Download the Full Version of textbook for Fast Typing at textbookfull.

com

Physical Layer Security for Cooperative Relay

Networks 1st Edition Yulong Zou

https://textbookfull.com/product/physical-layer-security-
for-cooperative-relay-networks-1st-edition-yulong-zou/

OR CLICK BUTTON

DOWNLOAD NOW

Download More textbook Instantly Today - Get Yours Now at textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Physical Layer Security in Random Cellular Networks 1st

Edition Hui-Ming Wang

https://textbookfull.com/product/physical-layer-security-in-random-
cellular-networks-1st-edition-hui-ming-wang/

textboxfull.com

5G Physical Layer Technologies (IEEE Press) 1st Edition

Abu-Rgheff

https://textbookfull.com/product/5g-physical-layer-technologies-ieee-
press-1st-edition-abu-rgheff/

textboxfull.com

Cyber Security for Cyber Physical Systems 1st Edition

Saqib Ali

https://textbookfull.com/product/cyber-security-for-cyber-physical-
systems-1st-edition-saqib-ali/

textboxfull.com

Cross Layer Design for Secure and Resilient Cyber Physical

Systems A Decision and Game Theoretic Approach Quanyan Zhu

https://textbookfull.com/product/cross-layer-design-for-secure-and-
resilient-cyber-physical-systems-a-decision-and-game-theoretic-
approach-quanyan-zhu/
textboxfull.com
Relay Tuning of PID Controllers For Unstable MIMO
Processes 1st Edition M. Chidambaram

https://textbookfull.com/product/relay-tuning-of-pid-controllers-for-
unstable-mimo-processes-1st-edition-m-chidambaram/

textboxfull.com

Cooperative Spectrum Sensing and Resource Allocation

Strategies in Cognitive Radio Networks Xavier Fernando

https://textbookfull.com/product/cooperative-spectrum-sensing-and-
resource-allocation-strategies-in-cognitive-radio-networks-xavier-
fernando/
textboxfull.com

Security and Privacy for Next-Generation Wireless Networks

Sheng Zhong

https://textbookfull.com/product/security-and-privacy-for-next-
generation-wireless-networks-sheng-zhong/

textboxfull.com

Multi Variable Calculus 1st Edition Yunzhi Zou

https://textbookfull.com/product/multi-variable-calculus-1st-edition-
yunzhi-zou/

textboxfull.com

Confrontational and Cooperative Regional Orders Managing

Regional Security in World Politics 1st Edition Ariel
Gonzalez Levaggi
https://textbookfull.com/product/confrontational-and-cooperative-
regional-orders-managing-regional-security-in-world-politics-1st-
edition-ariel-gonzalez-levaggi/
textboxfull.com
Wireless Networks

Yulong Zou
Jia Zhu

Physical-Layer
Security for
Cooperative
Relay Networks
Wireless Networks

Series Editor:
Xuemin (Sherman) Shen
University of Waterloo, Waterloo, Ontario, Canada

More information about this series at http://www.springer.com/series/14180

Yulong Zou • Jia Zhu

Physical-Layer Security for

Cooperative Relay Networks

123
Yulong Zou Jia Zhu
Nanjing University of Posts Nanjing University of Posts
and Telecommunications and Telecommunications
Nanjing, Jiangsu, China Nanjing, Jiangsu, China

ISSN 2366-1186 ISSN 2366-1445 (electronic)

Wireless Networks
ISBN 978-3-319-31173-9 ISBN 978-3-319-31174-6 (eBook)
DOI 10.1007/978-3-319-31174-6

Library of Congress Control Number: 2016936582

© Springer International Publishing Switzerland 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, express or implied, with respect to the material contained herein or for any
errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG Switzerland
Preface

Recently, cybercriminal activities in wireless communications systems are growing

due to the fact that more and more emerging malware programs (also known as
computer viruses) are targeted on the mobile terminals. Accordingly, an increasing
attention has been paid to the research of wireless security against various malicious
attacks. The radio propagation inherits the broadcast nature, leading to any receivers
within the coverage area of a radio transmitter being capable of overhearing the
wireless transmission. This makes the wireless communication systems extremely
vulnerable to the eavesdropping attack. Typically, cryptographic techniques relying
on secret keys are employed for preventing an eavesdropper from interpreting the
wireless transmissions.
However, classic cryptographic techniques including both public-key cryptogra-
phy and symmetric-key cryptography are only computationally secure and rely upon
the hardness of their underlying mathematical problems. Cryptography security
would be significantly compromised if an efficient method of solving its underlying
mathematical problem was to be discovered. Moreover, conventional secret key
exchange relies on a trusted key management center, which may not be always
applicable in wireless networks. To this end, physical-layer security is emerging as a
promising paradigm to secure wireless communications by exploiting the physical-
layer characteristics of wireless channels. It is proved from an information-theoretic
perspective that perfect secrecy can be achieved if the wiretap channel from a source
to an eavesdropper is a degraded version of the main channel from the source to its
intended destination. However, due to the time-varying fading effect of wireless
channels, the main channel may experience a deep fade, which makes the perfect
secrecy become impossible in some cases.
This book presents the concept and practical challenges of physical-layer security
as well as examines recent advances in cooperative relaying for the wireless
physical-layer security. In Chap. 1, we first review a range of physical-layer security
techniques, namely, information-theoretic security, artificial-noise-aided security,
security-oriented beamforming, and diversity-assisted security, along with an in-
depth discussion of cooperative relaying techniques for wireless networks. Next,
Chap. 2 investigates the physical-layer security for a wireless network consisting of

v
vi Preface

a source and a destination with the aid of multiple relays, where only the single
“best” relay is selected among the multiple relays to assist the source-destination
transmission against eavesdropping. In Chap. 3, we then examine joint relay and
jammer selection for enhancing the wireless physical-layer security of the source-
destination transmission with the help of multiple intermediate nodes in the presence
of an eavesdropper. In the joint relay and jammer selection, an intermediate node
is selected to act as the relay for assisting the source-destination transmission and
another intermediate node is chosen to act as the jammer for interfering with the
eavesdropper.
Additionally, Chap. 4 explores the security-reliability tradeoff (SRT) for a wire-
less network, where security and reliability are measured by using the intercept
probability experienced by the eavesdropper and outage probability encountered
at the legitimate destination, respectively. We present two relay selection schemes
for the SRT improvement, namely, single-relay selection (SRS) and multi-relay
selection (MRS). To be specific, in the SRS scheme, only the single “best” relay is
selected to assist the source-destination transmission, whereas in the MRS scheme,
multiple relays are invoked to participate in forwarding the source signal to the
destination. Finally, Chap. 5 re-examines the joint relay and jammer selection
from the SRT perspective, where a relay is used to help the source transmission
enhance wireless reliability and a friendly jammer is adopted to improve wireless
security through the emission of the artificial noise for confusing the eavesdropper.
It is shown that with an increasing number of relays and jammers, the security
and reliability of wireless communications relying on the joint relay and jammer
selection can be significantly enhanced concurrently.

