms xi

List of Figures xiii

List of Tables xxiii

Preface xxv

1 Introduction 1
1.1 Digital Communication System Structure. 2
1.2 Fundamental Limits . . . . . . . . . . . . 5

2 Block Codes 13
2.1 Block Codes 13
2.2 Linear Systematic Block Codes 15
2.3 Parity Check Matrix . . . . . . 16
2.4 The Minimum Distance of a Block Code 17
2.5 Maximum Likelihood Decoding of Block Codes for a
BSC Channel . . . . . . . . . . . . . . . . . . . .. 18
2.6 Maximum Likelihood Decoding of Block Codes for a
Gaussian Channel. . . . . . . . . . 19
2.7 Weight Distribution of Block Codes . . . . . . 20
2.8 Performance Upper Bounds . . . . . . . . . . 23
2.8.1 Word Error Probability Upper Bounds 23
2.8.2 Bit Error Probability Upper Bounds 26
2.9 Coding Gain . . . . . . . . . . . . . . . . . . 28
2.10 Soft Decision Decoding of Block Codes . . . . 30
2.11 Trellis Structure of Linear Binary Block Codes. 30
vi CONTENTS

3 Convolutional Codes 37
3.1 Introduction . . . . .............. 37
3.2 The Structure of (n,l) Convolutional Codes 38
3.3 The Structure of (n, k) Convolutional Codes 43
3.4 Systematic Form .. 45
3.5 Parity Check Matrix 50
3.6 Catastrophic Codes . 51
3.7 Systematic Encoders 53
3.8 State Diagram. . . . 58
3.9 Trellis Diagram . . . 60
3.10 Distance Properties of Convolutional Codes 62
3.11 Weight Distribution of Convolutional Codes 63
3.12 Punctured Convolutional Codes . . . . . . . 66

4 Turbo Coding Performance Analysis and Code De-

sign 73
4.1 Introduction . . . . . . . 73
4.2 Turbo Coding . . . . . . 74
4.2.1 A Turbo Encoder 74
4.2.2 Interleaving ... 76
4.2.3 Trellis Termination 77
4.2.4 High Rate Turbo Codes 78
4.3 Performance Upper Bounds of Turbo Codes 80
4.3.1 Conditional WEF's of Average Turbo Codes 80
4.3.2 Conditional WEF's of Component Codes . . 82
4.3.3 Average Upper Bounds on Bit Error Proba-
bility . . . . . . . . . . . . . . . 84
4.3.4 Interleaving Performance Gain . 87
4.3.5 Effective Free Distance . . . . 90
4.4 Turbo Code Performance Evaluation 92
4.5 Turbo Code Design . . . . . . . . . . . 98
4.5.1 Turbo Code Design at High SNR's 100
4.5.2 Turbo Code Design at Low SNR's . 103
4.5.3 Simulation Results ......... 105
4.6 Serial Concatenated Convolutional Codes. 107
4.6.1 A Serial Concatenated Encoder .. 107
4.6.2 Performance Analysis and Code Design . 108
CONTENTS vii

5 Trellis Based Decoding of Linear Codes 117

5.1 Introduction . . . . . . 117
5.2 System Model . . . . . 118
5.3 Optimization Criteria. 120
5.4 The Viterbi Algorithm 122
5.5 The Bidirectional Soft Output Viterbi Algorithm 126
5.6 Sliding Window SOYA . . . . . 135
5.7 The MAP Algorithm. . . . . . 138
5.8 The Max-Log-MAP Algorithm. 149
5.9 The Log-MAP Algorithm. . . . 151
5.10 Comparison of Decoding Algorithms 152

