Theory and Design of Digital Communication

Systems Ha T.T. pdf download

https://ebookgate.com/product/theory-and-design-of-digital-
communication-systems-ha-t-t/

Get Instant Ebook Downloads – Browse at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Improvisational Design Continuous Responsive Digital

Communication Suguru Ishizaki

https://ebookgate.com/product/improvisational-design-continuous-
responsive-digital-communication-suguru-ishizaki/

ebookgate.com

Digital Systems Design Using VHDL 2nd Edition Roth

https://ebookgate.com/product/digital-systems-design-using-vhdl-2nd-
edition-roth/

ebookgate.com

Digital Media and Innovation Management and Design

Strategies in Communication Second Edition Richard A.
Gershon
https://ebookgate.com/product/digital-media-and-innovation-management-
and-design-strategies-in-communication-second-edition-richard-a-
gershon/
ebookgate.com

Digital Design of Signal Processing Systems A Practical

Approach 1st Edition Shoab Ahmed Khan

https://ebookgate.com/product/digital-design-of-signal-processing-
systems-a-practical-approach-1st-edition-shoab-ahmed-khan/

ebookgate.com
Evolution of Communication Systems D. Kimbrough Oller

https://ebookgate.com/product/evolution-of-communication-systems-d-
kimbrough-oller/

ebookgate.com

Metadata for Digital Resources Implementation Systems

Design and Interoperability 1st Edition Muriel Foulonneau
And Jenn Riley (Auth.)
https://ebookgate.com/product/metadata-for-digital-resources-
implementation-systems-design-and-interoperability-1st-edition-muriel-
foulonneau-and-jenn-riley-auth/
ebookgate.com

Handbook of Reflector Antennas and Feed Systems Vol 1

Theory and Design of Reflectors 1st Edition Satish K
Sharma
https://ebookgate.com/product/handbook-of-reflector-antennas-and-feed-
systems-vol-1-theory-and-design-of-reflectors-1st-edition-satish-k-
sharma/
ebookgate.com

Rethinking Marketing Developing a New Understanding of

Markets 1st Edition Ha?Kan Ha?Kansson

https://ebookgate.com/product/rethinking-marketing-developing-a-new-
understanding-of-markets-1st-edition-hakan-hakansson/

ebookgate.com

Digital Design for Print and Web An Introduction to Theory

Principles and Techniques 1st Edition John Dimarco

https://ebookgate.com/product/digital-design-for-print-and-web-an-
introduction-to-theory-principles-and-techniques-1st-edition-john-
dimarco/
ebookgate.com
Providing the underlying principles of digital communication and the design

HA
techniques of real-world systems, this textbook prepares senior undergraduate
and graduate students for the engineering practices required in industry. Covering
the core concepts, including link analysis, modulation, demodulation, spread
spectrum, equalization, channel fading effects, and channel coding, it provides

THEORY AND DESIGN

step-by-step mathematical derivations to aid understanding of background material.
In addition to describing the basic theory, the principles of system and subsystem

COMMUNICATION SYSTEMS
THEORY AND DESIGN OF DIGITAL
design are introduced, enabling students to visualize the intricate connections
between subsystems and understand how each aspect of the design supports the
overall goal of achieving reliable communications. Throughout the book, theories
are linked to practical applications with over 250 real-world examples, whilst 370
varied homework problems in three levels of difficulty enhance and extend the text
OF DIGITAL
COMMUNICATION
material. With this textbook, students can understand how digital communication
systems operate in the real world, learn how to design subsystems, and evaluate
end-to-end performance with ease and confidence.

Tri T. Ha is a Professor in the Department of Electrical and Computer Engineering

at the Naval Postgraduate School (NPS), Monterey, California, a position he has SYSTEMS
held since 1989. Prior to joining NPS he worked at Fairchild Industries and GTE,
and he was an Associate Professor at Virginia Tech for four years. He is an IEEE
Fellow who has written two previous textbooks, and his current research interests TRI T. HA
are in wireless communications and cyber warfare.
Cover designed by Zoe Naylor

www.cambridge.org/Ha

• Detailed solutions manual for instructors

HA: THEORY AND DESIGN PPC CMYBLK

 Theory and Design of Digital Communication Systems

Providing the underlying principles of digital communication and the design techniques of
real-world systems, this textbook prepares senior undergraduate and graduate students for
the engineering practices required in industry. Covering the core concepts, including link
analysis, modulation, demodulation, spread spectrum, equalization, channel fading effects,
and channel coding, it provides step-by-step mathematical derivations to aid understanding
of background material. In addition to describing the basic theory, the principles of system
and subsystem design are introduced, enabling students to visualize the intricate connec-
tions between subsystems and understand how each aspect of the design supports the overall
goal of achieving reliable communications. Throughout the book, theories are linked to
practical applications with over 250 real-world examples, whilst 370 varied homework
problems in three levels of difficulty enhance and extend the text material. With this
textbook, students can understand how digital communication systems operate in the real
world, learn how to design subsystems, and evaluate end-to-end performance with ease and
confidence.

Tri T. Ha is a Professor in the Department of Electrical and Computer Engineering at the Naval
Postgraduate School (NPS), Monterey, California, a position he has held since 1989. Prior to
joining NPS he worked at Fairchild Industries and GTE, and was an Associate Professor at
Virginia Tech for four years. He is an IEEE Fellow who has written two previous textbooks,
and his current research interests are in wireless communications and cyber warfare.
Theory and Design of Digital
Communication Systems

TRI T. HA
Naval Postgraduate School, Monterey, California
 cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo, Mexico City
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521761741

© Cambridge University Press 2011

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2011

Printed in the United Kingdom at the University Press, Cambridge

A catalog record for this publication is available from the British Library

Library of Congress Cataloging in Publication data

Ha, Tri T., 1949–
Theory and design of digital communication systems / Tri T. Ha.
p. cm.
ISBN 978-0-521-76174-1 (hardback)
1. Digital communications. I. Title.
TK5103.7.H35 2011
384–dc22
2010024374

ISBN 978-0-521-76174-1 Hardback

Additional resources for this publication at www.cambridge.org/9780521761741

Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to
in this publication, and does not guarantee that any content on such
websites is, or will remain, accurate or appropriate.
World peace must develop from inner peace. Peace is not the absence of violence. Peace
is the manifestation of human compassion.
14th Dalai Lama of Tibet (Inscription on United States Congressional Medal)
 Contents

Preface page xvii

Acknowledgements xix
List of symbols xx
List of abbreviations xxx

1 Introduction 1
1.1 Brief overview 1
1.2 Scope 3
1.3 Summary 8
Bibliography 9

2 Deterministic signal analysis 10

Introduction 10
2.1 General description of deterministic signals 10
Continuous-time signals 11
Discrete-time signals 11
Periodic and aperiodic signals 12
Analog and digital signals 13
2.2 Power and energy 16
2.3 Orthogonal signals 21
2.4 Signal space 27
Gram–Schmidt procedure 27
Signal vectors 28
2.5 Linear time-invariant systems 34
2.6 Convolution 35
2.7 Fourier series of continuous-time periodic signals 37
Parseval relation for periodic signals 39
Unit step and unit impulse functions 39
Power spectral density 41
Bandwidth 41
Frequency shifting property 42
Response of an LTI system to a periodic signal 45
2.8 Fourier transform of continuous-time signals 47
Frequency shifting property 50
Parseval relation 52
Fourier transform of periodic signals 53
vii
viii Contents

Response of LTI systems 53

Tables of Fourier properties and Fourier transform pairs 56
2.9 Autocorrelation 56
Autocorrelation and energy spectral density 57
Autocorrelation and power spectral density 59
Output autocorrelation and power spectral density 61
2.10 Sampling 62
Sampling theorem 62
Impulse-train sampling 63
Reconstruction with an ideal lowpass filter 64
2.11 Bandpass signals 68
Representations 68
Response of an LTI bandpass system 70
2.12 Summary 72
Problems 73
Further reading 77
Bibliography 77

3 Random signal analysis 78

Introduction 78
3.1 Review of probability theory 78
Total probability theorem 79
Bayes theorem 79
Independence 79
Union bound 79
3.2 Random variables 79
Bayes theorem and total probability revisited 81
3.3 Random processes 82
Autocorrelation and autocovariance 82
Types of random processes 85
Power spectral density 92
3.4 Gaussian process 100
Linear transformation 100
Sampling 102
Sufficient statistics for signal processing in white Gaussian noise 103
Karhunen–Loeve expansion 105
Whitening filter 106
The central limit theorem 108
3.5 Gaussian-derived processes 108
Rayleigh and Rice processes 108
Squared envelope 111
Sum of squared Gaussian processes 112
Nakagami-m density function 113
Log-normal density function 114
ix Contents

3.6 Summary 117

Problems 118
Further reading 121
Bibliography 121

4 Information theory and channel coding 122

Introduction 122
4.1 Entropy of a discrete source 122
4.2 Source coding 124
Huffman codes 126
Shannon noiseless source coding theorem (first theorem) 127
4.3 Discrete channel 128
Mutual information 129
Channel capacity 131
Shannon noisy channel coding theorem (main theorem) 131
4.4 Gaussian channel 134
Differential entropy 136
Mutual information and channel capacity 138
Bandlimited Gaussian channel capacity 144
Gaussian channel with discrete inputs 144
4.5 Channel coding with block codes 147
4.6 Low-density parity-check codes (LDPC) 152
Tanner graph 153
Message passing algorithm (MPA) 154
4.7 Channel coding with convolutional codes 160
Hard decoding–BSC 163
Soft decoding–DMC 165
Soft decoding–Gaussian input vector 167
4.8 Summary 169
Problems 170
Further reading 175
Bibliography 175

5 Communication link analysis 177

Introduction 177
5.1 Basic wireless communication link 177
Thermal noise 178
Effective noise temperature 181
Receiver noise model 187
System signal-to-noise ratio 190
5.2 Cellular communication link 191
Frequency spectrum 192
Major cellular standards 192
Cell connection 195
x Contents

Path loss: two-ray ground reflection model 196

Hata model 199
Modified Hata model 202
Reciprocity and the reverse link 203
5.3 Co-channel interference in a narrowband cellular system 203
Combined signal-to-interference and noise ratio 206
Sectoring 208
Microcell-zoning 210
5.4 CDMA cellular link analysis 211
Forward link 211
Reverse link 214
5.5 Satellite communication link 215
5.6 Summary 219
Problems 220
Further reading 224
Bibliography 224

6 Modulation 225
Introduction 225
6.1 Review of double sideband-amplitude modulation (DSB-AM) 225
6.2 Digital modulation 229
6.3 Phase shift keying (PSK) 231
6.4 Differential phase shift keying (DPSK) 234
6.5 Amplitude shift keying (ASK) 236
6.6 Frequency shift keying (FSK) 238
Power spectral density 240
6.7 Minimum shift keying (MSK) 242
Power spectral density 243
Modulator 247
6.8 Gaussian minimum shift keying (GMSK) 249
6.9 The need for M-ary modulation 251
6.10 M-ary amplitude shift keying (MASK) 252
Signal space 252
Power spectral density 253
Modulator 255
6.11 M-ary phase shift keying (MPSK) 255
Signal space 256
Power spectral density 258
Modulator 260
Offset quadrature phase shift keying (OQPSK) 260
6.12 Differential M-ary phase shift keying (DMPSK) 262
Alternative differential encoding for DQPSK 264
Direct symbol mapping 265
Modulator 265
xi Contents

6.13 π/4-shifted differential quadrature phase shift keying (π/4-DQPSK) 266

Signal space 267
Direct symbol mapping 268
6.14 M-ary quadrature amplitude modulation (MQAM) 270
Signal space 270
Power spectral density 271
Differential MQAM (DMQAM) 274
6.15 Code shift keying (CSK) 275
Power spectral density 276
Modulator 276
6.16 M-ary frequency shift keying (MFSK) 277
Power spectral density 278
Modulator 279
6.17 Continuous phase modulation (CPM) 280
Power spectral density 281
Modulator 282
6.18 Orthogonal frequency division multiplexing (OFDM) 282
Practical baseband implementation 283
Cyclic prefix 287
6.19 Trellis coded modulation (TCM) 290
Ungerboeck TCM 291
Pragmatic TCM 295
6.20 Summary 298
Problems 299
Further reading 302
Bibliography 302

7 Demodulation 305
Introduction 305
7.1 The matched filter 306
Time domain interpretation 306
Frequency domain interpretation 310
Output signal and noise waveforms 310
Decision variable 311
Summary 312
7.2 The correlator 312
7.3 The matched filter–envelope detector (noncoherent matched filter) 313
Output signal-to-noise ratio 317
Decision variable 318
Summary 318
7.4 The quadrature correlator–square law detector (noncoherent correlator) 318
7.5 The threshold detector 320
Optimum threshold 320
Maximum likelihood criterion 321
xii Contents