Nanjing, China Yulong Zou

January 2016 Jia Zhu
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Wireless Physical-Layer Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Overview of Cooperative Relay Networks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Objective of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2 Relay Selection for Enhancing Wireless Secrecy Against
Eavesdropping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Relay Selection for Secrecy Enhancement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.1 Direct Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.2 Random Relay Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.3 Proposed Relay Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3 Asymptotic Intercept Probability and Secrecy Diversity . . . . . . . . . . . . . . 25
2.4 Numerical Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3 Joint Relay and Jammer Selection
for Wireless Physical-Layer Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Joint Relay and Jammer Selection for Physical-Layer Security . . . . . . 39
3.2.1 Pure Relay Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2.2 Pure Jammer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2.3 Joint Relay and Jammer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.3 Numerical Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

vii
viii Contents

4 Security-Reliability Tradeoff for Cooperative Relay Networks. . . . . . . . . 53

4.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2 Security-Reliability Tradeoff for Relay Selection . . . . . . . . . . . . . . . . . . . . . 57
4.2.1 Direct Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.2.2 Single-Relay Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2.3 Multi-relay Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3 Numerical Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5 Improving Security-Reliability Tradeoff Through Joint
Relay and Jammer Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2 Security-Reliability Tradeoff for Joint Relay and Jammer Selection . 77
5.2.1 Pure Relay Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.2 Pure Jammer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.2.3 Joint Relay and Jammer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.3 Numerical Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.1 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Acronyms

3G Third generation
AF Amplify-and-forward
AP Access point
AWGN Additive white Gaussian noise
BS Base station
CDF Cumulative distribution function
CF Compress-and-forward
CRC Cyclic redundancy check
CSI Channel state information
DF Decode-and-forward
DoS Denial-of-service
GSVD Generalized singular value decomposition
I.I.D Independent identically distributed
LoS Line-of-sight
LTE Long-term evolution
LTE-A Long-term evolution advanced
MAC Medium access control
MER Main-to-eavesdropper ratio
MIMO Multiple-input multiple-output
MISOME Multiple-input single-output multiple-eavesdropper
MRS Multi-relay selection
OFDMA Orthogonal frequency-division multiple access
PDF Probability density function
SNR Signal-to-noise ratio
SRS Single-relay selection
SRT Security-reliability tradeoff
TDMA Time-division multiple access

ix
Chapter 1
Introduction

Abstract Due to the broadcast nature of radio propagation, wireless transmissions

are accessible to any eavesdroppers and thus become extremely vulnerable to the
eavesdropping attack. Physical-layer security is emerging as a promising paradigm
to achieve the information-theoretic secrecy for wireless networks. This chapter
introduces the current research on physical-layer security for wireless networks. We
first discuss a range of physical-layer security techniques, including the information-
theoretic security, artificial noise aided security, security-oriented beamforming,
and diversity assisted security approaches. Then, we present an overview on
the cooperative relaying methods for wireless networks, namely the orthogonal
relaying, non-orthogonal relaying and relay selection. Additionally, the application
of cooperative relaying to wireless physical-layer security is also discussed for
protecting the wireless communications against eavesdropping.

1.1 Wireless Physical-Layer Security

Recent years have witnessed the widespread use of smartphones for accessing
various wireless networks, such as the third-generation (3G), long-term evolution
(LTE) and LET-advanced (LTE-A) mobile communications systems as well as
the Wi-Fi [1]. Meanwhile, it has been reported that an increasing number of
wireless terminals are compromised by the adversary for carrying out cybercriminal
activities, including malicious hacking, data forging, financial information theft,
and so on. Moreover, as discussed in [2] and [3], the broadcast nature of radio
propagation makes the wireless communication systems extremely vulnerable to
the eavesdropping attack. As shown in Fig. 1.1, an access point (AP) is considered
to transmit data packets to its associated legitimate users, which can be readily
overheard by an eavesdropper as long as it lies in the coverage area of AP.
Traditionally, cryptographic techniques relying on secret keys were adopted for
preventing an eavesdropper from interpreting the wireless data transmissions. There
are two main types of cryptographic techniques, namely the public-key cryptog-
raphy and symmetric-key cryptography, which are however only computationally
secure and rely upon the hardness of their underlying mathematical problems [4, 5].
The security of a cryptographic approach would be significantly compromised, if
an efficient method of solving its underlying hard mathematical problem was to be

© Springer International Publishing Switzerland 2016 1

Y. Zou, J. Zhu, Physical-Layer Security for Cooperative Relay Networks,
Wireless Networks, DOI 10.1007/978-3-319-31174-6_1
2 1 Introduction

Fig. 1.1 A wireless network

consisting of an access point
(AP) and multiple legitimate
users in the face of an
eavesdropper

Fig. 1.2 Radio wave

multipath propagation in an
indoor environment

discovered [6]. Moreover, the conventional secret key exchange requires a trusted
key management center, which may not be always applicable in wireless networks
(e.g., wireless ad hoc networks).
As a consequence, physical-layer security emerges as an effective means to
secure the wireless communications by exploiting the physical-layer characteristics
of wireless channels [7–13]. It was proved in [7] that if the wiretap channel (from
source to eavesdropper) is a degraded version of the main channel (from source to
destination), the prefect secrecy can be achieved in an information-theoretic sense.
In [8], the concept of secrecy capacity was further introduced, which is shown as the
difference between the capacity of the main channel and that of the wiretap channel.
More specifically, a positive secrecy capacity implies that the perfect secrecy is
achievable and vice versa. However, due to the time-varying fading effect of wireless
channels, the secrecy capacity of wireless communications is severely degraded,
especially when a deep fade is encountered in the main channel between the source
and destination.
As shown in Fig. 1.2, when a radio wave is emitted from the transmitter, it would
be propagated over multiple different paths due to the radio reflection and scattering
experienced by the obstacles as well as by the room ceiling and floor for indoor
environments. This leads to the fact that multiple radiowave components arrive at the
1.1 Wireless Physical-Layer Security 3

receiver, resulting in multipath reception. Since the differently delayed components

sometimes add destructively, sometimes constructively, the strength of the radio
signal combined at the receiver attenuates and fluctuates in time. The time-variant
attenuation of multipath radio propagation is referred to as the wireless fading,
which is modeled as a random process. There are three proper statistical models,
namely the Rayleigh fading, Rician fading and Nakagami fading, which may be
used for characterizing the wireless fading of radio propagation.
More specifically, Rayleigh fading is often used for modeling the magnitude
attenuation of a radio wave signal propagated through a wireless medium (e.g., free
space). This fading model is most applicable, when the propagation environment is
comprised of massive obstacles. In such an environment, a radio wave signal would
be heavily scattered and reflected, resulting in a large number of signal components
arriving at the receiver. According to the central limit theorem, if the number of
received radiowave components is sufficiently high, the combined signal containing
the sine and cosine parts can be modeled as a complex-valued Gaussian random
variable, regardless of the distribution of the individual components. The medium,
which conveys the radio signal propagating through a wireless channel from the
transmitter to receiver, is termed the channel response. To be specific, the real and
imaginary parts of the channel response are modeled by independent and identically
distributed (i.i.d) zero-mean Gaussian random variables. Hence, the amplitude of the
channel response, denoted by x, is Rayleigh distributed with a probability density
function (PDF) given by
 2
2x x
p.x/ D exp  2 ;x  0 (1.1)
2 