6 Iterative Decoding 157

6.1 Optimum Decoding of Thrbo Codes. . . . . . . .. 157
6.2 Iterative Decoding of Turbo Codes Based on the
MAP Algorithm. . . . . . . . . . . . . . . . 159
6.3 The Effect of the Number of Iterations on Turbo
Code Performance . . . . . . . . . . . . . . . . . . 164
6.4 The Effect of Interleaver Size on Turbo Code Perfor-
mance . . . . . . . . . . . . . . . . . . . . . . . .. 166
6.5 The Effect of Puncturing Component Codes on Turbo
Code Performance . . . . . . . . . . . . . . . . . . 167
6.6 Comparison Between Analytical Upper Bounds and
Simulation Results . . . . . . . . . . . . . 169
6.7 Asymptotic Behavior of Turbo Codes . . . . . . .. 170
6.8 Iterative SOYA Decoding of Turbo Codes. . . . .. 171
6.9 Comparison of MAP and SOYA Iterative Decoding
Algorithms. . . . . . . . . . . . . . . . . . . . . .. 177
6.10 Iterative MAP Decoding of Serial Concatenated Con-
volutional Codes . . . . . . . . . . . . . . . . . . . 178
6.11 Iterative SOYA Decoding of Serial Concatenated Con-
volutional Codes . . . . . . . . . . . . . . . . . . . 180
6.12 Serial Concatenated Convolutional Codes with Iter-
ative Decoding .. . . . . . . . . . . . . . . . . .. 182
6.12.1 The Effect ofInterleaver Size and the Number
of Iterations on AWGN Channels . . . . .. 182
viii CONTENTS

6.12.2 The Effect of Memory Order on AWGN chan-

nels. . . . . . . . . . . . . . . . . . . . . .. 184
6.12.3 Comparison of MAP and SOYA Decoding Al-
gorithms on AWGN Channels . . . . . . .. 186

7 Interleavers 193
7.1 Interleaving . . . . . . . . . . . . . . . . 193
7.2 Interleaving with Error Control Coding. 195
7.3 Interleaving in Turbo Coding . . . . . . 196
7.3.1 The Effect of Interleaver Size on Code Per-
formance. . . . . . . . . . . . . . . . . . .. 197
7.3.2 The Effect of Interleaver Structure on Code
Performance . . . . . . . 198
7.3.3 Interleaving Techniques. 200
7.4 Block Type Interleavers 200
7.4.1 Block Interleavers . . . . 200
7.4.2 Odd-Even Block Interleavers . 202
7.4.3 Block Helical Simile Interleavers . 204
7.5 Convolutional Type Interleavers . 206
7.5.1 Convolutional Interleavers 206
7.5.2 Cyclic Shift Interleavers 208
7.6 Random Type Interleavers . . . 209
7.6.1 Random Interleavers . . 209
7.6.2 Non-uniform Interleavers 210
7.6.3 S-random Interleavers . 211
7.7 Code Matched Interleavers . . . 213
7.8 Design of Code Matched Interleavers 214
7.9 Performance of Turbo Codes with Code Matched In-
terleavers . . . . . . . . . . . . . . . . . . . . . . . 220
7.10 Performance of Turbo Codes with Cyclic Shift Inter-
leavers . . . . . . . . . . . . . . . . . . . . . . . .. 222

8 Turbo Coding for Fading Channels 231

8.1 Introduction.......... 231
8.2 Fading Channels . . . . . . . 232
8.2.1 Multipath Propagation 232
8.2.2 Doppler Shift . . . . . 232
8.3 Statistical Models for Fading Channels 233
CONTENTS ix

8.3.1 Rayleigh Fading. . . . 233

8.3.2 Rician Fading . . . . . 235
8.4 Capacity of Fading Channels. 236
8.5 Performance Upper Bounds on Fading Channels 240
8.5.1 Upper Bounds on the Pairwise Error Proba-
bility . . . . . . . . . . . . . . . . . . . . .. 241
8.5.2 Average Upper Bounds on the Bit Error Prob-
ability . . . . . . . . . . . . . . . . 246
8.6 Iterative Decoding on Fading Channels . . 249
8.6.1 Modified MAP Decoding with CSI 252
8.6.2 Modified SOYA Decoding with CSI 254
8.7 Performance Simulation Results on Fading Channels 256
8.7.1 Performance Comparison Between MAP
and SOYA Algorithms on Independent
Fading Channels . . . . . . . . . . . . . 256
8.7.2 Performance Comparison Between Turbo and
Serial Concatenated Codes on Independent
Fading Channels . . . . . . . . . . . . . . . 256
8.7.3 Performance Comparison Between MAP and
SOYA Algorithms on Correlated Fading Chan-
nels. . . . . . . . . . . . . . . . . . . . . .. 258
8.7.4 Performance Comparison Between Turbo and
Serial Concatenated Codes on Correlated Fad-
ing Channels . . . . . . . . . . . . . . . . . 260