7.6 The maximum detector 322

Gaussian decision variables 323
Rice and Rayleigh decision variables 323
7.7 Binary demodulation 325
Coherent PSK 326
Coherent DPSK 328
Direct detection ASK 329
Coherent FSK 331
Coherent MSK and precoded MSK and GMSK 332
Noncoherent FSK and MSK 337
Noncoherent DPSK 339
Performance summary of binary modulation techniques 345
7.8 Minimum Euclidean distance detector 345
Symbol error probability 348
7.9 M-ary maximum detector 350
Gaussian decision variables 351
Rice and Rayleigh decision variables 353
7.10 M-ary demodulation 355
Coherent L-path demodulator: a signal space approach 355
Coherent M-path demodulator 358
Noncoherent M-path demodulator 360
7.11 Coherent MASK 362
7.12 Coherent MPSK 366
7.13 Coherent DMPSK 371
7.14 Noncoherent DMPSK 375
7.15 Noncoherent π/4-DQPSK 379
7.16 Coherent MQAM and DMQAM 380
7.17 Coherent CSK and MFSK 383
7.18 Noncoherent CSK and MFSK 384
7.19 Coherent CPM with sequence detection 386
7.20 Coherent CPM with symbol-by-symbol detection 395
7.21 Noncoherent CPM 397
7.22 Performance summary of M-ary modulation techniques 401
7.23 OFDM demodulation 402
Timing synchronization 403
Carrier phase synchronization 409
7.24 Binary demodulation with convolutional codes 410
Hard decoding 411
Soft decoding–Gaussian input vector 412
Soft decoding–χ2 input vector 414
7.25 TCM demodulation and decoding 416
7.26 Summary 422
Appendix 7A: The Q-function 422
xiii Contents

Problems 423
Further reading 430
Bibliography 430

8 Spread spectrum 432

Introduction 432
8.1 Direct sequence modulation 432
Orthogonal covering 436
IS-95 forward link 440
Code division multiple access (CDMA) 443
IS-95 reverse link 444
8.2 Direct sequence demodulation 446
Quadrature orthogonal covering demodulation 450
Noncoherent demodulation of DS-CSK 454
CDMA: performance evaluation 457
8.3 Frequency hop spread spectrum 460
Partial-band jamming 461
Multi-tone jamming 462
Follower jamming 464
8.4 Summary 466
Problems 466
Further reading 471
Bibliography 471

9 Intersymbol interference and equalization 473

Introduction 473
9.1 Intersymbol interference 473
Nyquist criterion for zero ISI 475
9.2 Optimum demodulator for bandlimited channel 477
Condition for maximum SNR0 480
Condition for zero ISI 480
Solution for |HT ( f )| and |HR ( f )| 480
9.3 Zero-forcing linear equalizer (ZF-LE) 482
Summary 493
9.4 Mean-square error linear equalizer (MSE-LE) 494
9.5 Zero-forcing decision-feedback equalizer (ZF-DFE) 501
9.6 Mean-square error decision-feedback equalizer (MSE-DFE) 506
9.7 Maximum likelihood sequence detection 509
9.8 Fractionally spaced equalizer (FSE) 516
9.9 Summary 517
Problems 518
Further reading 520
Bibliography 520
xiv Contents

10 Fading channels 522

Introduction 522
10.1 Physical characterization of fading channels 524
Time-varying effect 524
Space-varying effect 526
Summary of fading characteristics 527
10.2 Mathematical representation of fading channels 528
Channel impulse response 528
Multipath autocorrelation and Doppler profiles 530
Clarke–Doppler power spectrum 532
Generalized Doppler power spectrum 534
10.3 Coherent demodulation 536
Equivalent complex-valued demodulator 536
Rayleigh 539
Rice 540
Nakagami-m 541
Effect of channel tap error 543
10.4 Pilot symbol-aided decision-feedback demodulation 547
Differential and double-differential decision-feedback algorithms 549
10.5 OFDM 552
Low mobility 556
High mobility 557
10.6 Noncoherent demodulation 558
Rayleigh 558
Rice 558
Nakagami-m 559
Doppler tracking of MFSK 559
Doppler tracking of CSK 562
10.7 Pilot tone-aided demodulation of orthogonal covering signal 566
Complex spreading and despreading 566
Doppler analysis 567
10.8 Noncoherent demodulation of offset quadrature DS-CSK 572
10.9 Time diversity 574
Level crossing rate 575
Average fade duration 577
10.10 Maximal ratio combining (MRC) 578
Rayleigh 579
Rice 581
Nakagami-m 581
10.11 Selection combining (SC) 583
10.12 Equal gain combining (EGC) 585
Coherent EGC 585
Noncoherent EGC 586
xv Contents

10.13 Frequency diversity 587

Fast frequency hop (FFH) 588
OFDM subcarrier combining 588
Rake receiver 590
10.14 Receive antenna diversity (SIMO) 595
Mobile station antennas 596
Base station antennas 597
Performance 598
10.15 Transmit antenna diversity (MISO) 600
Space-time coding 600
Alamouti code 602
Rate ½ space-time code 603
Sufficient statistics 604
Rate ¾ space-time code 604
10.16 Transmit–receive antenna diversity (MIMO) 605
10.17 Channel capacity 608
Slow fading 609
Slow fading–receive antenna diversity 610
Slow fading–transmit antenna diversity 611
Slow fading–transmit and receive antenna diversity 612
Slow fading–OFDM 612
Fast fading 615
10.18 Summary 616
Appendix 10A: Complex-valued demodulators 617
Appendix 10B: Bit error probabilities 617
Rayleigh 618
Nakagami-m 618
Rayleigh–diversity: χ 2 -density function with 2L degrees of freedom 619
Problems 620
Further reading 627
Bibliography 627

Index 629
 Preface

This book was written with two goals in mind: to provide the underlying principles of digital
communication and to study design techniques integrated with real world systems. The
ultimate aim of a communication system is to provide reliable transmission of information to
the user(s). This fundamental foundation was established in 1948 by Claude Shannon, the
founding father of information theory, and led eventually to the development of modern
digital communication. Analog communication is near extinction or at the very gate of it.
The full spectrum dominance of digital communication has arrived and new frontiers are
being established every decade; from cellular systems to wireless LAN and MAN, the bit
rates are being pushed ever higher for ubiquitous mobile applications.
Knowing the limit of digital transmission is vital to the design of future communication
systems, particularly mobile wireless systems, where both spectrum and power are precious
resources, and design techniques can be used to manipulate these two main resources to fit
real world applications. No single technique can cover all the requirements of a modern
communication system, which makes it necessary for students to understand the intricate
web between subsystems, each designed to support others to achieve the common goal of
reliable communication.
The book contains more than 250 examples to help students achieve a firmer under-
standing of the subject. The problems at the end of each chapter follow closely the order of
the sections. They are designed for three levels: level one covers the straightforward
application of equations; level two requires patience and deep thinking; whilst level three
requires some research of the literature to assist in finding a solution. A solutions manual for
the instructor accompanies the book.
The book was written for both senior undergraduate and graduate students studying
communications at universities and colleges. The entire book is suitable for two-
semester courses in digital communications. The first course is typically a one-semester
senior course in digital communication, which may be taken by students new to
studying communications (the conventional wisdom is that students should learn analog
communication before learning digital communications) or after completing an intro-
ductory course in communication systems (one that is heavy in analog communication
systems such as AM and FM). The second course is a one-semester course for graduate
students who already have a firm background in random variables and processes. The
practical material included in this book (much of it focused on commercial and military
systems) will be helpful for practitioners and professionals in the digital communication
field.
As in the learning of any subject, some prerequisites are required for the reading of this
book. A first course in probability theory is necessary and exposures to random processes
xvii
xviii Preface

would be helpful. Readers should also be familiar with linear system analysis. A knowledge
of analog communication is helpful but not required. For readers who do not have the
patience to go through all the design techniques but would appreciate the beauty of the
underlying principles, we recommend our favorite book, Principles of Digital
Communication, authored by the legendary Robert G. Gallager.
 Acknowledgements

I would like to express my thanks to Dr. Phil Meyler of Cambridge University Press for his
enthusiastic support and for suggesting the title of the book to closely reflect its coverage.
I would like to thank the anonymous reviewers for their valuable comments that helped in
improving the book. I am grateful for the support of my friends Nathan Beltz, Professor
Vicente Garcia, Professor Jeff Knorr, Professor Frank Kragh, Donna Miller, Rita Painter,
Professor Clark Robertson, Professor Wei Su, Dr. Jan Tighe, and Dr. Charlie Victory. I also
would like to thank my students over the years for their valuable suggestions to improve the
original notes. I am grateful to my parents for providing me the opportunity to go to the
United States for higher education, and to Thuy, Khanh, Tuan, Huong and Dennis, Thu, and
Hien for their help and support over the years. This book is dedicated to the people who
work tirelessly for world peace.

xix
 Symbols

A amplitude, smallest signal amplitude in MQAM, azimuth angle

A matrix A
jjAjj2 squared Frobenius norm of matrix A
A conjugate and transpose of a matrix A
Ad number of paths of Hamming weight d that merge with the all-zero paths
Ae effective aperture area of the receiver antenna
A \ B; AB intersection of set A and set B, A and B
A [ B; A þ B union of set A and set B, A or B
a Gaussian filter parameter
a(hR) correction factor in Hata model
af frequency sensitivity
ak Fourier series coefficients
B bandwidth
Bd information weight (number of information bit errors) of all paths of
Hamming weight d
Bd free information weight (number of information bit errors) of all paths of
Euclidean distance d free of TCM
C channel capacity in coded bits/input symbol, correlation
C channel capacity in coded bits/second
Co outage capacity
cðDK Þ correlation metric in MLSD
CN correlation metric in CPM
CX ðt1 ; t2 Þ autocovariance of the random process x(t)
C(z) transfer function of causal ZF-LE, z-transform of the sequence c(k)
CovðNj Nk Þ covariance of two noise samples
c code word, PN sequence
ci a coded bit, a differentially coded bit
c(t) PN function of N chips
cðtÞ complex PN function
cI ðtÞ PN function of I-channel
cQ ðtÞ PN function of Q-channel
cðxjHi Þ metric for CPM demodulation
Di ith symbol in the symbol stream

xx
xxi List of symbols

d distance
d data sequence
dB decibel
d free free distance of a convolutional code
d free Euclidean free distance of TCM
di , d(i) normalized bit or symbol amplitude, di 2 f1; þ1Þ, di 2 f0; 1Þ
fdi g, fdðiÞg data sequence
dmin minimum Hamming distance, minimum Euclidean distance
d(u,v) Hamming distance between two code words
d ðx; si Þ Euclidean distance between two vectors
E energy (with or without a subscript), smallest symbol energy in MQAM or
MASK, electric-field wave, elevation angle
Eb bit energy
Eb diversity bit energy
E(d,t) free space E-field at distance d from transmitter and time t
Eh hop energy
EIRP effective isotropic radiated power
E0 free space E-field at distance d0 from transmitter
Es symbol energy
Es diversity symbol energy
EðX Þ expected value (mean value) of X
EðX 2 Þ mean-square value of X 2
e 2.718
e error word
e(k) error sequence of MSE-LE
e(t) error process
F noise figure
FfxðtÞg Fourier transform of x(t)
F ð zÞ minimum-phase transfer function, transfer function of a synthetic channel,
z-transform of the sequence f(k)
F  ð1=z Þ maximum-phase function
F 1 fX ðf Þg inverse Fourier transform of X(f )
2 F1 Gauss hypergeometric function
ð1Þ
Fjhj2 ðpo ÞSNR outage signal-to-noise ratio
FX ðxÞ distribution function of X
FXY ðx; yÞ joint distribution function of X and Y
fjhj ð xÞ density function of the channel tap magnitude jhj
fXY ðx; yÞ joint density function of X and Y
fX jY ðxjyÞ conditional density function of X given Y
f frequency, Doppler shift
xxii List of symbols

fc carrier frequency
fD Doppler spread
fj instantaneous carrier frequency of FH
fL;i instantaneous local carrier frequency of a frequency synthesizer
fm maximum Doppler shift
fs sampling frequency
fX ðxÞ density function of X
fX ðxÞ density function of vector X
G amplifier gain, gain of a two-port network
G generator matrix, space-time block code matrix
G TCM asymptotic coding gain
GDC gain of a downconverter
GLNA gain of a low-noise amplifier
GR receiver antenna gain
GR =Ts antenna gain-to-noise temperature ratio of the earth station
GS =Tsat antenna gain-to-noise temperature ratio of the satellite
GT transmitter antenna gain
g parity
g(t) pulse
gn ðtÞ orthonormal pulse shapes in OFDM
g(x) code generator polynomial
H Hadamard matrix (with or without a subscript), parity check matrix,
channel tap matrix
H source entropy in bits/second
Hi hypothesis
H(f ) transfer function or frequency response
HFE ðf Þ transfer function of the front-end filter of the receiver
jHðf Þj magnitude response (amplitude response)
HT ðf Þ transfer function of transmit filter
HR ðf Þ transfer function of receive filter
H(k) N-point DFT of the sequence h(n)
H(X) entropy of the discrete random variable (discrete source) X
h Planck constant, digital modulation index, complex channel tap
h row vector of a Hadamard matrix
^
h MMSE of the vector h
hi ; hðiÞ ith channel tap
~hðiÞ complex channel tap
hL ðtÞ complex envelope of the impulse response h(t)
h(n) sequence used in OFDM
hR receive antenna height
xxiii List of symbols