where  2 represents an expected value of x2 and can be viewed as the average

power of the received radio signal. Again, the Rayleigh fading is very suitable for
a propagation scenario in the presence of many scatterers to attenuate, reflect and
diffract the radio signal.
Rician fading is another stochastic model commonly used for characterizing the
multipath transmission when there is a line-of-sight (LOS) propagation path that
has a higher signal strength than the other non-line-of-sight paths. By contrast,
the above-mentioned Rayleigh fading is suitable when there is no LOS path. As
mentioned above, in Rayleigh fading, the channel responses are assumed to be zero-
mean Gaussian random variables. However, if there is a dominant LOS propagation
path, then the mean of the random channel response should be around the power
level of the dominant path, which is no longer zero. In such a situation containing
a dominant LOS path, Rician fading is more appropriate than Rayleigh fading in
modeling the transmission medium. In Rician fading, the channel amplitude x is
modeled by a Rician distribution, which is expressed as
  r !
2.K C 1/x .K C 1/x2 K.K C 1/
p.x/ D exp K  I0 2 x ;x  0 (1.2)
2 2 2
4 1 Introduction

where K is the ratio between the signal power of the direct LOS path and that of
the other scattered paths,  2 is the average power of the total propagation paths, and
I0 ./ is the first kind zero-order modified Bessel function. Additionally, when there
is no direct LOS path, this means K D 0. It can be shown that substituting K D 0
into (1.2) readily leads to (1.1), implying that the Rayleigh distribution is a special
case of the Rician distribution.
As discussed above, the multipath radio propagation is a random fading process,
implying that a radio signal propagating over a wireless channel may be sometimes
attenuated severely, sometimes only slightly. To this end, multiple antennas can be
employed at a receiver so that multiple independently faded copies of the radio
signal are received and combined together for the sake of enhancing the wireless
reception quality. In this case, if the wireless channels between the transmitter and
the receive antennas are modeled as Rayleigh fading, the amplitude of the combined
signal containing multiple Rayleigh-faded components obey the Nakagami distribu-
tion associated with a shape factor m, where m is the number of receive antennas.
Hence, Nakagami fading is used to model the amplitude of the sum of multiple i.i.d.
Rayleigh-faded signals. The PDF of Nakagami distribution can be written as

2mm  m 
2m1
p.x/ D x exp  2 x2 ; x  0 (1.3)
 .m/ 2m 

where  .m/ represents the gamma function and  2 is the expectation of x2 . It has
to be pointed out that when there is only a single receive antenna (i.e., m D 1), the
Nakagami fading becomes the same as the Rayleigh fading, which can be readily
validated by substituting m D 1 into (1.3) to obtain (1.1). The aforementioned
three stochastic models, namely the Rayleigh fading, Rician fading and Nakagami
fading, are generally utilized for characterizing the temporal behavior of a wireless
propagation channel.
The wireless channel fading is assumed to remain constant during the coherence
time by definition. However, in consecutive coherence time intervals, the wireless
fading varies considerably and is modeled as a stochastic process. By comparing the
coherence time of a wireless channel and the symbol period of a radio signal to be
transmitted over the channel, the fading imposed by the channel may be classified as
slow fading and fast fading. More specifically, slow fading is encountered when the
coherence time is higher than the symbol period, where again, the fading amplitude
and phase are considered to be constant over the symbol period. By contrast, fast
fading implies that the coherence time is shorter than the symbol period, where the
wireless fading varies during the symbol period.
Additionally, in a wireless channel, the coherence bandwidth represents the
range of frequencies, for which a radio signal transmitted over the channel
will experience comparable fading attenuation across the different frequencies. If
the coherence bandwidth of the channel is higher than the frequency bandwidth
of the radio signal, then all frequency components of the signal are deemed to
experience more or less the same fading effect, which is referred to as flat fading. On
the other hand, when the coherence bandwidth becomes lower than the bandwidth
1.1 Wireless Physical-Layer Security 5

of the signal, different frequency components of the signal will experience different
fading effects. This is referred to as frequency-selective fading. Therefore, the radio
propagation may also be divided into frequency-flat and frequency-selective fading.
In order to combat the fading effect for enhancing the wireless secrecy capacity,
considerable research efforts were devoted to the development of various physical-
layer security techniques, which can be classified into the following main research
categories: (1) the information-theoretic security [10–16], (2) artificial noise aided
security [17–21], (3) security-oriented beamforming [22–26], and (4) diversity
assisted security approaches [27–31].
Information-theoretic security is to examine the fundamental limits of physical-
layer security from information-theoretic perspective. Historically, the information-
theoretic security was first studied by Shannon in [10], with an emphasis on the
mathematical properties of communications secrecy systems. More specifically, a
secrecy system was achieved by mathematically transforming the plaintext into
the cryptograms with the aid of secret keys, where the transformation shall be
nonsingular so that unique deciphering is possible given the secret key. In [10], the
theoretical secrecy was developed to address the communications security against an
eavesdropper either with infinite or finite computing power. It was shown in [10] that
a perfect secrecy system is achievable with a finite secret key, where an eavesdropper
is unable to obtain a unique deciphering solution. In [7], Wyner investigated the
information-theoretic security for a discrete memoryless wiretap channel consisting
of a source, a destination and an eavesdropper. It was proved in [11] that when the
main channel has a better condition than the wiretap channel, there exists a positive
rate at which the source and destination can exchange information at perfect secrecy.
In [8], Wyner’s results were extended to the Gaussian wiretap channel, where the
secrecy capacity is shown as the difference between the capacity of the main channel
and that of the wiretap channel.
For the sake of improving the wireless secrecy in fading environments, the
multiple-input multiple-output (MIMO) was studied extensively to combat the
wireless fading and increase the secrecy capacity. In [14], Khisti and Wornell
investigated a multiple-input single-output multiple-eavesdropper (MISOME) sce-
nario as shown in Fig. 1.3, where the source and eavesdropper are assumed with
multiple antennas and the destination has only one antenna. By assuming that the
eavesdropper’s channel state information (CSI) is known, the secrecy capacity of
the MISOME was characterized in terms of the generalized eigenvalues. When
the CSI knowledge of the eavesdropper becomes unavailable, Khisti and Wornell
further presented a masked beamforming scheme and showed that it can obtain a
near-optimal security performance in high signal-to-noise ratio (SNR) regions. In
[15], the authors examined the information-theoretic security from the MISOME
case to a more general MIMO scenario compromised of the source, destination and
eavesdropper each equipped with multiple antennas. They considered two specific
cases: (1) the deterministic case in which the CSIs of both the man channel and
wiretap channel are fixed and known; and (2) the fading case in which both the
main channel and wiretap channel experience Rayleigh and only the statistical CSI
of the wiretap channel is assumed. The generalized-singular-value-decomposition
6 1 Introduction