9 Turbo Trellis Coded Modulation Schemes 263

9.1 Introduction...................... 263
9.2 Binary Thrbo Coded Modulation . . . . . . . . .. 264
9.2.1 Pragmatic Binary Thrbo Coded Modulation 264
9.2.2 Multilevel Thrbo Coding . . . . . . . . 267
9.3 Thrbo Trellis Coded Modulation. . . . . . . . 270
9.3.1 Schemes with Alternate Puncturing of
Parity Digits . . . . . . . . . . . . . . 270
9.3.2 Log-MAP Decoding Algorithm for Thrbo Trel-
lis Coded Modulation with Punctured Parity
Digits . . . . . . . . . . . . . . . . . . . .. 274
x CONTENTS

9.3.3 SOYA Decoding Algorithm for Turbo

Trellis Coded Modulation with
Punctured Parity Digits . . . . . . . . 278
9.3.4 Performance of Turbo Trellis Coded Modula-
tion with Punctured Parity Digits . . . . .. 279
9.3.5 Schemes with Puncturing of Systematic Bits 282
9.4 I-Q Thrbo Coded Modulation for Fading Channels 288
9.4.1 I-Q Coded Modulation Structure . . . . .. 290
9.4.2 The Decoder . . . . . . . . . . . . . . . . . 292
9.4.3 Performance of I-Q Turbo Coded Modulation
on Fading Channels. . . . . . . . . . . . .. 292

10 Applications of Turbo Codes 297

10.1 Thrbo Codes for Deep Space Communications 297
10.2 Turbo Codes for CDMA2000 . . . . . . . . 299
10.3 Turbo Codes for 3GPP . . . . . . . . . . . 300
10.4 Thrbo Codes for Satellite Communications 302

Index 307
List of Acronyms

3GPP 3rd Generation Partnership Project

APP a posteriori probability

AWGN additive white Gaussian noise

BER bit error rate

BPSK binary phase shift keying

BSC binary symmetric channel

bps bits per second

CCSOS Consultative Committee for Space Data Systems

COMA code division multiple access

CRC cyclic redundancy check

CSI channel state information

GCO greatest common divisor

IOWEF input-output weight enumerating function

IRWEF input-redundancy weight enumerating function

151 intersymbol interference

LLR log-likelihood ratio

xii LIST OF ACRONYMS

MAP maximum a posteriori

ML maximum likelihood

NRC nonrecursi ve convolutional

ODS optimal distance spectrum

PCCC parallel concatenated convolutional code

PSK phase shift keying

QAM quadrature amplitude modulation

RSC recursive systematic convolutional

SCCC serial concatenated convolutional code

SER symbol error rate

SISO soft-input soft-output

SNR signal-to-noise ratio

SOVA soft-output Viterbi algorithm

TCM trellis coded modulation

TTCM turbo trellis coded modulation

UEP unequal error protection

VA Viterbi algorithm
WEF weight enumerating function

WER word error rate

List of Figures

1.1 Model of a digital communication system. . . . .. 2

1.2 Spectral efficiency of various modulation and coding
schemes computed for the bit error rate of 10-5 8

2.1 Coded system model . . . . . . . . . . . 24

2.2 Performance upper bounds for the (7,4) Hamming
code . . . . . . . . . . . . . . . . . . . . 28
2.3 Trellis for the binary (5,3) code . . . . . 33
2.4 Expurgated trellis for the binary (5,3) code. 33