hT transmit antenna height

h(t) impulse response
h(X) differential entropy of a continuous random variable X
h(X|Y) conditional differential entropy of a continuous random variable X
h(X) differential entropy of a continuous n-dimensional random vector X
h(x) monic binary irreducible primitive polynomial, PN code polynomial
I interference power, photodiode current
Im interchannel interference
I MUI variable
I(X) self-information of the discrete random variable (discrete source) X
Iðui ; vj Þ pair-wise mutual information
IðU ; V Þ mutual information
I0 ðÞ modified Bessel function of the first kind of zero order
In ðÞ modified Bessel function of the first kind of nth order
J jamming variable
J0 , J0 ðÞ jamming spectral density, Bessel function of zero order
K number of simultaneous users in CDMA, Kelvin
K covariance matrix
k integer, Boltzmann constant, number of information bits in a block code,
number of inputs of a convolutional encoder
k0 free space wave number
k/n code rate
LC path loss
L diversity order, loss of a two-port network
L(f ) transfer function of an equalizer
Lr receive antenna diversity
Ls number of symbol times for m transmitted symbols in transmit diversity
LTL loss of a transmission line
Lt transmit antenna diversity
LdB mean value in decibels of the log-normal density variable
Lðλ; P1 ; :::; Pn Þ Lagrangian
l length of a code word
l average length of a code word
ln natural logarithm
ln ΛðX jyÞ conditional ln-likelihood ratio
log base-10 logarithm
log2 base-2 logarithm
M number of distinct M-ary symbols
M(r | c) path metric in Viterbi algorithm or log-likelihood function
m mean value of a random variable, Nakagami-m parameter
xxiv List of symbols

mA ðX 1 Þ arithmetic mean of 1=X ðf Þ

mG ðX Þ geometric mean of X(f )
mn metric for OFDM timing synchronization
m(t) message signal
ðlÞ
mij message sent by the bit node i to check node j at the lth iteration
^ ðlÞ
m ji message sent by the check node j to bit node i at the lth iteration
m2l sum of the squares of the means
mX ðtÞ mean value of the random process x(t)
N noise variable, available noise power, number of OFDM subcarriers,
period of a PN sequence
N noise vector, complex noise
N noise variable at detector input
N noise variable
N complex noise variable, complex noise vector
Nð f Þ power spectral density of the equivalent lowpass noise nðt Þ
ND number of branches in a frequency bin determinator
NH number of hop bins
Ni system noise power
Nn average number of nearest neighbors of a signal vector
N ð0; σ 2 Þ Gaussian random variable with zero mean and variance σ 2
Nð0; σ 2 Þ Gaussian vector with iid components of zero mean and variance σ 2
Nk noise sample
N0 =2 power spectral density of noise
NV average number of level crossings
n code word length, path loss exponent
nI ; N I in-phase noise variables
nQ ; NQ quadrature noise variables
(n,k) block code of k information bits and code word length n
n(t) noise
nL ðtÞ complex envelope of bandpass noise
n0 ðtÞ output noise of a matched filter
P power
Pc;h probability of correctly identifying the frequency bin
PðdÞ pair-wise error probability (Prðc ! c 0 Þ)
Pe error probability (bit, symbol, code word)
P(f ) energy spectrum, Fourier transform of pulse shape p(t)
Pj power of a jamming signal
Pp peak power
Pr(A) probability of A
Pr(A,B) joint probability of A and B
xxv List of symbols

Pr(A|B) conditional probability of A given B

Prðc ! c 0 Þ pair-wise error probability
Pr(c | r) a posteriori probability
Pr(r | c) likelihood of the transmitted code vector c
PT transmit power
P(z) linear predictor in MSE-DFE
p crossover probability of a BSC, probability of a binary symbol
pout outage probability
pUV ðui ; vj Þ joint distribution of ui ; vj
pðvj jui Þ transition probability of a discrete channel
pX ðxi Þ distribution of the discrete random variable X, i = 1, 2, . . . , n
p(t) pulse shape
Qða; bÞ Marcum Q-function
QðxÞ Gaussian integral Q-function of argument x
QðzÞ transfer function of the composite channel in suboptimum MSE-LE
q optimum number of jamming tones
R resistance, Rayleigh random variable, source rate in symbols/second
Rb bit rate
Rc chip rate
R responsivity
Re Earth’s radius (6378 km)
Rh ðτ; t 0 Þ multipath autocorrelation profile
Rhi ðt 0 Þ ith path autocorrelation
Rði  jÞ autocorrelation of the data sequence fdi g
Rs symbol rate
R(t) envelope of a bandpass process
RV level crossing rate
Rw Walsh chip rate
Rx ðτÞ autocorrelation of WSS random process x(t)
RX ðt1 ; t2 Þ autocorrelation of the random process x(t)
r value assumed by a random variable R, code rate, spectral efficiency
r received word
r2 signal-to-intertone interference ratio
re extinction ratio
S input variable of a Gaussian channel
S input vector of a Gaussian channel
S sample space
S(f ) power spectral density (with or without a subscript)
Sh ðτ; f 0 Þ multipath Doppler profile
Shi ðf 0 Þ ith path Doppler power spectrum
xxvi List of symbols

s syndrome vector, orthogonal covering symbol

sgn (x) signum function
sign½ x sign of x
s(t) digital signal
fsi ðtÞg set of M digital signals, i = 1, 2, . . . , M
fsi g set of M signal vectors in the signal space, i = 1, 2, . . . , M
ksi k norm of signal vector si
sik coefficients of the Gram–Schmidt orthogonal expansion, I–Q values of a
two-dimensional signal vector
sL ðnÞ time samples of an OFDM signal
sL ðtÞ complex envelope a bandpass signal s(t)
s0 ðtÞ output signal of a matched filter
sp pilot symbol
T time interval, period, sampling period
T phase error rotation matrix
TA antenna noise temperature
Tb bit time
Tc chip time, channel coherence time
Td time delay, multipath delay spread
TDC effective noise temperature of a downconverter
TDM effective noise temperature of a demodulator
Te effective noise temperature
Th hop time
TLNA effective noise temperature of a low-noise amplifier
Tn physical temperature of the resistor
T0 reference temperature, time interval, period
Tp pulse width
Tr(A) trace of matrix A
Ts symbol time, system noise temperature
t time, error-correcting capability of a block code
U set of M input symbols of a discrete channel
SM
Ai union of Ai set
i¼1
u(t) unit step function
u message vector
V voltage, set of Q output symbols of a discrete channel
VarðX Þ variance of X
Vrms root mean-square voltage
VT threshold voltage
v radial velocity
v(t) voltage signal
xxvii List of symbols

W bandwidth, watt
fwn ðtÞg set of Walsh functions, n = 1, 2, . . . , M
X random variable, discrete source, output variable of a Gaussian channel or
a matched filter
X random vector, output vector of a Gaussian channel
X mean value (expected value) of X
X2 mean-square value of X 2
 0
X ej2πf discrete-time Fourier transform of the sequence x(k)
 
X ðf Þ2  energy spectral density of the energy signal x(t)
X(f ) Fourier transform of x(t), folded spectrum
X(k) N-point DFT of the sequence x(n)
Xn nth extension of the discrete source X
XT ðf Þ Fourier transform of xT ðtÞ, 2T-truncation of x(t), T  t  T
X(z) z-transform of the sequence x(k), transfer function of the composite
channel in optimum MSE-LE
x value assumed by a random variable X
x value assumed by a random vector X
b xc integer part of x
xðnÞ discrete-time signal, sequence used in OFDM
x(t) continuous-time signal (with or without a subscript)
xI ðtÞ in-phase component of the bandpass signal x(t)
fxk ðtÞg set of L orthonormal basis functions, k = 1, 2, . . . , L
fxk g set of L orthonormal basis vectors, k = 1, 2, . . . , L
xL ðtÞ complex envelope (equivalent lowpass signal) of the bandpass signal x(t)
xp ðtÞ periodic signal
xQ ðtÞ quadrature component of the bandpass signal x(t)
xs ðtÞ sampled function
xT ðtÞ 2T-truncation of x(t), T  t  T
Y(k) N-point DFT of the sequence y(n)
y(n) sequence
y(t) continuous-time function
Z pre-mapped vector at the input of the combiner
Z(k) frequency samples of an OFDM signal (the I–Q values of symbols of
OFDM subcarriers)
Z0 amplifier transimpedance
zk ðtÞ complex envelope of the kth OFDM subcarrier
 linear convolution
 circular convolution
ð Þ complex conjugate
α arbitrary constant
xxviii List of symbols

αn normalized signal amplitude in MQAM

αp complex Doppler factor
 
αp  Doppler loss factor
β proportionality constant, roll-off factor of a raised-cosine filter
γ threshold
γk;m complex Doppler loss factor
γn MSK data stream
Γ ground reflection coefficient, gamma function
ΔF frequency offset in OFDM
Δf peak frequency deviation
Δ"l differential Doppler phase error
Δ^"l1 post-estimated differential Doppler phase error
Δ"l  Δ^
"l1 double-differential Doppler phase error
δ jamming pulse duty cycle, fraction of FH bandwidth being jammed,
fraction of a hop being jammed
δij 0 for i 6¼ j and 1 for i = j
δðtÞ unit impulse function
" phase error
"l Doppler phase error
θ phase
θk azimuth angle of the kth wave
θL Earth station longitude
θ‘ Earth station latitude
θS GEO satellite longitude
θðtÞ phase function
λ wavelength, Lagrange multiplier
μ conditional mean value
ΛðXjyÞ conditional likelihood ratio
 product, fractional coverage area
ρ spatial correlation coefficient
ρX ðtÞ normalized autocovariance of the random process x(t)
σ2 variance of noise
σ dB standard deviation of the log-normal density variable in decibels
σ 2X variance of the random variable X
σ 2s power of the diffuse paths
τ time delay variable, average fade duration
τ i ðtÞ path delay
’ phase state in CPM
’k polar angle of the kth wave
f impossible event, null set, phase in MFSK and CPM, phase shift
xxix List of symbols

F0 ð f Þ power spectral density of the equalizer output noise

ff k ðtÞg set of orthonormal eigenfunctions of the noise autocorrelation
χ voice activity factor or data duty cycle
χ2 chi-square
ψ angle of mobile direction with respect to the x-axis, Doppler phase error
Cð f Þ power spectral density of sampled noise
C0 ð f Þ power spectral density of output noise of ZF-LE
CðtÞ phase of a bandpass process
O mean-square value of the envelope of the Nakagami-m process
 Abbreviations

2G second generation
3G third generation
A/D analog/digital conversion
AGN additive Gaussian noise
AMPS advanced mobile phone system
APD avalanche photodiode
ASK amplitude shift keying
AWGN additive white Gaussian noise
BCH Bose–Chaudhuri–Hocquenghem code
BEC binary erasure channel
BPA belief propagation algorithm
BSC binary symmetric channel
CDM code division multiplexing
CDMA code division multiple access
CDMA 2000 3G CDMA
CP cyclic prefix
CPM continuous phase modulation
CP-MFSK continuous phase M-ary frequency shift keying
CRC cyclic redundancy check
CSI channel side information
CSIR channel side information at the receiver
CSK code shift keying
D/A digital/analog conversion
DD-DF double-differential decision-feedback algorithm
D-DF differential decision-feedback algorithm
DEMUX demultiplexer
DFS decision-feedback selection
DFT discrete Fourier transform
DMC discrete memoryless channel
DMPSK differential M-ary phase shift keying
DMQAM differential quadrature amplitude modulation
DPSK differential phase shift keying
DQPSK differential quadrature phase shift keying

xxx
xxxi List of abbreviations

DSB-AM double sideband–amplitude modulation

DS direct sequence
DS-CSK direct sequence–code shift keying
DS-PSK direct sequence–phase shift keying
DS-SS direct sequence spread spectrum
DTFT discrete-time Fourier transform
EGC equal gain combining
EIRP effective isotropic radiated power
ESN electronic serial number
ETACS extended total access cellular system
FCC Federal Communications Commission
FDM frequency division multiplexing
FDMA frequency division multiple access
FFH fast frequency hop
FFT fast Fourier transform
FH frequency hop
FIR finite impulse response
FM frequency modulation
FSE fractionally spaced equalizer
FSK frequency shift keying (binary frequency shift keying)
fT frequency–time product
GEO geostationary orbit
GMSK Gaussian minimum shift keying
GPS global positioning system
GSM global system for mobile communication
2
jhj SNR instantaneous SNR
ICI intercarrier interference
ICI interchannel interference
IDFT inverse discrete Fourier transform
IEEE Institute of Electrical and Electronics Engineers
IFFT inverse fast Fourier transform
iid independent and identically distributed
IIR infinite impulse response
IPI intrapath interference
IS interim standard
ISI intersymbol interference
ISI intersample interference
JDC Japanese digital cellular system
JTACS Japanese total access communication system
LDPC low-density parity-check code
xxxii List of abbreviations