Fig. 1.3 A multiple-input

single-output
multiple-eavesdropper
(MISOME) communications
system

(GSVD) approach was utilized to achieve the secrecy capacity for the deterministic
case, which was then extended to the fading case, showing that the corresponding
secrecy capacity approaches to zero, if the ratio of the number of eavesdropper’s
antennas to that of the source’s antennas is larger than two.
The artificial noise aided security [17–21] enables the source to generate an
interfering signal (referred to as artificial noise) in a manner such that the artificial
noise only interferes with the eavesdropper, while the legitimate destination keeps
unaffected. This results in a decrease of the capacity of the wiretap channel without
affecting the capacity of the main channel, thus leading to an increase of the
secrecy capacity. In [17], the authors considered a wireless network comprised
of a source and a destination in the face of an eavesdropper and investigated the
use of the artificial noise for defending the source-destination transmission against
eavesdropping. More specifically, a certain transmit power at the source is allocated
to generate the artificial noise, which is sophisticatedly designed such that only
the wiretap channel is degraded without affecting the main channel. To this end,
the authors of [17] proposed the employment of multiple antennas to generate the
artificial noise, and proved that the artificial noise would not degrade the main
channel only when the number of transmit antennas at the source is more than the
number of receive antennas at the eavesdropper. It was shown in [17] that a positive
secrecy capacity can always be achieved by using the artificial noise, even if the
wiretap channel is better than the main channel. Although the artificial noise aided
security is able to guarantee the secrecy of wireless communications, it comes at the
expense of additional power consumption since a certain transmit power is needed
for producing the artificial noise. In [18], the authors studied the power allocation
between the information-bearing signal (that carries the desired information from
source to destination) and the artificial noise (that is employed to interfere with the
eavesdropper). It was shown in [18] that the simple equal power allocation is a near-
optimal strategy when the eavesdroppers are uncoordinated and independent of each
other. In addition, when the CSI estimation error is considered, it was observed that
using more power for the artificial noise can achieve better secrecy performance
than increasing the transmit power for the information-bearing signal.
1.1 Wireless Physical-Layer Security 7

The security-oriented beamforming [22–26] allows the source to transmit its

signal in a particular direction to the destination such that the signal arrived at
an eavesdropper experiences destructive interference and becomes much weaker
than that received at the destination experiencing constructive interference. In this
way, the capacity of the main channel from the source to destination relying on
the security-oriented beamforming would be much higher than that of the wiretap
channel from the source to eavesdropper, leading to the enhancement of secrecy
capacity. In [22], the authors examined the use of relay nodes to form a beamforming
system with the perfect CSI knowledge of the main channel and wiretap channel.
The relay beamforming design was performed to maximize the secrecy rate under
the total transmit power constraint, which was addressed by using the well-known
semi-definite programming technique. It was shown in [22] that the proposed
beamforming method significantly increases the secrecy capacity of wireless com-
munications. In [23], multiple antennas were utilized for the beamforming design to
improve the secrecy capacity of the source-destination transmission in the presence
of an eavesdropper. Differing from the beamforming design of [22] relying on the
perfect CSI knowledge of the wiretap channel, Mukherjee et al. [23] carried out
the beamforming design without knowing the eavesdropper’s CSI knowledge. In
addition, the authors of [24] investigated the transmit beamforming in a wireless
relay network consisting of a source, a destination and an untrusted relay, where
the relay may potentially assist the signal transmission from source to destination.
But, the relay is untrusted in the sense that it may become a passive eavesdropping
attack. Two secure beamforming schemes, namely the noncooperative beamforming
and cooperative beamforming, were proposed to maximize the wireless secrecy
capacity. Simulation results showed that both the noncooperative and cooperative
beamforming schemes performs better than the conventional approaches in terms of
the secrecy capacity.
As an alternative, diversity techniques [27–31] can also be utilized to enhance
the wireless security against eavesdropping attack, called the diversity assisted
security. For example, considering that the source has multiple transmit antennas, an
optimal antenna can be selected and used to transmit the desired signal depending
on whether the CSI of the main channel and wiretap channel is available [27]. To
be specific, if the CSI of both the main channel and wiretap channel is known
at the source, a transmit antenna with the highest instantaneous secrecy capacity
can be chosen to transmit the desired signal, which can significantly improve
the secrecy capacity of wireless transmission. If only the main channel’s CSI is
available, we can choose a transmit antenna with the highest instantaneous capacity
of the main channel to send the desired signal. Since the transmit antenna selection
is based on the main channel’s CSI only and the wiretap channel is typically
independent of the main channel, the capacity of the main channel will be increased
with the transmit antenna selection, but no capacity improvement can be achieved
for the wiretap channel. This finally results in an increase of the secrecy capacity
through the transmit antenna selection.
In addition, the multiuser diversity [30, 31] is another effective means of
improving the wireless physical-layer security against eavesdropping. As shown
8 1 Introduction

Fig. 1.4 A cellular network

consisting of a base station
(BS) and multiple users in the
presence of an eavesdropper

in Fig. 1.4, we consider a cellular network comprised of a base station (BS)

and multiple associated users. Typically, a multiple access approach, such as the
time-division multiple access (TDMA) and orthogonal frequency-division multiple
access (OFDMA) is employed to enable the multiple users to communicate with
the BS. Taking the TDMA as an example, we should first perform the multiuser
scheduling to determine which user is scheduled to access a given time slot for
data transmissions. In order to effectively protect the wireless transmissions against
eavesdropping, the multiuser scheduling should be aimed at minimizing the capacity
of the wiretap channel, while maximizing the capacity of the main channel. If the
CSIs of both the main channel and the wiretap channel are known, a specific user
with the highest instantaneous secrecy capacity would be chosen to access the given
time slot. If only the main channel’s CSI is known, then a user that maximizes the
capacity of the main channel is selected for accessing the slot, which does not need
the wiretap channel’s CSI.