3.1 A rate 1/2 convolutional encoder . . . . . . 38

3.2 A general (n, 1, v) convolutional code feedforward en-
coder. . . . . . . . . . . . . . . . . . . . . . . . .. 41
3.3 Encoder for a (3,2,1) code. . . . . . . . . . . . .. 46
3.4 The controller canonical form of a rational transfer
function a(D)/q(D) . . . . . . . . . . . . . . . . . 48
3.5 The observer canonical form of a rational transfer
function a(D)/q(D) . . . . . . . . . . . . . . . . . 49
3.6 The controller canonical form ofthe systematic (2,1)
encoder with the generator matrix G1(D) . . . . . 49
3.7 The observer canonical form of the systematic (2,1)
encoder with the generator matrix G 1 (D) 50
3.8 Nonsystematic encoder in Example 3.9 . . . . . . , 55
3.9 Systematic encoder in Example 3.9 . . . . . . . .. 56
3.10 A systematic encoder with the generator matrix in
(3.73) . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.11 Observer canonical form of an (n, n - 1) systematic
encoder . . . . . . . . . . . . . . . . . . . . . . . . 58
xiv LIST OF FIGURES

3.12 State diagram for the (2,1) nonsystematic convolu-

tional encoder from Fig. 3.1 . . . . . . . . . . . .. 59
3.13 State diagram for the (2,1) systematic encoder in
Fig. 3.7 .. . . . . . . . . . . . . . . . . . . . . .. 60
3.14 Trellis diagram for the (2,1) nonsystematic encoder
in Fig. 3.1 . . . . . . . . . . . . . . . . . . . . . .. 61
3.15 Augmented state diagram of Fig. 3.12 . . . . . . . 64
3.16 Trellis diagram of a rate 2/3 punctured code pro-
duced by periodically deleting symbols from a rate
1/2 code . . . . . . . . . . . . . . 67
3.17 Encoder for a rate 2/3 code . . . 68
3.18 Trellis diagram of a rate 2/3 code 68

4.1 A turbo encoder. . . . . . 75

4.2 A rate 1/3 turbo encoder. 77
4.3 Trellis termination . . . . 78
4.4 A rate 1/2 turbo encoder. 80
4.5 A compound error path. . 83
4.6 Bit error probability upper bounds for a turbo code
with interleaver size 500 .. . . . . . . . . . . . .. 86
4.7 Bit error probability upper bounds for a turbo code
with various interleaver sizes 91
4.8 Turbo encoder TC1 . . . . . . . . . . . . . . . . .. 94
4.9 Turbo encoder TC2 . . . . . . . . . . . . . . . . .. 94
4.10 Distance spectra for component code of TC1 and
turbo code TC1 with interleaver sizes of 20 and 50 . 95
4.11 Bit error probability upper bounds for component
code of TC1 and turbo code TC1 with interleaver
sizes of 20 and 50 . . . . . . . . . . . . . . . . . .. 96
4.12 Relative contributions of various distance spectral
lines to overall bit error probability for turbo code
TC1 with interleaver size 20 . . . . . . . . . . . .. 97
4.13 Relative contributions of various distance spectral
lines to overall bit error probability for turbo code
TC1 with interleaver size 50 . . . . . . . . . . . .. 98
4.14 Distance spectra for turbo codes TC1 and TC2 with
interleaver sizes of 20 and 50 . . . . . . . . . . . .. 99
LIST OF FIGURES xv

4.15 Bit error probability upper bounds for turbo codes

TC1 and TC2 with interleaver sizes of 20 and 50.. 100
4.16 Distance spectra for ODS turbo codes with inter-
leaver size 40 . . . . . . . . . . . . . . . . . . . .. 105
4.17 Bit error probability upper bounds for ODS turbo
codes with interleaver size 40 . . . . . . . . . . . . 106
4.18 Performance of ODS and BM turbo codes with rate
1/3 and memory order 4 on AWGN channels. 107
4.19 A serial concatenated encoder 108

5.1 System model . . . . . . . . . 118

5.2 A convolutional encoder and its graphical represen-
tation . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.3 The branch metrics in Example 5.1 . . . . . . . .. 128
5.4 The survivors and their path metrics in Example 5.1 128
5.5 The branch metrics in Example 5.2 . . . . . . . .. 133
5.6 The forward recursion in Example 5.2, the ML path
is shown by the thick line . . . . . . . . . . . . . . 133
5.7 The backward recursion in Example 5.2, the ML
path is shown by the thick line. . . . . . . . . . .. 133
5.8 Forward and Backward processing for the simplified
SOYA . . . . . . . . . . . . . . . . . . . . . . .. 137
5.9 A rate 1/2 memory order 2 RSC encoder. . . .. 139
5.10 State transition diagram for the (2,1,2) RSC code 140
5.11 Trellis diagram for the (2,1,2) RSC code . . . .. 141
5.12 Graphical representation of the forward recursion 146
5.13 Graphical representation of the backward recursion 147
5.14 Trellis diagram for the encoder in Example 5.3 150
5.15 Performance comparison of MAP and SOYA 154