LFSR linear feedback shift-register

LLR ln-likelihood ratio
LR likelihood ratio
L-REC rectangular pulse of duration L symbols
L-RC raised cosine pulse shape of duration L symbols
LSB lower sideband
LTI linear time-invariant
MAP maximum a posteriori
MASK M-ary amplitude shift keying
MFSK M-ary frequency shift keying
MIMO multiple-input multiple-output
MIN mobile identification number
MIP multipath intensity profile
MISO multiple-input single-output
ML maximum likelihood
MLSD maximum likelihood sequence detection
MMSE minimum mean-square error
MPA message passing algorithm
MPSK M-ary phase shift keying
MQAM quadrature amplitude modulation
MRC maximal ratio combining
MSC mobile switching center
MSE-DFE mean-square error decision-feedback equalizer
MSE-LE mean-square error linear equalizer
MSK minimum shift keying
MUI multi-user interference
MUX multiplexer
NAMPS narrowband advanced mobile phone system
NRZ non-return-to-zero
NTACS narrowband total access communication systems
OFDM orthogonal frequency division multiplexing
OOK on–off keying
OQPSK offset quadrature phase shift keying
PCS personal communication system
PD pin photodiode
PDC Pacific (or personal) digital cellular system
PDF probability distribution function
pdf probability density function
π=4-DQPSK π=4 shift differential quadrature phase shift keying
PLL phase-locked loop
xxxiii List of abbreviations

PN pseudo-noise
PSK phase shift keying (binary phase shift keying)
PSTN public switched telephone network
QPSK quadrature phase shift keying
RS Reed–Solomon code
SC selection combining
SCM station class mark
SFH slow frequency hop
SIMO single-input multiple-output
SINR signal-to-interference and noise ratio
SINR path signal-to-interference-and-noise ratio
SINR0 output signal-to-interference plus noise ratio
SIR signal-to-interference ratio
SIRi input signal-to-interference ratio
SIR0 output signal-to-interference ratio
SJNR0 output signal-to-jamming-plus-noise ratio
SJRi input signal-to-jamming ratio
SJRi; p input signal-to-pulse jamming ratio
SJR0 output signal-to-jamming ratio
SJR0; p output signal-to-pulse jamming ratio
SNR signal-to-noise ratio
SNR diversity symbol signal-to-noise ratio
SNR0 output signal-to-noise ratio
SPA sum product algorithm
TCM trellis coded modulation
TDMA time division multiple access
TIA/EIA Telecommunication Industry Association/Electronic Industry Association
USB upper sideband
USDC US digital cellular
VCO voltage-controlled oscillator
WCDMA wideband CDMA (3G CDMA)
WLAN wireless local area network
WMAN wireless metropolitan area network
WSCS wide-sense cyclostationary
WSS wide-sense stationary
ZF-DFE zero-forcing decision-feedback equalizer
ZF-LE zero-forcing linear equalizer
1 Introduction

If the mind is in tranquility, time and space cease to exist.

Essence of Buddhism

1.1 Brief overview

This book provides the principles of digital communication and studies techniques to design
and analyze digital communication systems for point-to-point and point-to-multipoint
transmission and reception. Other than for radio broadcasting, modern communication
systems are going digital, and in the USA the conversion of analog TV broadcasting into
digital HDTV broadcasting at the beginning of 2009 signified the coming end of analog
communications. Communications between living beings began with the voice, and the
three biggest voice systems in the world are the telephone, and the cellular and radio
broadcasting systems.
The dissemination of visual activities then propelled the development of TV broadcasting
systems. The pioneer telephone network and radio broadcasting systems employed analog
communication techniques, such as AM and FM, for transmission of analog voice, as did the
analog TV broadcasting systems, which employed VSB-AM for picture transmission. The
quality of the message, such as voice and images, at the analog receiver depends on
how well the waveform that carries the message over the physical channel (twisted-pair
telephone wires, coaxial and fiber-optic cables, space, and water) can be reproduced. In
addition, the fidelity of the received message depends on the signal-to-noise ratio at the
receiver input. For good analog communications, the signal-to-noise ratio must be large, and
this requires high-power transmitters, such as are used in AM radio and TV broadcasting.
For FM radio broadcasting a large frequency spectrum is used, such as 200 kHz for radio
broadcasting, which shows that analog communications do not utilize power and bandwidth
efficiently. Furthermore, the advent of the Internet requires audio, video, imagery, and text
messages to be integrated for transmission over a common channel and this in effect rules
out analog communications such as AM and FM.
In analog communications, the message signal requires an infinite set of continuous-time
waveforms for transmission over a physical channel. This is because the message itself, such
as audio or video, must first be converted into a voltage baseband waveform with a
continuous range in amplitude that has countless possible values. When the baseband
voltage waveform is used to modulate an RF carrier for transmission, such as in AM or
FM, the modulated RF signal transmitted over the physical channel also has countless
possible values in both its amplitude and frequency ranges. The only way to recover the
1
2 Introduction

message signal is to faithfully reproduce the baseband waveform from the modulated signal.
This can be done easily in the case of no noise and no equipment imperfections, but
otherwise the fidelity of the message signal may be reduced. Digital communication does
not involve the faithful reproduction of the baseband waveform in the presence of noise
and equipment imperfections. Digital communication operates instead with a finite set of
continuous-time modulation waveforms for transmission over a physical channel. This
implies that the message signal must be represented by a finite set of voltage baseband
waveforms. Mathematically, a finite set of waveforms can only represent a finite set
of alphabets, commonly referred to as symbols. A symbol consists of a fixed number of
binary digits or bits. For example, the set of four distinct symbols {00, 01, 10, 11} can be
represented by four distinct waveforms fA cos 2πfc t, A sin 2πfc tg. The time separation of
consecutive waveforms that represent a symbol stream is called the symbol time, which is
the inverse of the symbol rate. If the waveforms are of finite duration then this duration is the
symbol time. This begs the question of how to obtain the bits or symbols that represent the
message. The process of converting a voltage baseband waveform that represents an audio
or video message into bits is referred to as the analog-to-digital conversion (or A/D). Text
messages generated by computers are inherently in bits, so with A/D conversion, audio,
video, text, and imagery can all be integrated into a single digital stream of bits. The process
of A/D, bit-symbol mapping, baseband waveform shaping, and modulation is referred to
as digital transmission. The process of demodulating the modulated signal, detecting the
symbol, symbol–bit mapping, and digital-to-analog conversion (or D/A) is called digital
reception.
Digital communication makes no attempts to reproduce the finite set of voltage baseband
waveforms. Instead, the receiver detects the energy content of each baseband waveform in
the presence of noise and equipment imperfections, and then makes a best estimate of which
transmitted symbol was received. If the signal-to noise ratio per symbol is reasonably large,
a symbol will most likely be detected correctly with high probability. If not, a symbol error
may occur. This is the essence of digital communication. For a given signal-to-noise ratio, an
analog communication receiver attempts to reproduce the voltage baseband waveform with
certain subjective fidelity. On the other hand, for a given signal-to-noise ratio per symbol, a
digital communication receiver produces symbols with a quantitative error rate. It is
important to know in advance the lower bound of the signal-to-noise ratio per symbol for
a specified error rate irrespective of the type and size of the set of modulation waveforms. In
1948 Claude Shannon established this lower bound and also provided the channel capacity
for reliable transmission [1]. Shannon’s work gives the designers of digital communication
systems the freedom to choose the set of modulation waveforms that achieve either the
best power or bandwidth efficiency, or a trade-off combination of both. As long as the
transmission rate is below the channel capacity and the signal-to-noise ratio per symbol is
above the Shannon limit, reliable communication is possible with an arbitrarily small error
rate. Guided by the Shannon channel capacity theorem (main theorem), the designer can
further integrate error-correction codes with modulation techniques to lower the signal-to-
noise ratio per symbol to achieve a specified error rate. The first error-correction code, the
Hamming code, was discovered by Richard W. Hamming in 1950, two years after Shannon
published his landmark work [2]. In addition to the main theorem, the Shannon first theorem
 3 1.2 Scope

Analog
message A/D Baseband
bit-symbol
compression waveform Modulation
mapping
encoding shaping

Channel

Analog
message Decoding Demodulation
Symbol-bit Equalization signal
decompression
mapping detection processing
D/A

Figure 1.1 Conceptual block diagram of a digital communication system.

provided the framework for encoding a discrete source of a finite set of symbols to minimize
the bit rate at the source encoder output. This allows the compression of the A/D samples of
the message signal to remove redundancy and any insignificant information not perceptible
by the human eye or ear.
The most common compression algorithms in use today are MP3 for music, JPEG for
pictures, and MPEG for video. Figure 1.1 shows the conceptual block diagram of a digital
communication system. The material in the book is organized to cover the transmitter,
receiver, and channel.

1.2 Scope

Chapter 2 provides a general study of deterministic signals that can be analyzed with Fourier
transform and Fourier series. Simple classification of signals and the concept of power and
energy are reviewed. One important class of signal, namely orthogonal signals, such as the
Walsh functions employed in IS-95, CDMA 2000, and WCDMA, is discussed in detail.
The majority of continuous-time and finite-energy signals in practice can be conveniently
analyzed via their signal spaces. These signal spaces are displays of the signal vectors in
their respective constellations. The signal vectors which can be viewed as the A/D versions
of a signal set contain all the information about the signal set. Practical communication
systems are inherently linear time-invariant systems operating in the small-signal range.
They can be analyzed by Fourier series and Fourier transform to provide a frequency-
domain snapshot of the signal bandwidth. The concept of autocorrelation and its relationship
with energy or power spectral density are discussed for linear time-invariant systems.
The sampling theorem that governs the A/D conversion of an analog signal and the
4 Introduction

Nyquist–Shannon interpolation for reconstruction of the analog signal are presented.

Finally, the representations of a bandpass signal, that is, the signal sent over a physical
channel, are discussed. The complex envelope (equivalent lowpass signal) of a bandpass
signal that can be employed to simplify the analysis of a communication system is also
included.
Chapter 3 studies random signals and their statistics. Although a finite set of deterministic
signals is employed to represent a finite set of information symbols, the transmitted symbols
are truly random, with each one in the set occuring with a fixed probability. Therefore, the
infinite series of signals transmitted over the channel is indeed a random process with finite
power. The study of random processes allows the establishment of the Fourier transform
relationship between the autocorrelation of the random process and its power spectral
density via the Einstein–Wiener–Khinchine theorem. The emphasis here is on cyclo-
stationary processes, which encompass all digitally modulated signals. The Gaussian
process that is used to represent channel noise is discussed in sufficient detail. Sampling
of bandlimited white Gaussian process, sufficient statistics for white Gaussian samples, the
Karhunen–Loeve theorem, and whitening filter are studied. To study the performance of
wireless communication via a fading channel we look at a variety of processes derived
from the Gaussian process, such as the Rayleigh, Rice, Nakagami-m, χ 2 , and log-normal
processes.
Chapter 4 provides a general study of information theory developed by Shannon, and
addresses both source and channel coding. The concept of source entropy and prefix code is
discussed. Shannon–Fano and Huffman prefix codes are used as examples. The Shannon
first theorem is presented with a proof, and the concept of mutual information is presented
together with the Shannon main theorem for a discrete channel. The concept of differential
entropy for a Gaussian channel is introduced, and leads to mutual information and Shannon
channel capacity. Vector Gaussian channels and the water filling strategy are presented to
highlight the concept of channel coding, which is also interpreted via the sphere packing
bound. The channel capacity of a bandlimited Gaussian channel and the channel capacity of
a Gaussian channel with discrete inputs are derived. The latter provides a snapshot of how
efficient digitally modulated signals perform as compared to the Shannon capacity. Channel
coding can be done with error-correction codes such as block codes and convolutional
codes. Performance of coded digital signals is presented for both block codes and convo-
lutional codes, with the emphasis on low-density parity-check codes (LDPC) and con-
volutional codes. The decoding of LDPC codes is implemented via the message passing
algorithm. The decoding of convolutional codes is carried out via the Viterbi algorithm,
which includes hard decoding, and quantized or unquantized soft decoding.
Chapter 5 examines methods for establishing a communication link between the trans-
mitter and receiver, commonly referred to as link analysis. The link budget involves the
allocation of power to the transmitter and noise temperature (or noise figure) to the receiver
so that a signal-to-noise ratio is established at the receiver to match a specified error rate
range. Given the transmitter power and the channel attenuation, the power of the received
signal can then be established. The channel attenuation is unique for each physical medium.
The chapter begins with the concept of the noise temperature of a two-port network,
which leads to the concept of the system noise temperature of a cascade of two-port
5 1.2 Scope