1.2 Overview of Cooperative Relay Networks

This section presents an overview of cooperative relay techniques for wireless

networks. In general, there are two different types of relays, namely the full-duplex
relay and the half-duplex relay [32, 33]. More specifically, the full-duplex relay
implies that a relay can transmit and receive radio signals simultaneously over the
same channel. By contrast, with the half-duplex relay, two orthogonal channels
are needed for the relay to transmit and receive the radio signals. It is obvious
that the full-duplex relay saves the spectrum resource and doubles the spectrum
utilization, as compared to the half-duplex relay. However, with the full-duplex
relay, an incoming signal received at an antenna will be interfered with its outgoing
signal, which is known self-interference. It is challenging to cancel out such a self-
interference due to the significant difference in the power levels of incoming and
outgoing signals [34]. Thus, although the half-duplex relay has a lower spectrum
utilization than the full-duplex relay, it is still used in practical wireless systems
due to its simplicity in implementation. To alleviate the loss of spectrum utilization,
1.2 Overview of Cooperative Relay Networks 9

the two-way relay was examined by employing the physical-layer network coding
[35, 36], where the exchange of two messages can be completed between two source
nodes via the relay over two orthogonal channels.
As shown in Fig. 1.5, when a relay node assists the message transmission from
a source to a destination, it typically needs two phases [37–39]: (1) multicast
phase, where the source multicasts its signal to the relay and destination, and (2)
retransmission phase, where the relay forwards its received signal to the destination
by using a certain relaying protocol, i.e., the amplify-and-forward (AF) protocol
and decode-and-forward (DF) protocol. Figure 1.6 shows a comparison between the
two AF and DF protocols. It can be seen from Fig. 1.6 that with the AF protocol,
the relay node just simply retransmits a scaled version of its received noisy signal
to the destination. By contrast, the DF protocol enables the relay first to decode its
received signal and then to forward its decoded result to the destination, as shown
in Fig. 1.6.
More specifically, the AF is a simple relaying protocol, which allows a relay
node to amplify and retransmit its received noisy version of the source signal

Fig. 1.5 A wireless network

consisting of a source, a relay
and a destination

Fig. 1.6 Comparison

between the amplify-
and-forward (AF) and
decode-and-forward (DF)
protocols: (a) AF relaying
and (b) DF relaying
10 1 Introduction

to the destination. On the one hand, the AF protocol has the advantage of
simple implementation, since the AF relay just forwards its received noisy signal
without relying on any kind of decoding operations. On the other hand, the main
disadvantage of the AF protocol is that the noise received at the relay node may also
be amplified and forwarded to the destination. This would result in a performance
degradation at the destination in decoding the source signal. Considering that the
source transmits its signal denoted by x at a power of P, we can express the received
signal at the relay as
p
yr D Phsr x C nr ; (1.4)

where hsr denotes the fading coefficient of the channel spanning from the source to
the relay and nr represents the zero-mean additive white Gaussian noise (AWGN)
with a variance of N0 .pNext, the AF relay amplifies its received signal yr with a
scaling factor ˛ D 1=. Pjhsr j/ and retransmits the scaled signal to the destination.
Thus, the received signal at the destination can be written as
p
d D
yAF P˛hrd yr C nd ; (1.5)

where hrd denotes the fading coefficient of the channel spanning from the relay
to the destination and nd represents the zero-mean AWGN with a variance of N0 .
From (1.5), we can obtain the channel capacity of the source-relay-destination
transmission relying on the AF protocol as

1 jhsr j2 jhrd j2
AF
Csrd D log2 .1 C  /; (1.6)
2 jhrd j2 C jhsr j2

where  D P=N0 and the factor 12 arises from the fact that two time slots are needed
for completing the transmission of one message from the source to the destination
AF
via the relay. It can be observed from (1.6) that the channel capacity Csrd is affected
by both the source-relay channel and the relay-destination channel.
Differing from the AF relay, the DF relay needs to decode its received signal
transmitted from the source and then forwards its decoded result to the destination
node. Following [37], we can obtain the channel capacity of the source-relay-
destination transmission relying on the AF protocol as

1
DF
Csrd D log Œ1 C  min.jhsr j2 ; jhrd j2 /; (1.7)
2 2
which shows that the channel capacity of the DF protocol is the minimum between
the capacity of the source-relay channel and that of the relay-destination channel.
Comparing (1.6) and (1.7), one can easily conclude that the channel capacity of the
DF protocol is higher than that of the AF protocol, showing the advantage of the
DF over AF in terms of its capacity performance [40]. It is worth mentioning that
the capacity improvement achieved by the DF over AF comes at the expense of
1.2 Overview of Cooperative Relay Networks 11

Fig. 1.7 A source transmits

to its destination with the aid
of multiple relays

communications latency, since the decoding operations required by the DF protocol

incur an additional time delay, whereas the AF just simply forwards its received
signal without any sort of decoding.
So far, we have discussed two basic relaying protocols, namely the AF and
DF for a single-relay wireless network. As shown in Fig. 1.7, multiple relays
may become available to assist the transmission from the source to destination.
Given multiple relay nodes available, it is of particular interest to explore how
the relays can be utilized in an efficient manner so that a significant performance
enhancement is achieved while the corresponding cost is minimized. When multiple
relays become available, we may allow all the relays to forward their received
signals from the source to the destination in two different ways, namely the
orthogonal relaying and non-orthogonal relaying [41–43]. More specifically, in the
orthogonal relaying process, all the relays retransmit their received source signals
over mutually orthogonal channels. This can effectively avoid the signal interference
among different relays, but scarifies the precious channel resources, since an
increasing orthogonal channels are required as the number of relays increases.
By contrast, the non-orthogonal relaying enables the multiple relays to transmit
simultaneously over the same channel, which significantly saves the orthogonal
channel resources. However, it requires complex symbol-level synchronization
among multiple spatially distributed relays so that the different signals transmitted
by different relays can be synchronized and combined at the destination without
causing interference. Moreover, as the number of relays increases, the complexity
of such symbol-level synchronization becomes extremely high.
Alternatively, the relay selection was also studied to address the multiple-relay
scenario, where only the single “best” relay is chosen to participate in forwarding the
source signal to the destination [44–46]. To be specific, the source first broadcasts its
message to all relay nodes, which attempt to decode the source message from their
received signals, if the DF protocol was used at the relays. Let these relays which
successfully decode the source message constitute a so-called decoding set. It needs
to be pointed out that the cyclic redundancy check (CRC) coding may be invoked
for determining whether a relay node succeeds in decoding its received signal or
not. Particularly, if a relay nodes passes the CRC checking, it is considered to be
successful in decoding and added to the decoding set. If the decoding set is not
empty, a single relay would be chosen among the decoding set for forwarding its
12 1 Introduction

decoded result to the destination. Besides, if the AF protocol was considered at the
relays, we may select a relay that maximizes the received SNR yAF d for assisting the
source-destination transmission. It can be observed that the relay selection avoids
the complex synchronization issue and only requires two orthogonal channels (i.e.,
the source-to-relay channel and the best-relay-to-destination channel) regardless
of the number of relays. More importantly, the relay selection can achieve the
same diversity gain as the aforementioned non-orthogonal relaying and orthogonal
relaying approaches.
Besides, the relay selection can also be employed to enhance the physical-layer
security of wireless communications in the presence of an eavesdropper [47, 48].
More specifically, considering a wireless network consisting of a source and a
destination with the help of multiple relay nodes, we can choose the “best” relay
to assist the source-destination transmission against eavesdropping. To elaborate a
litter further, if the CSIs of both the main channel and wiretap channel are available,
a relay node with the highest instantaneous secrecy capacity may be invoked to
forward the source signal to the legitimate destination. If the eavesdropper’s CSI
is unknown and only the main channel’s CSI is available, then a relay node with
the highest instantaneous capacity of the main channel is selected to assist the
source transmission to the destination. It was shown in [47] that the wireless
secrecy capacity can be significantly improved relying on the relay selection,
which explicitly demonstrates the physical-layer security benefits of exploiting the
cooperative relays against eavesdropping.