6.1 Basic turbo encoder. . . . . . . . . . . . . . 158

6.2 An iterative turbo code decoder based on the MAP
algorithm . . . . . . . . . . . . . . . . . . . . . .. 160
6.3 BER performance of a 16 state, rate 1/3 turbo code
with MAP algorithm on an AWGN channel, inter-
leaver size 4096 bits, variable number of iterations. 164
xvi LIST OF FIGURES

6.4 BER performance of a 16 state, rate 1/3 turbo code

with MAP algorithm on an AWGN channel, inter-
leaver size 16384 bits, variable number of iterations 165
6.5 BER performance of a 16 state, rate 1/3 turbo code
with MAP algorithm on an AWGN channel, inter-
leaver size N, the number of iterations 18. . . . .. 167
6.6 BER performance of a 16 state, rate 1/2 turbo code
with MAP algorithm on an AWGN channel, inter-
leaver size N, the number of iterations 18. . . . .. 168
6.7 BER performance of a 16 state, rate 2/3 turbo code
with MAP algorithm on an AWGN channel, inter-
leaver size N, the number of iterations 18 . . . . . 169
6.8 Simulation result of a 16 state, rate 1/3 turbo code
with MAP, interleaver size 1024 bits, variable num-
ber of iterations I and the theoretical bound on an
AWGN channel. . . . . . . . . . . . . . . . . . . .. 170
6.9 Simulation result of a 16 state, rate 1/3 turbo code
with MAP, interleaver size 1024 bits, the number of
iterations 10 and the theoretical bound on an AWGN
channel. . . . . . . . . . . . . . . . . . . . . . . .. 172
6.10 An iterative turbo code decoder based on the SOYA
algorithm . . . . . . . . . . . . . . . . . . . . . .. 174
6.11 BER performance of a 16 state, rate 1/3 turbo code
with MAP, Log-MAP and SOYA algorithm on an
AWGN channel, interleaver size 4096 bits, the num-
ber of iterations 18 . . . . . . . . . . . . . . . . .. 178
6.12 Iterative MAP decoder for serial concatenated codes 179
6.13 Iterative SOYA decoder for serial concatenated codes 181
6.14 Performance of a rate 1/3 serial concatenated code,
with a rate 1/2, 4 state nonrecursive convolutional
code as the outer code, a rate 2/3, 4 state recursive
convolutional code as the inner code, AWGN chan-
nel, SOYA decoding algorithm, various interleaver
size N, and the number of iterations 20 . . . . . . 184
LIST OF FIGURES xvii

6.15 Comparison of a rate 1/3, memory order 2 turbo

code with interleaver size 4096 bits and a rate 1/3
serial concatenated code with memory order 2 outer
code, interleaver size 4096 bits on an AWGN channel,
SOYA decoding algorithm, the number of iterations
18. .......................... 185
6.16 BER performance of a rate 1/3 serial concatenated
code with rate 1/2, 4 state outer code and rate 2/3, 4
state inner code with SOYA algorithm on an AWGN
channel, interleaver size 4096 bits, variable number
of iterations . . . . . . . . . . . . . . . . . . . . .. 186
6.17 Comparison of a rate 1/3 turbo code for different
memory order with SOYA algorithm on an AWGN
channel, interleaver size 1024 bits, the number of it-
erations 12 . . . . . . . . . . . . . . . . . . . . . .. 187
6.18 Comparison of a rate 1/3 turbo code for different
memory order with SOYA algorithm on an AWGN
channel, interleaver size 4096 bits, the number of it-
erations 18 . . . . . . . . . . . . . . . . . . . . . .. 188
6.19 Comparison of a rate 1/3 serial concatenated code
for different outer code memory order with SOYA al-
gorithm on an AWGN channel, interleaver size 1024
bits, the number of iterations 12. . . . . . . . . . . 188
6.20 Comparison of a rate 1/3 serial concatenated code
for different outer code memory order with SOYA al-
gorithm on an AWGN channel, interleaver size 4096
bits, the number of iterations 18 . . . . . . . . . .. 189
6.21 Performance comparison of MAP and SOYA for a
rate 1/3 serial concatenated convolutional code 189