networks modeling a receiver. The system noise temperature allows the evaluation of noise
power in the receiver bandwidth, and hence the system signal-to-noise ratio. The physical
channels investigated in this chapter are the cellular and satellite channels. For the cellular
channel, we adopt the well-known Hata model to estimate the median path loss between
the transmitter and receiver. The presence of co-channel interference between cells is also
taken into account. Both narrowband cellular systems (IS-136, GSM) and wideband CDMA
cellular systems (IS-95, CDMA-2000, WCDMA) are covered. For a satellite channel,
the communication link is a point-to-point link, consisting of up- and downlinks. The
Friis formula for free-space attenuation is employed to establish the uplink or downlink
attenuation.
Chapter 6 presents modulation techniques for transmitting information over the physical
channel. The chapter essentially has two parts, namely binary modulation and M-ary
modulation. The structure of each modulation technique is studied via the signal waveform,
the power spectral density, and the modulator. For binary modulation we investigate phase
shift keying (PSK), differential phase shift keying (DPSK), amplitude shift keying (ASK)
(commonly referred to as intensity-modulated on–off keying (OOK), a technique used in
fiber optic communication), frequency shift keying (FSK), minimum shift keying (MSK),
and Gaussian MSK employed by the GSM cellular standard. Many practical applications
require either the higher spectral efficiency or higher power efficiency that binary modu-
lation techniques can provide; M-ary modulation can accommodate both. The second part of
this chapter covers M-ary amplitude shift keying (MASK), M-ary phase shift keying
(MPSK), offset quadrature phase shift keying (OQPSK), differential M-ary phase shift
keying (DMPSK), π=4 shifted differential quadrature phase shift keying (π=4-DQPSK),
M-ary quadrature amplitude modulation (MQAM), code shift keying (CSK), M-ary fre-
quency shift keying (MFSK), and continuous phase modulation (CPM). The chapter con-
tinues with a treatment of the dominant multiplexing-modulation technique, namely
orthogonal frequency division multiplexing (OFDM), which is used in many wireless
standards. The chapter ends with a look at trellis coded modulation (TCM) for bandlimited
channels. Both Ungerboeck and pragmatic TCM are investigated.
Chapter 7 provides a treatment of digital demodulation. A generic digital demodulator
consists of two major subsystems, namely the signal processor and the detector. There are
four types of signal processor: the matched filter, the correlator, the noncoherent matched
filter, and the noncoherent correlator. The first two types are employed in coherent de-
modulation while the last two types are used in noncoherent demodulation. For binary
demodulation the two fundamental detectors are threshold and maximum detectors. For
M-ary demodulation the two fundamental detectors are the minimum Euclidean distance
detector and the M-ary maximum detector. Combining the signal processor(s) and the
detector in that order produces an L-path demodulator for the set of digital signals with
L orthonormal basis functions and an M-path demodulator for the set of M orthogonal
signals. The bit error probability analysis is carried out for binary modulation techniques
such as coherent PSK, coherent DPSK, direct-detection ASK (for fiber optic communica-
tion), coherent FSK, coherent MSK, precoded MSK and GMSK, noncoherent FSK and
MSK, and noncoherent DPSK. For M-ary demodulation, the bit error probability analysis is
carried out for coherent MASK, coherent MPSK, coherent DMPSK, noncoherent DMPSK,
6 Introduction

coherent MQAM and DMQAM, coherent CSK and MFSK, noncoherent CSK and MFSK,
coherent CPM with sequence detection, coherent CPM with symbol-by-symbol detection,
and noncoherent CPM. The chapter continues with OFDM demodulation, with emphasis on
the IEEE 802.11a,g standards. Finally, the demodulation and decoding of TCM are studied
and performance analysis is investigated. The Viterbi algorithm is again used to illustrate the
decoding process.
Chapter 8 investigates two major spread spectrum communication techniques for both
commercial and military applications: direct sequence (DS) and frequency hop (FH). The
chapter begins with a presentation of the pseudo-noise (PN) sequences needed for spreading
the modulated signal. Next the concept of quadrature orthogonal covering using Walsh
functions of the same length for multiplexing DS signals with an identical symbol rate is
discussed. This concept is then extended to variable-length orthogonal covering for variable
symbol rates. IS-95 is used as a real life example for the study of the direct sequence
spread spectrum. The demodulation of DS signals in the presence of tone jamming, broad-
band jamming, and pulse jamming is analyzed. Demodulation of quadrature orthogonal
covering (IS-95 forward link) as well as noncoherent DS-CSK (IS-95 reverse link) is
presented. The analysis of code division multiple access (CDMA) with random spreading
sequences is presented together with a closed form expression and a tight upper bound for bit
error probability. For frequency hop signals, three jamming strategies are studied: partial-
band jamming, multi-tone jamming, and follower jamming. Both slow and fast hops are
considered for follower jamming.
Chapter 9 deals with intersymbol interference (ISI) in a bandlimited channel. The Nyquist
criterion for zero ISI is stated together with the corresponding pulse shapes that satisfy it.
The design of an optimum demodulator for a bandlimited channel with Gaussian noise is
carried out. The optimum demodulator relies on the signal pulse shape implemented at the
modulator. The channel is converted to an ideal channel via an equalizer implemented at the
modulator (the equalizer is a filter with a transfer function equal to the inverse transfer
function of the channel). At the demodulator, a matched filter matched to the signal pulse
shape simultaneously achieves both the maximum signal-to-noise ratio and zero ISI as long
as the pulse shape at the matched filter output satisfies the Nyquist criterion of zero ISI.
In practice, because the channel transfer function is not known or varies with time, ISI
removal is instead implemented at the demodulator. The equalizer implemented at the
demodulator can be classified into two types: linear and nonlinear. The treatment of linear
equalizers covers zero-forcing and mean-square error equalizers. The latter alleviates the
noise enhancement effect that severely degrades the former in channels with deep attenu-
ation in the passband. Nonlinear equalizers such as zero-forcing decision-feedback and
mean-square error decision-feedback can avoid the noise enhancement effect altogether,
although in channels with severe distortion the error propagation due to decision feedback
could worsen the performance. To obtain optimum performance, maximum likelihood
sequence detection may be employed to mitigate the ISI. The motivation behind sequence
detection is to use the symbol energy that resides in the ISI portion of the symbol to aid the
detection instead of throwing it away. The Viterbi algorithm is employed in practice for
sequence detection. Finally, a fractionally spaced equalizer that can mitigate timing error
is presented.
7 1.2 Scope

Chapter 10 studies the transmission and reception of a digitally modulated signal over a
fading channel. Unlike the AWGN channel, where noise is the only problem, a fading
channel is a greater challenge, as it may cause signal envelope variations, phase errors, and
intersymbol interference, all of which are detrimental to the performance of the signal. Thus,
it is necessary to understand the mechanism that causes these unwanted effects and find
ways to mitigate them. A fading channel arises from the movement of the transmitter and
receiver, commonly referred to as the time-varying effect or Doppler effect. This may cause
random amplitude attenuation and random phase rotation of the signal space. In underwater
acoustic communication, the medium also changes over time, further compounding the
problem. A fading channel also exhibits a space-varying effect, where the locations of the
transmitter and receiver and the physical structures in the environment dictate the paths that
the signal may travel. A transmitted signal representing an arbitrary symbol may arrive at the
receiver via multiple paths. Depending on the time delay between paths, signal echoes of a
symbol may overlap the next several symbols causing intersymbol interference. Both time-
varying and space-varying effects can be classified into four fading characteristics: slow
fading with random amplitude attenuation and negligible phase error, fast fading with
random amplitude attenuation and large phase error, flat fading with random amplitude
attenuation and negligible intersymbol interference, and frequency-selective fading with
random amplitude attenuation and intersymbol interference. A mobile wireless channel may
have two of these four characteristics where the random amplitude attenuation is described
by a special distribution (Rayleigh, Rice, and Nakagami-m). For analysis, it is convenient to
model a fading channel with a channel impulse response that includes both time- and space-
varying effects. From the channel impulse response, the multipath autocorrelation and
Doppler profiles are derived, which lead to the concept of the Doppler power spectrum.
Clarke–Doppler and Aulin–Doppler spectra are studied as examples. Using a mathematical
model, the performance of a modulated signal in a fading channel is analysed. First, ideal
coherent demodulation (assuming the carrier phase is always available for symbol-by-
symbol detection) is investigated and the fading channel is assumed to produce only random
amplitude attenuation. Channel tap estimation as well as the channel tap error effect is
studied to reflect real world situations.
Next, the slow fading channel with random amplitude attenuation is investigated for pilot
symbol-aided demodulation (the pilot symbols are periodically transmitted in the symbol
stream; this represents a less desirable situation than ideal coherent demodulation but
remains in line with practical applications). These investigations are extended to OFDM
(the major waveform that was adopted by IEEE 802.11a-g, 802.16 to name a few) where a
slow and frequency-selective fading channel is assumed. The fundamentals of coherent
demodulation are extended to noncoherent demodulation, where the Doppler tracking of
orthogonal signals is investigated. The next discussion centers on another major waveform
that was adopted by IS-95, CDMA 2000, and WCDMA (for use either in the forward
channel or reverse channel or both) for their respective cellular systems, namely, orthogonal
covering and spread spectrum signals. Complex spreading and despreading as well as
Doppler analysis and tracking are presented. For completeness the demodulation of the
signal used in the reverse channel of IS-95 is also presented. Once the Doppler phase error
resulting from either slow or fast fading is corrected via Doppler tracking, and the ISI
8 Introduction

resulting from frequency-selective fading is mitigated via OFDM and equalization, only
random amplitude attenuation remains to be dealt with. This particular effect can be
effectively alleviated via the use of time, frequency, antenna, or multipath delay diversity.
Diversity is a signal combining method that makes use of uncorrelated signal redundancy
for both transmission and reception to enhance symbol detection in the presence of a
deep fade which may destroy a non-diversity symbol. Diversity can be achieved via
redundant symbol interleaving for time diversity, or via uncorrelated subcarrier com-
bining in OFDM for frequency diversity. It can also be achieved via multiple transmit
antennas for transmit antenna diversity, and via multiple receive antennas for receive
antenna diversity or using a Rake receiver for multipath delay diversity. Combinations of
these methods are also possible. Three main signal combining methods are studied:
maximal ratio combining (MRC), selection combining (SC), and equal gain combining
(EGC). MRC is the optimum combining scheme for coherent or pilot symbol-aided
demodulation in AWGN and is the most commonly used method for wireless LAN,
MAN, WAN, and cellular systems. It is superior to SC and EGC, although it cannot be
used for noncoherent demodulation, unlike the other two schemes. Wireless communi-
cation in a fading channel favors the use of multiple transmit antennas for performance
enhancement. This type of antenna diversity employs orthogonal space-time block codes
with rates of ½, ¾, and 1 with MRC at the receiver. The Alamouti code with unity rate
was recommended for the IEEE 802.16 family. Integrating both transmit and receive
antenna diversity provides the receiver with a powerful method to combat random
amplitude attenuation.
The remaining parts of this chapter investigate the capacity of a fading channel. Since
a fade causes an outage that may drive the instantaneous channel capacity to zero, the
average channel capacity defined for an AWGN channel does not exist for a slow fading
channel. Therefore, the outage channel capacity is defined instead. The evaluation of
outage capacity for a slow fading channel, as well as slow fading channels with receive
antenna diversity (SIMO), transmit antenna diversity (MISO), both receive and transmit
antenna diversity (MIMO), and OFDM are presented. For a fast fading channel, the
average channel capacity is well defined since symbols fade independently and there
exists a coding system that ensures a maximum reliable rate, which is the ensemble
average rate. Fast fading is less detrimental than slow fading from the capacity point of
view but requires more complex channel coding to deal with both Doppler phase error
(virtually eliminated via Doppler tracking in slow fading) and random amplitude
attenuation.

1.3 Summary

The structure of the book can be summarized as follows:

* Acquiring the prerequisite knowledge of communication signals: Chapters 2 and 3.
* Packaging the message and introducing the concept of signal-to-noise ratio and bandwidth: Chapter 4.
9 Bibliography

* Measuring and establishing the required signal-to-noise ratio for a given communication coverage:
Chapter 5.
* Sending the message based on the required signal-to-noise ratio and bandwidth: Chapter 6.
* Receiving the message and providing the best detection: Chapter 7.
* Enhancing the survivability of narrowband modulation in the presence of interference via bandwidth
spreading, i.e., spread spectrum modulation: Chapter 8.
* Pulse shaping and equalizing the effect of a bandlimited channel for modulated signals: Chapter 9.
* Dealing with sending and receiving signals over a mobile channel for previously discussed modulation
techniques: Chapter 10.

Bibliography

1. C. E. Shannon, “A mathematical theory of communication,” Bell Syst. Tech. J., Vol. 27,
pp. 379–423, pp. 623–56, 1948. Also available at http://cm.bell-labs.com/cm/ms/what/
shannonday/paper.html.
2. R. W. Hamming, “Error detecting and error correcting codes,” Bell Syst. Tech. J.,
Vol. 29, pp. 147–60, 1950.
3. R. W. Hamming, Coding and Information Theory, 2nd edition, Hemel Hempstead:
Prentice-Hall, 1986.
4. R. G. Gallager, Principles of Digital Communications, Cambridge: Cambridge
University Press, 2008.
2 Deterministic signal analysis

Introduction

In this chapter we lay the foundation for the analysis and design of communication systems,
and digital communication systems in particular. We employ deterministic signals to carry
information from the transmitter to the receiver. These deterministic signals contain certain a
priori features sufficiently adequate for the receiver to retrieve the information. Note that the
information always appears random to the receiver, that is, it does not know which data it
will receive; otherwise, communications would not be needed. Deterministic signals form
a very broad class of signals; therefore, the first step is to categorize them so that their
characterization can be fully exploited. The categorization leads to the labels continuous-
time, discrete-time, periodic, aperiodic, analog, digital, energy, and power signals. Further
study leads us to orthogonal signals and the use of signal space to represent digital signals as
vectors. We also review linear time-invariant (LTI) systems and the important convolution
operation that relates the inputs and outputs of an LTI system.
We then investigate Fourier series representation of continuous-time periodic signals,
and Fourier transform of continuous-time aperiodic signals. The Fourier transform is
indispensable in the analysis and design of LTI systems. The energy spectral density of an
energy signal and the power spectral density of a power signal are studied. From here the
autocorrelation functions of both energy and power signals are examined.
The process of representing a continuous-time signal by its samples is then studied using
the sampling theorem. We also discuss the process of recovering a continuous-time signal
from its samples. Finally, we study various representations of bandpass signals, which are
commonly used in the analysis of communication systems.