1.3 Objective of This Book

Cryptographic techniques relying on secret keys are typically employed in order

to guarantee the transmission confidentiality against eavesdropping, which are only
computationally secure and depend on the hardness of their underlying mathemat-
ical problems. Moreover, the key distribution relies upon a trusted infrastructure,
which may be unavailable and even compromised in some cases. To this end,
the emerging physical-layer security exploits physical characteristics of wireless
channels for achieving the perfect secrecy against eavesdropping. However, the
physical-layer security of wireless transmissions is severely limited and degraded
by the time-varying multipath fading.
The objective of this book is to present the concept and practical challenges of
physical-layer security as well as examine recent advances in cooperative relaying
designs for the wireless security enhancement. The motivations and concepts of
physical-layer security are first explored along with a review of the wireless
security threats in cooperative relay networks. Then, the relay-selection designs
are discussed in details for improving wireless secrecy against eavesdropping in
time-varying fading environments. Finally, the security-reliability tradeoff (SRT) is
mathematically characterized for wireless communications. Also, the relay selection
as well as the joint relay and jammer selection are examined for the wireless SRT
improvement.
References 13

References

1. O. Aliu, A. Imran, M. Imran, and B. Evans, “A Survey of self organisation in future cellular
networks,” IEEE Commun. Surveys & Tutorials, vol. 15, no. 1, pp. 336–361, Feb. 2013.
2. A. Mukherjee, S. A. Fakoorian, J. Huang, and A. L. Swindlehurst, “Principles of physical
layer security in multiuser wireless networks: A survey,” IEEE Commun. Surveys & Tutorials,
vol. 16, no. 3, pp. 1550–1573, Sept. 2014.
3. Y. Zou, J. Zhu, L. Yang, Y.-.C. Liang, and Y.-D. Yao, “Securing physical-layer communications
for cognitive radio networks,” IEEE Commun. Mag., vol. 53, no. 9, pp. 48–54, Sept. 2015.
4. S. Mathur, W. Trappe, N. Mandayam, C. Ye, and A. Reznik, “Radio-telepathy: Extracting
a secret key from an unauthenticated wireless channel,” in Proceedings of The 14th Annual
International Conference on Mobile Computing and Networking (MobiCom 2008), California,
USA, September 2008.
5. S. Jana, et al., “On the effectiveness of secret key extraction from wireless signal strength in
real environments,” in Proceedings of The 15th Annual International Conference on Mobile
Computing and Networking (MobiCom 2009), Beijing, China, September 2009.
6. Q. Wang, K. Xu, and K. Ren, “Cooperative secret key generation from phase estimation in
narrowband fading channels,” IEEE J. Sel. Areas Commun., vol. 30, no. 9, pp. 1666–1674,
Sept. 2012.
7. A. D. Wyner, “The wire-tap channel,” Bell Syst. Tech. J., vol. 54, no. 8, pp. 1355–1387, 1975.
8. S. K. Leung-Yan-Cheong and M. E. Hellman, “The Gaussian wiretap channel,” IEEE Trans.
Inf. Theory, vol. 24, no. 7, pp. 451–456, Jul. 1978.
9. Y. Zou, X. Li, and Y.-C. Liang, “Secrecy outage and diversity analysis of cognitive radio
systems,” IEEE J. Sel. Areas Commun., vol. 32, no. 11, pp. 2222–2236, Nov. 2014.
10. C. E. Shannon, “Communications theory of secrecy systems,” Bell Syst. Tech. J., vol. 28,
pp. 656–715, 1949.
11. H. Mahdavifar and A. Vardy, “Achieving the secrecy capacity of wiretap channels using polar
codes,” IEEE Trans. Inf. Theory, vol. 57, no. 10, pp. 6428–6443, Oct. 2011.
12. P. K. Gopala, L. Lai, and H. E. Gamal, “On the secrecy capacity of fading channels,” IEEE
Trans. Inf. Theory, vol. 54, no. 10, pp. 4687–4698, Oct. 2008.
13. Y. Liang, H. V. Poor, and S. Shamai, “Secure communication over fading channels,” IEEE
Trans. Inf. Theory, vol. 54, no. 6, pp. 2470–2492, Jun. 2008.
14. A. Khisti and G. W. Wornell, “Secure transmission with multiple antennas: The MISOME
wiretap channel,” IEEE Trans. Inf. Theory, vol. 56, no. 7, pp. 3088–3104, Jul. 2010.
15. F. Oggier and B. Hassibi, “The secrecy capacity of the MIMO wiretap channel,” IEEE Trans.
Inf. Theory, vol. 57, no. 8, pp. 4961–4972, Aug. 2011.
16. S. Shafiee, N. Liu, and S. Ulukus, “Towards the secrecy capacity of the Gaussian MIMO wire-
tap channel: The 2-2-1 channel,” IEEE Trans. Inf. Theory, vol. 55, no. 9, pp. 4033–4039, Sept.
2009.
17. S. Goel and R. Negi, “Guaranteeing secrecy using artificial noise,” IEEE Trans. Wirel.
Commun., vol. 7, no. 6, pp. 2180–2189, Jul. 2008.
18. X. Zhou and M. McKay, “Secure transmission with artificial noise over fading channels:
Achievable rate and optimal power allocation,” IEEE Trans. Veh. Tech., vol. 59, no. 8,
pp. 3831–3842, Aug. 2010.
19. D. Goeckel, et al., “Artificial noise generation from cooperative relays for everlasting secrecy
in two-hop wireless networks,” IEEE J. Sel. Areas Commun., vol. 29, no. 10, pp. 2067–2076,
Oct. 2011.
20. S.-C. Lin, et al., “On the impact of quantized channel feedback in guaranteeing secrecy with
artificial noise: The noise leakage problem,” IEEE Trans. Wirel. Commun., vol. 10, no. 3,
pp. 901–915, Mar. 2011.
21. Q. Li and W.-K. Ma, “Spatially selective artificial-noise aided transmit optimization for
MISO multi-eves secrecy rate maximization,” IEEE Trans. Signal Process., vol. 61, no. 10,
pp. 2704–2717, Oct. 2013.
14 1 Introduction