7.1 An interleaver device . . . . . . . . . . . . . . . 194

7.2 An interleaver mapping. . . . . . . . . . . . . . 195
7.3 Distance spectra for a turbo code with various inter-
leaver sizes. . . . . . . . . . . . . . . . . . . . . .. 198
7.4 Bit error probability upper bounds for a turbo code
with various interleaver sizes 199
7.5 A block interleaver . . . . . . . . . . . . . . . . .. 201
xviii LIST OF FIGURES

7.6 A weight-4 square input pattern of block interleavers 202

7.7 A convolutional interleaver and deinterleaver . .. 207
7.8 A convolutional interleaver with L = 3 and B = 2. 208
7.9 A cyclic shift interleaver . . . . . . . . . . . . . .. 209
7.10 A general m-stage shift register with linear feedback 210
7.11 A weight-2 input sequence pattern. . . . . . . . .. 217
7.12 A weight-4 input sequence pattern. . . . . . . . .. 218
7.13 BER performance of the 4-state, rate 1/3, (1, 5/7)
turbo code with random, S-random and code matched
interleavers on an AWGN channel . . . . . . . . .. 221
7.14 BER performance of the 8-state, rate 1/3, (1, 17/15)
turbo code with random, S-random and code matched
interleavers on an AWGN channel. . . . . . . . .. 222
7.15 BER performance of the 16-state, rate 1/3, (1,33/31)
turbo code with random, S-random and code matched
interleavers on an AWGN channel. . . . . . . . .. 223
7.16 BER performance ofthe 16-state, rate 1/3, (1,33/31)
turbo code with S-random and cyclic shift inter-
leavers on an AWG N channel .. 224

8.1 The pdf of Rayleigh distribution. 234

8.2 The pdf of Rician distributions with various K . 237
8.3 Capacity of independent Rayleigh fading channels
with coherent BPSK signalling. . . . . . . . . . .. 239
8.4 Coded system block diagram . . . . . . . . . . . .. 241
8.5 Bit error probability upper bound for the 4 state,
rate 1/3 turbo code with interleaver size 100 on in-
dependent Rician fading channels with ideal channel
state information. The curves are for Rician chan-
nels with K =0, 2, 5, 50, starting from the top, with
the bottom one referring to an AWGN channel. .. 249
8.6 Bit error probability upper bound for the 4 state,
rate 1/3 turbo code with interleaver size 100 on in-
dependent Rician fading channels without channel
state information. The curves are for Rician chan-
nels with K =0, 2, 5, 50, starting from the top, with
the bottom one referring to an AWGN channel. .. 250
LIST OF FIGURES xix

8.7 Bit error probability upper bound for the 4 state,

rate 1/3 serial code with information size 100 on in-
dependent Rician fading channels with ideal channel
state information. The curves are for Rician chan-
nels with K =0, 2, 5, 50, starting from the top, with
the bottom one referring to an AWGN channel. .. 251
8.8 Bit error probability upper bound for the 4 state,
rate 1/3 serial code with information size 100 on in-
dependent Rician fading channels without channel
state information. The curves are for Rician chan-
nels with K =0, 2, 5, 50, starting from the top, with
the bottom one referring to an AWGN channel. .. 252
8.9 Distance spectrum comparison of the 4 state, rate
1/3 turbo and serial concatenated codes with infor-
mation size 100 . . . . . . . . . . . . . . . . . .. 253
8.10 Bit error probability upper bound comparison of the
4 state, rate 1/3 turbo and serial concatenated codes
with information size 100 on independent Rayleigh
fading channels . . . . . . . . . . . . . . . . . . " 254
8.11 Performance comparison of MAP and SOYA, with
and without CSI, for the 16 state, rate 1/3 turbo
code on an independent Rayleigh fading channel, in-
formation size 1024, the number of iterations 8 . .. 257
8.12 Performance comparison for the 4 state, rate 1/3
turbo and serial codes on an independent Rayleigh
fading channel . . . . . . . . . . . . . . . . . . .. 258
8.13 Performance comparison of MAP and SOYA, with
and without CSI, for the 16 state, rate 1/3 turbo
code on a correlated Rayleigh fading channel, the
fading rate normalized by the symbol rate is 10- 2 ,
information size 1024, the number of iterations 8.. 259
8.14 Performance comparison for the turbo and serial codes
on a correlated Rayleigh fading channel, the fading
rate normalized by the symbol rate is 10- 2 , informa-
tion size N, the number of iterations I 260