2.1 General description of deterministic signals

A deterministic signal is completely specified at any instant of time t. There is no uncertainty

about its value at t. The transmitter employs deterministic signals to carry random informa-
tion. When the receiver receives a transmitted signal that has been corrupted by noise
(a random signal), it attempts to detect the information by stripping away the deterministic
signals. A deterministic signal can fall into a number of categories, which are described
below.
10
 Another Random Document on
Scribd Without Any Related Topics
woodland music—all this thrilled them with the joy of living. “‘When
the Red Gods call,’” whispered George, happily, as Kipling’s poem
came into his mind.
Then they heard again the low, warning rumble of distant rapids,
and once more their hearts beat fast. Anxiously they peered ahead
for a sight of the long lane of “white caps.” The noise became
louder; and, rounding a turn of the river, they saw the rapids tossing
in front of them.
This time they had no fear when the canoe, with Ben standing in the
stern, raced down through the center of that wild course. They had
implicit confidence in the skill of the guide, and they enjoyed each
moment as the little bark plunged and careened in its uncertain
passage among the waves. As before, Ben brought them safely
through, and paddled on down the river.
It was late in the afternoon when two sturdy figures emerged from
the edge of the woods and hailed the canoe. Ben replied, and told
the boys that they were lumbermen. He said they would soon reach
the great camp itself, now but a short distance farther on.
“We’ve made a whole lot better time coming down than we’ll be able
to make going back,” he said, when Ed expressed surprise that they
had finished the trip so soon. “Fact is, we’ve ridden down on the
back of the flood; but we’ll return with what is left of it pushing us in
the face.”
On both sides of the river were many logs lying along the bank close
to the water. They had been cut and dragged there during the
winter, and when the water subsided to the desired level they would
be rolled into it and floated down to the mill, many miles below.
Other men now appeared along the shores and waved their hands
cheerfully at the canoemen.
“They have their booms stretched,” said Ben, pointing to a long line
of floating logs chained one to the other. “That means they’re
intending to send the logs down—probably to-morrow.”
The boys found that the booms were used to guide the logs in their
course, and to hold them back at certain stations until the stream
below was cleared for their passage or a jam broken up. He told
them that a jam was a great tangled pile-up of logs, caused by one
or two logs grounding, or jamming, and obstructing the progress of
the hundreds afloat behind them.
It appeared that patrols of lumbermen were stationed along the
river, while the logs were “running,” to watch for just such
emergencies. These men would go fearlessly to work to break a jam,
a hard and dangerous task. If unsuccessful, they would run to the
nearest of the telephone-boxes, which the company had placed at
intervals along the shore, and summon aid. Sometimes a bad jam
required the work of several days to break it, and dynamite was
often used in such cases.
Twilight was gathering when they spied a clearing some distance
along on the right bank of the river. As they approached they saw
that it contained several long, low, log cabins.
“There’s the camp,” said Ben, and he headed the bow of the canoe
toward shore.
As he drew closer he shouted. Immediately some men made their
appearance in the doorway of one cabin, and, seeing the canoe,
came outside. Two of them left the group and started slowly toward
the water. When they had come near enough to recognize Ben, they
called to him and pointed to a suitable landing-place. As soon as the
bow of the canoe touched bottom, they pulled it from the water, and
the three occupants stepped out.
“Howdy, Ben,” greeted the two lumbermen.
“Helloa, Ned! Helloa, Jim!” replied the guide, shaking each of them
by the hand. Then he motioned to the boys. “These are young
friends of mine, Ed Williams and George Rand; they’ve been
spending the winter with me. Boys, shake hands with Ned Crawford
and Jim Halliday. Crawford is foreman of this crew, and Jim is the
greatest log-rider in the country.”
Ed and George unloaded their packs from the canoe and, manfully
refusing an offer of assistance, shouldered the loads and followed
Ben toward the cabins. They saw Ned wink at the guide and nod
toward them approvingly, and they were glad they had declined help
and “rustled” for themselves.
“The boys are at supper,” explained the big foreman, indicating the
cabin with the smoking chimney. “I’ll take you fellows over to the
bunk-house, and you can leave your ‘turkeys,’ and wash up a bit.
Then we’ll mosey over and have something to eat.”
The lads looked about them curiously when they entered the great
house with its rows of bunks. Along the first half of each side were
long benches. In the center of the room was a huge, round stove,
and, although it contained no fire—for the nights were not cold
enough to require one—many articles of clothing were hung upon it
to dry. A large kerosene-filled lamp, suspended from overhead in the
middle of the room, furnished what light there was.
“You can leave your ‘turkeys’ over there; take those two bunks,” said
Crawford, designating the proper ones.
Ben promptly tossed his pack into one of them and smiled broadly
when the boys looked at him in some confusion.
“Your pack, or ‘duffle,’ is your ‘turkey,’” laughed the guide. “Put yours
in that lower bunk.”
“Yep, that’s what we call it,” the foreman explained. “Now, you’ll find
the basin and a pail of water outside by the door, and here’s a clean
towel. When you’re all fixed up handsome like, come over and join
us in the ‘grub-shanty,’” he said, and left them.
When they had stowed away their belongings, they went outside
and washed at the battered tin basin. While they were going through
this interesting and necessary ceremony several lumbermen came
up to shake hands with Ben. Then they went to the “grub-shanty,” or
dining-cabin, where the foreman was waiting to welcome them.
A long pine table, flanked on either side by low, continuous benches,
extended almost the length of the room. At it sat the lumber crew—
deep-chested, quick-eyed men of the wilderness. Ben readily called
most of them by name as he and the boys took seats.
Soon a big, swarthy-faced man, who wore a soiled apron and had
his sleeves rolled above his elbows, came scuffling in and placed
three smoking bowls of oatmeal—“oats” he called it—before the
new-comers. Then he struck the guide a resounding whack between
the shoulders, as proof of his delight at seeing him.
“Helloa, Charley!” said Ben, when he had recovered sufficient breath.
“Why didn’t you hit me with an ax?—it wouldn’t have hurt so much.”
The cook roared his delight at the compliment to his strength, and
Ben introduced Ed and George. “Remember to always keep on the
proper side of the cook, and you’ll come out all right,” laughed the
guide.
They watched Charley disappear into a sort of out-shanty, where
several other men in aprons seemed to be fussing about with pots
and pans. Presently he reappeared and supplied them with cups of
tea, “sour-dough bread,” and a plentiful portion of baked beans.
“Pile into it strong,” he urged. “You’ll never get the gout from any
fancy dishes of mine.”
The boys did not delay. The trip and the spring air had sharpened
their appetites. They instantly became popular with the cook by their
devotion to the substantial fare set before them, and from time to
time they cast curious glances at the long rows of jolly, brown-faced
men with whom they sat.
They had heard and read so much about the “lumber-jacks,” and
they wanted to find out what sort of fellows they were. They were
compelled to laugh outright at the quaint expressions used by these
men in asking for the various things on the table.
“Hey, Joe, chase the cow down here, will you?” And at once Joe
understood and passed the milk down the table toward several of his
friends, who were calling, “Co boss, co boss.” “Roll along the spuds”
meant to pass the potatoes. “Say, Charley, I’m shy a stabber,” was
replied to by the gift of a fork. A spoon was alluded to as a “dipper,”
and so on through the entire list. Ben laughingly explained each
phrase as it was employed, and the boys memorized it with the
purpose of trying it on the family at home.
When the meal was finished, they accompanied Ben to the canoe for
the string of fish, which he presented to Ned, with the compliments
of the young anglers.
“There’s sure one dandy fish in that bunch, and that’s calling it
something, ’cause they’re all dandies,” declared the foreman, holding
the trout at arm’s length, so that all might admire it.
Then they went into the bunk-house and took places on “the
mourners’ bench,” which was what Ben said the lumbermen had
christened the seats along the sides of the cabin.
“When do you figure on running your logs?” he asked, when Ned
came in and seated himself beside them.
“To-morrow, if things go as we expect. Got all our booms stretched,
and the water ought to be right if she slides down a notch or two
before morning. Quite a gang of the boys along the river now—
boom gangs,” he explained.
“Well, boys, we’re just in time to see the fun, and I guess Ned won’t
mind if we stay around a day or two,” said Ben. “You see, Ned, I’d
like these fellows to see something of a real log-drive before they go
home.”
“The latch-string of this here camp is always out for you and your
friends, Ben; and the longer your hat hangs on the peg, the better
we like it,” was the foreman’s reply.
For some time the men talked together in little groups ranged along
the wall. The guide seized the opportunity to make Ed and George
familiar with some famous characters of the logging country. There
was “Shorty” Brundage, a square-shouldered, stockily built young
fellow, who bore the proud distinction of having loosened more jams
than any other man in the crew. Several times he had escaped death
by the merest margin. Next to him sat “Red” Thompson, who had
achieved fame by “riding” a log through the first set of rapids.
Slightly farther along, a dark-skinned man was stooped over unlacing
his “larrigans.” Ben said that he was the renowned Pierre La Valley,
known throughout the big woods wherever an ax was swung. With a
double-headed ax he could fell a tree quicker than any two men. At
each swing he turned the blade so that every stroke was made with
an alternate edge. His fame as an axman had traveled abroad over
the entire lumber country, and scores of good men had been
matched against him; but as yet he was undefeated. At the far end
of the bench was “Jake” Grant, champion “birler” of the crew. The
boys asked what “birler” meant, and were told they would learn
before they left the camp.
This roll-call of heroes was interrupted when some one called for
Tony and his fiddle. A tall youth, with the features and hair of an
Indian, brought forth a violin and seated himself at the head of “the
mourners’ bench.”
“He’s a half-breed,” whispered Ben.
“Cut her loose, Tony!”
“Open her up wide!”
“Wat you fellows want?” asked the fiddler.
“‘Turkey in the Straw,’” cried some.
“‘Old Dan Tucker,’” urged others.
Tony favored those making the most noise, and started the lively
strains of “Turkey in the Straw,” a jig dear to the heart of every
backwoodsman. The men kept time with feet and hands, and before
long the boys joined in. “Lanky” Jack Stewart brought out a mouth-
organ, and added his efforts to those of Tony. Then the foreman
produced an accordion, and the home-made orchestra was
complete. The music was loud and lively; and, unable to restrain
their buoyant spirits, several of the “lumber-jacks” jumped to their
feet and began to dance a “shake-down.”
The lads soon found themselves yelling like the rest, for the fun was
contagious. Ben looked at them and smiled across at the foreman,
who was rocking his body to the accompaniment of the notes from
his accordion.
The merrymaking finally wound up with an impromptu Virginia Reel,
in which the three visitors were compelled to dance. It was no
gentle affair, that “going down the line.” Resounding whacks stung
the bodies of the good-natured victims who gamely ran the gantlet
between lines of whooping dancers. The boys, too, were soundly
spanked before they reached the end of it. They retaliated
enthusiastically on the couples that followed.
“Red Thompson and Miss Hank Davis are going down the line!”
shouted Crawford, as the two “lumber-jacks,” one of whom played
“lady,” turned to make their rush. And again: “Mr. and ‘Mrs.’ Ben
Adams are going down the line,” as Ben and his partner, Jim
Halliday, made their way to the end of the column. The boys laughed
heartily when their turn came, and Ned called out, “A pair of
bantams are going down the line.”
At last the music ceased, and the foreman said he guessed it was
time to turn in. Pushing and wrestling good-naturedly, the men
made their way to the bunks. Then Crawford extinguished the lamp,
and the cabin became dark and quiet.
 XXIV
A DARING RESCUE