22. J. Zhang and M. Gursoy, “Collaborative relay beamforming for secrecy,” in Proc. The 2010
IEEE Intern. Conf. Commun., Cape Town, South Africa, May 2010.
23. A. Mukherjee amd A. Swindlehurst, “Robust beamforming for security in MIMO wiretap
channels with imperfect CSI,” IEEE Trans. Signal Process., vol. 59, no. 1, pp. 351–361, Jan.
2011.
24. C. Jeong, I. Kim, and K. Dong, “Joint secure beamforming design at the source and the
relay for an amplify-and-forward MIMO untrusted relay system,” IEEE Trans. Signal Process.,
vol. 60, no. 1, pp. 310–325, Jan. 2012.
25. Y. Yang, et al., “Cooperative secure beamforming for AF relay networks with multiple
eavesdroppers,” IEEE Sig. Process. Lett., vol. 20, no. 1, pp. 35–38, Jan. 2013.
26. X. Liu, et al., “Joint beamforming and user selection in multicast downlink channel under
secrecy-outage constraint,” IEEE Commun. Lett., vol. 18, no. 1, pp. 82–85, Jan. 2014.
27. Y. Zou, J. Zhu, X. Wang, and V. Leung, “Improving physical-layer security in wireless
communications through diversity techniques,” IEEE Netw., vol. 29, no. 1, pp. 42–48, Jan.
2015.
28. Y. Zou, B. Champagne, W.-P. Zhu, and L. Hanzo, “Relay-selection improves the security-
reliability trade-off in cognitive radio systems,” IEEE Trans. Commun., vol. 63, no. 1,
pp. 215–228, Jan. 2015.
29. A. Jindal, C. Kundu, and R. Bose, “Secrecy outage of dual-hop AF relay system with relay
selection without eavesdropper’s CSI,” IEEE Commun. Lett., vol. 18, no. 10, pp. 1759–1762,
Oct. 2014.
30. Y. Zou, X. Wang, and W. Shen, “Physical-layer security with multiuser scheduling in cognitive
radio networks,” IEEE Trans. Commun., vol. 61, no. 12, pp. 5103–5113, Dec. 2013.
31. X. Ge, et al., “Secrecy analysis of multiuser downlink wiretap networks with opportunistic
scheduling,” in Proc. 2015 Intern. Conf. Commun., London UK, Jun. 2015.
32. B. Rankov and A. Wittneben, “Spectral efficient protocols for half-duplex fading relay
channels,” IEEE J. Sel. Areas Commun., vol. 25, no. 2, pp. 379–389. Feb. 2007.
33. I. Krikidis, et al., “Full-duplex relay selection for amplify-and-forward cooperative networks,”
IEEE Trans. Wirel. Commun., vol. 11, no. 12, pp. 4381–4393, Dec. 2012.
34. S. W. Peters, A. Y. Panah, K. T. Truong, and R. W. Heath, “Relay architectures for 3GPP LTE-
advanced,” EURASIP J. Wireless Commun. and Net., Vol. 2009, doi:10.1155/2009/618787.
35. S. Zhang, S. C. Liew, and P. P. Lam, “Hot topic: Physical-layer network coding,” in Proc. The
12th ACM Annual Intern. Conf. Mobile Comp. Net., Los Angeles, CA, Sept. 2006.
36. R. Louie, Y. Li, and B. Vucetic, “Practical physical layer network coding for two-way relay
channels: Performance analysis and comparison,” IEEE Trans. Wirel. Commun., vol. 9, no. 2,
pp. 764–777, Feb. 2010.
37. J.N. Laneman, D.N.C. Tse, and G.W. Wornell, “Cooperative diversity in wireless net-
works: Efficient protocols and outage behavior,” IEEE Trans. Inf. Theory, vol. 50, no. 12,
pp. 3062–3080, Dec. 2004.
38. A. Sendonaris, E. Erkip, and B. Aazhang, “User cooperation diversity part I: System
description,” IEEE Trans. commun., vol. 51, no. 11, pp. 1927–1938, Nov. 2003.
39. Y. Zou, Y.-D. Yao, and B. Zheng, “Opportunistic distributed space-time coding for decode-and-
forward cooperation systems,” IEEE Trans‘ Signal Process., vol. 60, no. 4, pp. 1766–1781, Apr.
2012.
40. M. Souryal and B. Vojcic, “Performance of amplify-and-forward and decode-and-forward
relaying in Rayleigh fading with turbo codes,” in Proc. 2006 IEEE Intern. Conf. Acoustics,
Speech and Signal Process. (IEEE ICASSP 2006), Toulouse, France, May 2006.
41. J. Laneman and G. W. Wornell, “Distributed space-time-coded protocols for exploiting cooper-
ative diversity in wireless networks,” IEEE Trans. Inf. Theory, vol. 49, no. 10, pp. 2415–2425,
Oct. 2003.
42. G. Scutari and S. Barbarossa, “Distributed space-time coding for regenerative relay networks,”
IEEE Trans. Wirel. Commun., vol. 4, no. 5, pp. 2387–2399, May 2005.
43. Y. Jing and B. Hassibi, “Distributed space-time coding in wireless relay networks,” IEEE Trans.
Wirel. Commun., vol. 5, no. 12, pp. 3524–3536, Dec. 2006.
References 15

44. A. Bletsas, H. Shin, M. Z. Win, and A. Lippman, “A simple cooperative diversity method based
on network path selection,” IEEE J. Select. Areas Commun., vol. 24, no. 3, pp. 659–672, Mar.
2006.
45. Y. Zou, J. Zhu, B. Zheng, and Y.-D. Yao, “An adaptive cooperation diversity scheme with
best-relay selection in cognitive radio networks,” IEEE Trans. Sig. Process., vol. 58, no. 10,
pp. 5438–5445, Oct. 2010.
46. Y. Zou, Y.-D. Yao, and B. Zheng, “Cognitive transmissions with multiple relays in cognitive
radio networks,” IEEE Trans. Wirel. Commun., vol. 10, no. 2, pp. 648–659, Feb. 2011.
47. Y. Zou, X. Wang, and W. Shen, “Optimal relay selection for physical-layer security in
cooperative wireless networks,” IEEE J. Sel. Areas Commun., vol. 31, no. 10, pp. 2099–2111,
Oct. 2013.
48. Y. Zou, J. Zhu, X. Li, and L. Hanzo, “Relay selection for wireless communications against
eavesdropping: A security-reliability tradeoff perspective,” IEEE Net., accepted to appear, Oct.
2015.
Chapter 2
Relay Selection for Enhancing Wireless
Secrecy Against Eavesdropping

Abstract In this chapter, we consider a wireless network consisting of a source and

a destination with the aid of multiple relay nodes, where an eavesdropper is assumed
with an intention to tap the confidential transmission from the source to destination.
Considering multiple relays available, we present a relay selection scheme to protect
the source-destination transmission against eavesdropping, where only the single
“best” relay is selected to help the source transmit the signal to the destination. For
comparison purposes, we also consider the conventional direct transmission and
random relay selection as benchmark schemes. As the name implies, the direct
transmission allows the source to directly transmit its signal to the destination
without relying on the relays. By contrast, in the random relay selection, a relay
is randomly selected to assist the source-destination transmission. Closed-form
intercept probability expressions are derived for the conventional direct transmission
and random relay selection as well as the proposed relay selection schemes over
Rayleigh fading channels. We also present the secrecy diversity analysis of these
there schemes and show that the proposed relay selection obtains the full secrecy
diversity, whereas the direct transmission and random relay selection methods
achieve the secrecy diversity order of only one. Numerical results demonstrate that
the proposed relay selection performs better than both the direct transmission and
random relay selection in terms of the intercept probability. Finally, as the number of
relays increases, the intercept performance of the proposed relay selection improves
significantly, showing the security benefits of exploiting the relay selection to defend
against eavesdropping.