9.1 Pragmatic turbo TCM encoder .... 265

xx LIST OF FIGURES

9.2 Pragmatic turbo TCM decoder 265

9.3 16-QAM with Gray mapping. 266
9.4 Multilevel turbo encoder . . . . 268
9.5 Multilevel turbo decoder . . .. 269
9.6 Turbo TCM encoder with parity symbol puncturing 271
9.7 Example of a turbo trellis coded 8-PSK with parity
symbol puncturing . . . . . . . . . . . . . . . . .. 273
9.8 Turbo TCM decoder with parity symbol puncturing 275
9.9 Performance of the rate 2/3, 4-state turbo trellis
coded 8-PSK with various interleaver sizes on an
AWGN channel, SOYA decoding algorithm, the num-
ber of iterations I, bandwidth efficiency 2 bits/s/Hz 280
9.10 Performance of the rate 2/3, 8-state turbo trellis
coded 8-PSK with various interleaver sizes on an
AWGN channel, SOYA decoding algorithm, the num-
ber of iterations I, bandwidth efficiency 2 bits/s/Hz 281
9.11 Performance of the rate 2/3, 16-state turbo trellis
coded 8-PSK with various interleaver sizes on an
AWGN channel, SOYA decoding algorithm, the num-
ber of iterations I, bandwidth efficiency 2 bits/s/Hz 282
9.12 Performance comparison of the Log-MAP and SOYA
for the rate 2/3, 8-state turbo trellis coded 8-PSK
with interleaver size 1024 on an AWGN channel,
bandwidth efficiency 2 bits/s/Hz . . . . . . . . . . 283
9.13 Performance comparison of the Log-MAP and SOYA
for the rate 3/4, 4-state turbo trellis coded 16-QAM
with various interleaver sizes on an AWGN channel,
bandwidth efficiency 3 bits/s/Hz . . . . . . . . . . 284
9.14 Performance comparison of the Log-MAP and SOYA
for the rate 3/4, 8-state turbo trellis coded 16-QAM
with various interleaver sizes on an AWGN channel,
bandwidth efficiency 3 bits/s/Hz . . . . . . . . . . 285
9.15 Performance comparison of the Log-MAP and SOYA
for the rate 3/4, 16-state turbo trellis coded 16-QAM
with various interleaver sizes on an AWGN channel,
bandwidth efficiency 3 bits/s/Hz . . . . . . . . . . 286
LIST OF FIGURES xxi

9.16 Turbo trellis coded 16-QAM with systematic symbol

puncturing. . . . . . . . . . . . . . . . . . . . . .. 287
9.17 Performance comparison of the turbo trellis coded
16-QAM with systematic symbol puncturing and the
pragmatic turbo coded 16-QAM with bandwidth ef-
ficiency 2 bits/s/Hz and interleaver size 32768 on an
AWGN channel, the number of iterations 8 . . . .. 288
9.18 Turbo trellis coded 8-PSK with systematic symbol
puncturing. . . . . . . . . . . . . . . . . . . . . .. 289
9.19 Performance of the turbo trellis coded 8-PSK with
systematic symbol puncturing with bandwidth effi-
ciency 2 bits/s/Hz and interleaver size 16384 on an
AWGN channel, the number of iterations I . . . .. 290
9.20 I-Q turbo trellis coded 16-QAM . . . . . . . . . .. 291
9.21 Performance of the I-Q turbo coded 16-QAM with
bandwidth efficiency 2 bits/s/Hz and various inter-
Ie aver sizes on a Rayleigh fading channel . . . . .. 293
9.22 Performance comparison of the I-Q turbo coded 16-
QAM and the pragmatic turbo coded 16-QAM with
bandwidth efficiency 2 bits/s/Hz and interleaver size
4096 on a Rayleigh fading channel. . . . . . . . 294