T he boys were awakened before daylight by the “turning out” of

the crew. They found Ben up and ready for breakfast, and,
hastening into their clothes, they joined the line of jolly men waiting
their turn at the tin wash-basin. Having finally reached it, they
completed their toilet and followed Ben into the “grub-shanty.”
The merry meal was soon over, and, with an invitation to come and
see them at work, the lumbermen started for the woods and the
river. Ned, the foreman, was too busy to give much attention to his
guests, and telling the guide where he might find him at work, he
quickly followed after his men.
Left alone, Ben and the boys, feeling thoroughly at home, began a
closer inspection of the camp. Charley, the cook, came out and
joined them, and they found him a quaint and interesting character.
He told of his experiences in the lumber woods, and, of course,
made a great impression on Ed and George. When they had been
sitting there some little time, the guide rose and pointed toward the
river.
“They’re running; here comes the head of the drive!” he cried.
Plowing their way swiftly down the current, the boys saw an
advance-guard of huge logs. Close behind were others, and as the
boys hurried to the water’s edge, they saw the river was dotted with
logs as far up as they could look. The majority drifted rapidly past,
well out in the middle of the stream. Occasionally, however, one
would swerve and bang against the bank, or become temporarily
stranded on a sand-bar or pebbly beach.
Suddenly they saw Jim Halliday sweep into view around a turn of the
river. He was standing erect on one of the drifting logs, boldly
“riding” it down the stream. The boys watched him in wide-eyed
amazement as he came gliding along, balancing himself with a long
peavey-pole, which he held horizontally. When he was opposite, the
“lumber-jack” waved his arm, and they cheered him. Jim skilfully
steered the log into a quiet eddy beyond, and, jumping into the
water, sent his “wooden horse” down the river and waded briskly
ashore.
“That’s what they call log-riding,” explained Ben.
 “RIDING” A LOG DOWN THE STREAM
Halliday had been despatched back to camp on an errand, and had
taken advantage of the opportunity for his first ride of the season.
When he was ready to return up-river, he asked Ben and the boys to
accompany him, declaring that there was “a barrel of fun” going on
up there.
They followed Jim several miles along a winding river trail, until they
came to a large clearing where the men were at work. Here the
crew were rolling logs down the steep bank into the water. Strong-
armed, quick-footed fellows started a huge log, and then jumped
nimbly out of its way as it went bumping down the incline to land in
the river. There other men, immersed to their waists, tugged and
pushed till they worked it into the current and started it on its long,
uncertain course to the distant mill. The great collection of logs
extended well back into the woods. As fast as the front ones were
moved more were dragged forward by the teams. It was an
absorbing scene of bustling activity. As Jim had said, there was
plenty of sport. They saw several of the crew pushed from the bank
by their frolicsome comrades. The victims always took their ducking
good-humoredly, and scrambled from the water determined to
retaliate.
Just before noon a long, high-prowed boat, resembling a fisherman’s
dory in its general lines, was seen slowly approaching against the
current. This was the bateau. Two broad-shouldered fellows were at
the oars, and in the bow was another, pole in hand, prepared to fend
off drifting logs. In the stern sat no less a personage than Charley,
the cook.
At sight of the boat Ned called a halt in the work, and then the real
fun began. The men seized each other in rough sport, until almost
the whole crew were wrestling about on the ground.
“This way, Charley; bring it up here,” ordered the foreman.
The cook and his three assistants struggled up the bank with the
supplies for a hot dinner. Grouped in a circle on the ground, each
man was equipped with a tin plate, a knife, a fork, and a spoon, and
a large tin bowl which was speedily filled with hot stew. After that
came heaping dishes of hot beans and steaming cups of coffee. Like
the others, Ben and the boys ate the outdoor meal with keen relish.
As neither Ed nor George had seen a birling contest, and had no
idea what it was like, Ned arranged one for their benefit.
A large log was towed out into fairly deep water in a near-by eddy of
the river. Then Jake Grant, the champion of this particular sport,
jumped from the bank and landed on the log. He caught his balance
and drove the long, nail-like calks of his shoes deep into the bark.
His action was intended as a challenge, or “defi,” to any one to jump
on the other end and enter the contest.
There was a cry of “Sandy,” and, amid cheers from his comrades,
“Sandy” Donaldson accepted the challenge. Moving their feet up and
down together, the men whirled the log over and over beneath
them. From time to time one or the other would jam his spiked
shoes down hard in an endeavor to stop the log and throw his
adversary into the water. First one way, then the other, they spun
the log faster and faster. The excitement on shore was intense, for
each contestant had his partisans. Once Grant lost his balance for a
second, and a wild yell went up. It looked as though his long term of
championship was about to end. By wonderful agility he saved
himself, and another cheer broke forth.
“Toss him, toss him, Jake!” cried Grant’s friends.
“Bump him off, Sandy; you nearly caught him that time!” yelled the
partisans of Donaldson.
Then the experience of the champion came to his aid. He worked
backward toward the extreme end of the log, and started it spinning
as fast as his nimble feet could work. Donaldson kept stride with
him, and those on shore waited in breathless suspense for the
outcome of what they believed a clever bit of stratagem.
Suddenly the champion jammed his spikes down hard, and
Donaldson tottered. Before he could recover his balance Grant
jumped into the air. Instantly his end of the log rose, and that on
which his opponent swayed sank—not much, but enough to do the
trick—and Donaldson went over backward into the river. Jake was
still champion, for he landed squarely on top of the log and waved
his hand to the cheering crowd.
“I wouldn’t have missed that for anything,” said Ed, when he had
ceased cheering.
“Nor I,” declared George. “Say, we’ll try that some time, Ed. What do
you call it, Ben?”
“Birling,” explained the guide.
The cook, who with his crew had waited to see the outcome of the
contest, approached Ben and offered to take them back in the boat.
That was certainly better than “hiking,” and presently they were
gliding swiftly down the river.
That night there was less frolicking in the bunk-house. The first day’s
river work had tired the men, and they sat about quietly smoking
and telling yarns and singing a few lumbermen’s songs before they
turned in.
Next morning they paddled their canoe some distance down-river to
see a big restraining boom. They were obliged to dodge floating
logs, which dotted the river as far as they could see. A collision
would have smashed their light craft. However, Ben kept safely out
of the way, and, searching the water far in advance, he chose open
channels, down which he piloted them in safety.
They saw many logs which had grounded along the shore, but they
learned that these would be all found and set afloat by the
“reardrive” men, who followed the last of the logs down the river for
that purpose. Occasionally they passed members of the river patrol,
who stood on the banks and waved to them as they floated by. Ben
pointed out several bark shanties, or lean-tos, in which these men
sheltered themselves until the drive was over.
Then, at a narrow place in the river, they saw a great jagged pile of
logs. Others were constantly crashing into it and momentarily adding
to the tangle.
“There’s a jam, and a nasty one!” said Ben, carefully working the
canoe toward shore. “Wonder if any one is about?”
“Yes, I see a man,” declared George. “There he is, right under the
big log that sticks out. Gee whiz! Did you see that one rise right up
on end and sail past his head?”
They landed and ran along the bank until they drew near the spot.
At the foot of the towering pyramid a red-shirted man was balancing
on a slippery log and prying and pulling with all his might in an effort
to free the log which was the key of the jam. Each moment he
stayed there he was risking death from the grinding, crunching,
splintering logs which the river was raising on end and throwing into
the pile behind him. Calm and undismayed by his peril, he turned a
flushed, perspiring face and called to Ben:
“Hey, run back up the trail a piece, till you see a box nailed to a big
white pine. You’ll find a telephone inside. Tell ’em to send some men
down here, quick, an’ to stop shovin’ in any more sticks till we git
this straightened out.”
Bidding the boys remain where they were, Ben ran up the trail on his
important errand. Anxious to help the plucky lumberman, but
knowing themselves powerless, the boys, fearful but fascinated,
could only stand and watch the reckless worker out there in that
inferno of flying logs.
A great black hulk rose from the foaming water, shot into the air, and
came straight at him. An exclamation of horror came from the white-
faced spectators on shore. His quick eye and alert brain proved
equal to the emergency, however, and he jumped back and just
escaped being crushed. A cheer sounded from the lads on the bank,
and the “lumber-jack” turned and waved his appreciation.
“It’s ‘Shorty’ Brundage, the champion jam-breaker!” cried Ed.
They watched him in awed silence while he went on with his
hazardous task. Dodging and climbing, he seemed to escape
destruction by simply the luck for which he was noted. Above him
towered the great mass of piled logs. Should it give way, he would
be buried beneath an avalanche. On each side great logs shot past
within reach of his arm. Below, the river caught and tugged at his
legs in an effort to sweep him to destruction. Still he worked on, his
one thought the breaking of the jam and the clearing of the stream.
Suddenly he slipped, lost his balance, and fell into the swirling,
foam-tossed water. They saw him reappear, a long, red streak
showing down the side of his pallid face. He made a desperate effort
to climb upon the log from which he had fallen; but two floating
timbers caught him between them, and with a despairing gesture
“Shorty” collapsed. Half in the water, half across one of the logs, he
was in peril of being crushed to a pulp by the massive logs which
reared themselves from the water and crashed down on all sides of
him.
For a moment the boys stood paralyzed with horror. Then they
realized that they were standing there without an effort to save the
unconscious man. There was one startled glance at the towering log
pile, the raging, white-capped water, and the crashing logs. Then
their gaze settled on the helpless red-shirted figure in deadly danger.
Instantly they made their choice. With white, set faces the lads ran
down the bank and along the edge of the racing water toward the
jam.
Out along the top of a slippery log they crawled, one behind the
other. They dared not stand erect, for fear of falling into the
seething, log-studded pool beneath. The noise was terrific. In some
places the raging torrent surged above their waists and threatened
to sweep them from the log.
“Hold on with your legs!” shouted Ed.
Then a long, spear-like splinter was thrust at him like a lance. He
dodged just in time, and the splinter flew over his back. Again and
again the whirling logs nearly crushed their legs. But at last only two
feet of open water intervened between them and the log on which
lay the victim of the jam.
As they were hesitating, the butt of another log was driven into the
space and for the moment wedged fast. In an instant Ed had thrown
himself across it to the log that supported Brundage. Quickly he
crawled to the prostrate figure, and, placing his hands beneath the
powerful shoulders, he tried to pull the man from his perilous
position. In an instant George reached his side. After much effort
they managed, between them, to lift “Shorty” from the water and
drag him some distance along the slippery log over which they had
come.
No sooner had they removed him than two great logs were lifted by
the water and sent crashing down upon the very log across which
his unconscious form had lain.
Lifting, pulling, and tugging with all their strength, Ed and George
managed, somehow, to crawl over the wet logs, dragging “Shorty”
with them. They had many narrow escapes, but at last the bank was
reached. Pulling their man up beyond the reach of the angry river,
they dropped beside him, too exhausted to move.
Then Ben, who, as he was returning, had seen the whole daring
rescue, rushed breathlessly down the bank and hugged both boys in
his arms.
“You fellows are the real thing!” he cried to the blushing boys.
“You’ve got sand, and you know when to use it. That was one of the
pluckiest pieces of work I ever saw done.”
“What’s—the—matter?” asked Brundage, feebly, opening his eyes
and staring about him in bewilderment.
“The matter is that the jam came powerful near getting you, ‘Shorty.’
It would have got you, too, only for these young fellows. They got to
you, somehow, and, what’s more wonderful, they got back and
brought you with them! That splintered log out there, the one with
the three big ones lying across it, is the log you were on,” said Ben.
The “lumber-jack” passed his hand weakly across his forehead,
blinked, and sat up. He beckoned the boys to come close; and when
they had done so, he reached up and grasped their hands.
“Much—obliged—pardners,” he said.
They thrilled at the last word. It was the greatest compliment this
big, brave man of the woods could have paid them—he had placed
them on an equality of manhood with himself.
“What about the jam?” he queried, in a half-dazed manner. “Did you
tell them to stop the ‘sticks’?”
“Yes,” Ben assured him, “and Crawford and a picked crew are on
their way down. Here they come now.”
Down the middle of the river came the bateau filled with
lumbermen. The big foreman was in the bow. Spying the trouble
ahead, he bawled his orders to the stalwart oarsmen, and the boat
was quickly beached beside the little canoe.
The crew at once leaped out and came running to attack the huge
pile of obstructing logs. They were armed with peavey-poles, axes,
and steel bars. The boys could hardly keep from cheering these
heroes of the river as they rushed forward to grapple with the jam.
“What’s up? Did it get you, ‘Shorty’?” inquired the foreman, bending
anxiously over the stricken river-man.
“Pretty nigh got me, Ned,” laughed “Shorty,” feebly. “Guess I’d have
gone if it hadn’t been for these lads. They ran out there and got
me.”
The lumbermen had gathered about their injured comrade, and as
he spoke they turned to the boys. They slapped them affectionately
on the back and praised them for their bravery. Then they gave
three mighty cheers which roared and echoed up and down the river
for a mile.
“Well, let’s sail into it!” yelled Crawford.
His crew went to work with a will, and Ed and George stood by,
interested witnesses of the determined onslaught. They saw daring
chances taken by reckless men, and several miraculous escapes. The
foreman himself led in the perilous work. He it was who, when the
pile of logs began to creak and give and totter, stood calmly below
them until the last of his crew had jumped to safety. Then, as the
jam crumbled and broke, he, too, leaped from under the crash and
ran nimbly across the tops of plunging logs to shore.
When the obstacle had been overcome, Crawford despatched a man
up the river to the telephone-box, with orders for the crew above to
resume operations. It was important that no time should be lost in
getting the logs through while the high water lasted. Much against
his wishes, “Shorty” was ordered to camp, and a new man was left
on watch at the dangerous turn of the river.
Assisted by the boys, Ben paddled the canoe upstream, following in
the wake of the bateau. The men in the latter struck up a familiar
river song, and the music came drifting back over the water.
“I should think they would be too tired to sing, after what they’ve
just been through,” said George.
“They’ve forgotten about it long ago; it’s all part of the day’s work,”
laughed Ben.
The boat soon drew away in advance of the canoe. When the
occupants of the latter disembarked at the camp, the larger craft
had already landed “Shorty” and disappeared up the river.
That night was a gala one in the bunk-house. The whole crew of
“lumber-jacks,” having learned of the boys’ gallant rescue of
Brundage, came into camp to do them honor. Men arrived from far
down the river to have a “look at the youngsters” who had made
good. The merrymaking was shared by all, and there was a general
thanksgiving because “Shorty,” the best jam-breaker in the country,
had been spared to his crew. Nor was Charley, the cook, to be
outdone. When the fun had reached its height, he marched solemnly
to the door of the bunk-house and, waving his arm toward the
“grub-shanty,” invited them all over for a surprise party.
With loud cheers and whoops of approval, the noisy gathering sat
down at the long table laden with good things which the cook and
his assistants had prepared for the occasion. There were several rare
treats; and when the feast was over, three rousing cheers were
given for “the best cooks in the lumber country.”
Big-hearted Ned Crawford and his hospitable crew were for keeping
their guests several days longer, but Ben explained that they must
leave next day. Warmly seconded by the boys, he thanked the
lumbermen for their hospitality and this glimpse of a new life.
Early next morning, when the canoe had been loaded and the
departing guests were about to step into it, “Shorty” Brundage came
forward and addressed them.
“I’m not much on speeches,” he laughed, nervously, as he grasped
the hands of his rescuers; “but I want you fellows to know that I feel
what you did for me. It sure was a white thing to do, and any time I
can do anything for you, just call on me like a pardner. I reckon Ben
can tell you what that means in this country.”
The lads thanked him, and tried to make light of their exploit. Then
they took their places in the canoe, and Ben put away from shore.
Charley and his little crew waved their aprons and beat a loud salute
on tin pans, and the foreman and some of his men accompanied the
departing guests in the bateau as far as the log pile. Here more
rousing cheers were given the voyagers, until they rounded a turn of
the river and disappeared.
“Well, you fellows have made a hit with that crowd. Your names will
be known along this river for a good many years to come,” said Ben,
proudly. “How did you enjoy it all, anyway?”
“Ben, I’ve had the time of my life!” declared Ed. “As for the ‘lumber-
jacks,’ as you call them, I think they are the manliest, jolliest, best
all-round lot of fellows I ever met. As ‘Shorty’ would say, they sure
treated us white.”
“They did that,” chimed in George; “and I want to say I have
enjoyed it all immensely.”
“Well, I feel some better about having taken you, now, than I did
yesterday, when I was running along the river-bank while you were
out there below the jam. I want to tell you that you had a mighty
close shave, boys, and I’m powerful thankful it turned out as it did,”
he added, soberly.
 XXV
CALLED HOME