10.1 CCSDS turbo encoder block diagram . . . . . . 298

10.2 The reverse link turbo encoder for CDMA2000 . 300
10.3 The turbo encoder for 3GPP . . . . . . . . . . . 301
10.4 The serial concatenated convolutional encoder for
3GPP . . . . . . . . . . . . . . . . . . . . . . . . . 303
List of Tables

2.1 A (6, 3) linear block code ... 15

3.1 Punctured convolutional codes. 70

4.1 Best rate 1/3 turbo codes at high SNR's [14] 103
4.2 Rate 1/3 ODS turbo codes at Low SNR's . 104

5.1 Decoder complexity comparison . . . . . . 153

8.1 Channel capacity limits for coherent BPSK . 240

9.1 Rate 2/3 turbo trellis coded 8-PSK schemes 279

9.2 Rate 3/4 turbo trellis coded 16-QAM schemes 281
Preface

This book grew out of our research, industry consulting and con-
tinuing education courses.
Turbo coding initially seemed to belong to a restricted research
area, while now has become a part of the mainstream telecommu-
nication theory and practice. The turbo decoding principles have
found widespread applications not only in error control, but in de-
tection, interference suppression and equalization.
Intended for use by advanced students and professional engi-
neers, involved in coding and telecommunication research, the book
includes both basic and advanced material. The chapters are se-
quenced so that the knowledge is acquired in a logical and progres-
sive way. The algorithm descriptions and analysis are supported
by examples throughout the book. Performance evaluations of the
presented algorithms are carried out both analytically and by sim-
ulations.
Basic material included in the book has been taught to students
and practicing professionals over the last four years in the form of
senior undergraduate or graduate courses, lecture series and short
continuing education courses.
Most of the presented material is a compilation of the various
publications from the well established literature. There are, how-
ever, original contributions, related to decoding algorithms, inter-
Ie aver design, turbo coded modulation design for fading channels
and performance of turbo codes on fading channels. The bidirec-
tional SOYA decoding algorithm, presented in the book, had been
developed for soft output detection and originally applied to cellu-
lar mobile receivers, but was subsequently modified for decoding of
turbo codes. We have published various versions of the algorithm
xx.vi PREFACE

at a number of conferences, but never as a journal paper, so it has

not been widely known. Its distinguishing features are excellent
performance, which is only slightly worse than the optimum MAP
algorithm, and very low complexity, making it attractive for prac-
tical implementation. It should be of particular value to telecom-
munication system designers.
A great deal of effort has been put into ensuring consistency
and continuity between chapters.

Special Thanks
We would like to thank everyone who has been involved in the
process of writing, proof reading and publishing this book. In par-
ticular we would like to thank Dr Lei Wei, Dr Steven Pietrobon,
Dr Adrian Barbulescu, Dr Miroslar Despotovic, Prof Shu Lin, and
Prof Dusan Drajic for reading the manuscript and providing valu-
able feedback. We would also like to thank Dr Akihisa Ushirokawa
for constructive discussions and Enrico Vassallo for providing the
details on the CCSDS standard.
We are pleased to acknowledge the students' contribution to
advancing the understanding of turbo coding. In particular, we
thank Wen Feng for her work reported in Chapters 6 and 7, Jade
Kim for her work reported in Chapter 6, Mark Tan for his work
reported in Chapter 7 and Lei Wan for her comments on Chapters
5 and 6.
We express our appreciation to Wen Feng for providing simula-
tion results as well as to Maree Belleli and Zhuo Chen for typing
the manuscript and preparing illustrations for the book.
We owe special thanks to the Australian Research Council, NEC,
DSTO, Motorola and other companies, whose support enables grad-
uate students and the staff of Sydney University to pursue contin-
uing research in this important field.
Alex Greene, senior editor, of Kluwer, helped and motivated us
during all phases of the preparation of the book.
Finally, we would like to thank our families for providing the
most meaningful content in our lives.