I t was well toward the evening of the day following when they
reached Ben’s cabin. The trip up the river had taken them the
best part of two days. The canoemen were agreeably surprised to
see Bill and Moze awaiting them at the landing.
“Helloa, you fellows! Where have you been?” asked the trapper, as
they stepped ashore and grasped him warmly by the hand.
“Down at the lumber camp; I promised to take the boys there before
they went home,” explained Ben.
Ed and George at once started a good-natured wrestling bout with
old Moze, who seemed delighted at seeing them. Barking and
whining, he jumped about, wagging his tail, until George declared he
would lose it.
“When did you get here, Bill?” inquired the guide of his friend, after
they had entered the cabin.
“Same morning you left, I reckon. When I saw you were away, I left
a note and went on into the settlement. Got back late last night, and
I figured I’d mosey around for a day or two and see if you came
home. I brought out some mail for Westbrook and you fellows.
There’s yours over on the table yonder. By the way, Ben, Westbrook
said to tell you he expected to go in with his team and wagon the
first part of next week. He said he’d come by here on the old wood-
road, so if you wanted anything you could let him know.”
The boys quickly found the letters from home and began to read
them. From their manner Ben knew that they had received news of
some importance. When they had finished the letters they held a
low-toned conference, and then turned toward the guide, who was
watching them keenly.
“Nothing wrong, I hope?” he said.
“Nothing, except that we are ordered home,” answered Ed,
endeavoring to force a smile.
“The deuce you say!” cried Ben, in dismay.
“Yes, I guess our time is up,” declared George, rather gloomily. “You
see, our folks are going away in about two weeks, and, as we expect
to enter college in the fall, they won’t see much of us for a while.
Consequently they want us with them this summer.”
“Guess that’s only natural,” said Ben, quietly. “Say, it’s going to be
some lonesome after you fellows pull out.”
For a time the little group of friends sat about in silence. While they
were anxious and eager to see their parents, the boys nevertheless
felt badly about leaving Ben, and Bill, and Moze, and the great
peaceful forest they had learned to love.
“Well, say, there’s no need of everybody getting the ‘blues,’” laughed
the guide. “I’ll hustle around and cook some supper, and then maybe
we’ll all feel better.”
That night the boys sat up late, and Ben told Bill of their plucky
adventure at the log-jam. It was decided that they should go to the
station with Tom Westbrook and his team early the next week. Bill
said he would wait over at the cabin to see them off.
It rained all the next day, and the boys busied themselves with
packing their belongings. Not the least important were several
trophies of the hunt, which they were taking home in proof of their
prowess.
The following days were spent in the woods and on the lake with
Ben and Bill. Rare golden days they were, filled with flowers and
sunshine and song, for the long, dreamy days of early summer had
arrived. The boys learned more of the songs and calls of the birds,
and the names and uses of the many wild flowers which were
constantly bursting into bloom in the woods about the cabin. They
saw fishes guarding their gravelly nests of spawn at the bottom of
the lake. They found the eggs of turtles at the end of long, tunnel-
like excavations in sandy banks. The time sped rapidly, and at length
the day for their departure was but a night away.
“Well, this is our last snooze on balsam tips,” said Ed, when they
were finally in their bunk.
“Yes, and, do you know, sometimes it all seems like a dream. We
certainly have been through a few experiences since we left the city.
And let me tell you, Ed, each of them has done us good; I feel that
we can take care of ourselves anywhere now,” replied George. “Say,
we never found out who ‘The Old Man of the Woods’ is.”
“Well then, I’ll tell you now,” laughed Ben, who had overheard their
conversation. “You met him right here the first night, when you
worried about the flying squirrels. He has been with you ever since,
until you know him well enough to call him by his right name, which
is Experience.”
“And is that how you met him?” inquired the boys.
“That is how we must all meet him, if we expect to ever know him
well. He’s a rough old fellow, and he don’t make friends easily.
You’ve got to prove your worth before he accepts you. If you’re
game, he’ll take care of you in great shape and tell you all he knows.
But if you’re a quitter, he’ll soon drive you out of his country and
make things so unpleasant you’ll never wish to return. Now you
know him, and, furthermore, you can consider him your friend,
because you’ve made good. Good night.”
“Thank you, Ben. Good night.” And they went to sleep, happy over
the well-earned compliment.
They had just finished breakfast next morning when the loud
barking of Moze announced the arrival of Westbrook with his team.
A moment later the big woodsman stood in the doorway and
greeted them.
“Have some breakfast,” urged Ben, hospitably.
“No, thanks; I had my breakfast before I started. Thought I’d just
stop a minute to see if I could do anything for you in town.”
“You can take the boys and myself and this luggage, if you’re
traveling in light,” said Ben.
“I sure will; but, say, now, it’s too bad you fellows have to pull up
stakes so sudden like,” he said, turning to the lads. “This is the
prettiest time of the year, too. Guess you’ve had quite a time of it,
though, haven’t you?”
“Yes, indeed,” replied Ed; “we’ve enjoyed ourselves very much.”
After Ben had insisted, Westbrook drank a cup of coffee. Then they
proceeded to load the baggage into the same wagon that had
conveyed it from the station almost eight months before.
“Different-looking boys from those that came out in the fall, eh,
Ben?” laughed Westbrook, when everything was in and they were
ready to climb aboard.
“Some browner,” said Ben.
Bill came forward to say good-by, and Ed and George grasped his
hand affectionately. They thanked him for all the pleasure he had
given them, and said they hoped some day to come out and see him
again. Old Moze came wriggling toward them, and they stooped and
petted him.
Then Westbrook spoke to his team, and the wagon bumped away
over the long road. The boys waved their hats to Bill, who stood in
the doorway, holding Moze by an ear to prevent him from following
his departing friends.
At noon they halted, and Ben cooked a simple meal in the woods.
This time the boys built the fire, and the two woodsmen declared it
a proper one.
“Couldn’t have done that when you came in,” laughed Ben.
“We couldn’t do much of anything then,” replied Ed.
When they resumed their journey, the lads recalled many familiar
spots along the way. They went safely over the very place where
they had mired to the hubs before, and saw their own pile of rocks
and the poles with which they had freed the wagon. They forded the
stream which had come near capsizing the load. This time they had
no trouble. They bounced and bumped over the rough “corduroy”
bridges which had so excited their curiosity on the way to camp.
Several times they flushed grouse from close beside the road, and as
the birds rose with a roar of wings Ben looked at the lads and
smiled, and George was reminded of his first fright in the woods.
They started a deer from the edge of a swamp, and watched the
nimble-footed creature go bounding along the road ahead of them.
Thus the interesting ride continued, until late in the afternoon they
drew up before the tiny log station.
The agent remembered them, and said they were lucky to reach
there at that time, as an express was due to stop in thirty minutes.
Had they missed it they would have been obliged to remain over
until the next day. The boys quickly checked their baggage, and then
came outside to await the arrival of the train.
“Ben, I’m sure we can never thank you half enough,” said Ed, with
feeling.
“Don’t try,” laughed Ben; “I’ve enjoyed it as much as you have. The
worst of it all is this breaking up of camp; a woodsman always hates
it.” And as he turned to look up the track the boys thought they saw
moisture in his kindly eyes. “Well, it hasn’t hurt you any,” he added,
quickly recovering himself.
“I should say not,” said George. “I feel as if I could wrestle with a
bear.”
“You almost did, once,” Ben chuckled.
“Here she comes. Stand back!” cried the agent, as the great
whistling engine made its appearance around a curve.
The boys shook hands with Ben and Westbrook and then stood on
the rear platform of the last car and waved their hats at the two
broad-shouldered men left behind when the train pulled out for
home.
Two days later they were met at the great city terminal by their
fathers. Both Dr. Williams and Mr. Rand were surprised and delighted
at the improvement in the boys. They took critical note of the firm,
bronzed skin, the broadened shoulders and deepened chests, the
direct, keen glance of the eyes, and, above all, the erect, confident
carriage and free, swinging stride. The inspection ended in approval
—Ben had molded well the raw material placed in his hands, and the
result was a credit to him.
The experiment was a success; the theory was proven. The lads had
returned with interest on the investment. They had gained not only
in health and strength, but in much besides. They had a finer,
clearer, broader view of life and the living. They had learned peace
and beauty and quietness from God’s temple, the wilderness. They
had gained strength from the pine-scented air, courage from
exposure to hardship and peril, resourcefulness from dependence
upon themselves, and a sound knowledge of sound truths from
honest old Ben.
“By Jove, Doctor, your plan has worked out wonderfully,” declared
Mr. Rand. “I wouldn’t have believed it possible. Why, look at George;
he is fit for the freshman crew.” And the proud father was delighted
at the prospect, for he, too, had worked from freshman to ’varsity
oarsman in his college days.
“Well, you think it has done them as much good as your tutoring trip
through Europe, do you?” laughed the Doctor.
“More!” declared Mr. Rand, enthusiastically. “In fact, I’m inclined to
agree to that Western proposition of yours for next summer, now
that I’ve seen the results of this trip.”
“What is that?” demanded the boys, aroused at the idea of a future
trip to the wilds.
“Never mind,” laughed Dr. Williams, winking slyly; “we’ll give you a
surprise when the time comes.”
 Transcriber’s Note:
Illustrations have been moved to paragraph breaks near where they are mentioned.
Punctuation has been made consistent.
Variations in spelling and hyphenation were retained as they appear in the original
publication, except that obvious typographical errors have been corrected.
A change has been made as follows:
p. 306: Illustration caption added to agree with the Table of Illustrations.
*** END OF THE PROJECT GUTENBERG EBOOK CAMPING IN THE
WINTER WOODS: ADVENTURES OF TWO BOYS IN THE MAINE
WOODS ***

Updated editions will replace the previous one—the old editions

will be renamed.

Creating the works from print editions not protected by U.S.

copyright law means that no one owns a United States
copyright in these works, so the Foundation (and you!) can copy
and distribute it in the United States without permission and
without paying copyright royalties. Special rules, set forth in the
General Terms of Use part of this license, apply to copying and
distributing Project Gutenberg™ electronic works to protect the
PROJECT GUTENBERG™ concept and trademark. Project
Gutenberg is a registered trademark, and may not be used if
you charge for an eBook, except by following the terms of the
trademark license, including paying royalties for use of the
Project Gutenberg trademark. If you do not charge anything for
copies of this eBook, complying with the trademark license is
very easy. You may use this eBook for nearly any purpose such
as creation of derivative works, reports, performances and
research. Project Gutenberg eBooks may be modified and
printed and given away—you may do practically ANYTHING in
the United States with eBooks not protected by U.S. copyright
law. Redistribution is subject to the trademark license, especially
commercial redistribution.

START: FULL LICENSE

Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com