2.1 System Model and Problem Formulation

Recently, extensive efforts have been devoted to the research of cooperative relays
for improving the performance of wireless networks from different perspectives,
including the network coverage [1–3], transmission reliability [4–10] and spectrum
utilization [11–15]. More specifically, in [1], the authors studied the deployment
of relays for the sake of maximizing the wireless network coverage for a given
data transmission rate in Gaussian relay channels, where the network coverage
is shown to be sensitive to the locations of relays and the path loss. Later on,
in [2] and [3], the network coverage region was developed for an amplify-and-

© Springer International Publishing Switzerland 2016 17

Y. Zou, J. Zhu, Physical-Layer Security for Cooperative Relay Networks,
Wireless Networks, DOI 10.1007/978-3-319-31174-6_2
18 2 Relay Selection for Enhancing Wireless Secrecy Against Eavesdropping

forward (AF) relay network strategy and the relay location has been optimized
for maximizing the coverage region. Alternatively, the employment of cooperative
relays was studied for enhancing the reliability of wireless transmissions in [4] and
[5], where several relay protocols are proposed, namely the fixed relay, incremental
relay and selective relay. It was shown in [6–10] that the outage probability of
wireless communications relying on the proposed relay protocols can be largely
reduced as compared to conventional direct transmission. Additionally, in [11–15],
cooperative relays were exploited in cognitive wireless networks for enabling
dynamic spectrum sharing between the primary and secondary users so that a better
spectrum utilization can be achieved.
The relay selection technique was first examined to combat the wireless fading
effect and thus improve the wireless reliability e.g. [16–22], where only the single
“best” relay is selected for participating in forwarding the signal transmission from a
source node to its destination. In this way, only two orthogonal channels (namely the
source-to-destination channel and the best-relay-to-destination channel) are needed
regardless of the number of cooperative relays. Meanwhile, it was shown in [16–22]
that the full diversity gain can be achieved by the best relay selection approach,
showing its significant advantage in terms of enhancing the wireless reliability.
More recently, there are some research works on improving the wireless physical-
layer security by using cooperative relays. For example, In [23–25], the authors
studied the physical-layer security of wireless communications in the face of an
eavesdropper with the help of a relay node, where the AF, decode-and-forward (DF),
and compress-and-forward (CF) protocols are investigated and compared with each
other. Moreover, a so-called noise-forwarding approach was devised in [26], where
a relay node is employed to send an artificially-designed noise for confusing the
eavesdropper without affecting the legitimate receiver.
In this chapter, we investigate the physical-layer security of a cooperative
wireless network consisting of a source transmitting to its destination with the aid
of multiple DF relays, where an eavesdropper is considered for tapping the source-
destination transmissions. The main contributions of this chapter can be summarized
as follows. First, we propose a single relay selection scheme, where only the
“best” relay is chosen among the multiple DF relays for assisting the source-
destination transmission against eavesdropping. For comparison purposes, we also
consider the conventional direct transmission and random relay selection as our
benchmarks. Second, we analyze the closed-form intercept probability performance
for the conventional direct transmission, random relay selection and proposed relay
selection over Rayleigh fading channels. Finally, some numerical simulations are
carried out, showing the advantage of the proposed relay selection over the direct
transmission and random relay selection in terms of the intercept probability.
We first present the system model of a wireless relay network in the face
of an eavesdropper and then formulate the physical-layer security problem for
wireless relay transmissions. As shown in Fig. 2.1, we consider a wireless relay
network, where multiple decode-and-forward (DF) relays are available to assist the
transmission from a source (S) to its destination (D). Presently, the relay architecture
2.1 System Model and Problem Formulation 19

Fig. 2.1 A source (S)

transmitting to its destination
(D) with the aid of N relays
in the face of an eave-
sdropper (E)

has been adopted in various commercial wireless network standards e.g. IMT-
advanced and IEEE 802.16j. For notational convenience, the set of DF relays is
denoted by R D fR1 ; R2 ;    ; RN g.
As depicted in Fig. 2.1, when S transmits its message denoted by x to D with
the help of N relays, an eavesdropper (E) appears and intends to tap the source
transmission. Throughout this book, we assume that both the D and E are beyond
the coverage of S and a relay node is exploited to forward the source signal x to D.
Although the D becomes capable of receiving the source signal from the relay
node, the E can also overhear and decode the relay transmission for the sake of
intercepting the source signal. Note that the solid and dash lines represent the
main channel and wiretap channel, respectively. Assume that all the wireless links
between any two network nodes of Fig. 2.1 are modeled as independent Rayleigh
fading channels. Moreover, all the receivers are considered to have the zero-mean
additive white Gaussian noise (AWGN) with a variance of N0 .
Following the physical-layer security literature, the E is assumed to know
everything about the S-D transmission, including the waveform, carrier frequency
and bandwidth, encoding and modulation, except that the source message x is
confidential. We assume that S transmits its signal x at a power of Ps . Thus, the
signal received at a relay node denoted by Ri is expressed as
p
yi D hsi Ps x C ni ; (2.1)

where hsi represents the wireless fading of the S-Ri channel and ni represents AWGN
received at Ri . Using the Shannon’s capacity formula, the capacity of the S-Ri
channel is obtained from (2.1) as

1
Csi D log2 .1 C jhsi j2 s /; (2.2)
2

where s D Ps =N0 represents the transmit signal-to-noise ratio (SNR) and the factor
1
2
is due to the fact that two orthogonal time slots are required for transmitting the
20 2 Relay Selection for Enhancing Wireless Secrecy Against Eavesdropping

source signal to the D via a relay node. If the relay Ri succeeds in decoding x and is
chosen to transmit the source signal, the received signal at D can be given by
p
yd D hid Ps x C nd ; (2.3)

where hid represents the wireless fading of the Ri -D channel and nd represents
AWGN received at Ri . Meanwhile, due to the openness nature of the wireless
propagation, the signal transmission of Ri can be overheard by E. Hence, the
received signal at E is expressed as
p
ye D hie Ps x C ne ; (2.4)

where hie represents the wireless fading of the Ri -E channel and ne represents the
AWGN at E. From (2.3), the capacity of the Ri -D channel is obtained as

1
Cid D log2 .1 C jhid j2 s /: (2.5)
2
Similarly, from (2.3), the capacity of the Ri -E channel is given by

1
Cie D log2 .1 C jhie j2 s /: (2.6)
2
As discussed in [26], when the channel capacity Cid is higher than Cie , a perfect
secrecy can be achieved in an information-theoretic sense. However, if Cid is lower
than Cie , the information-theoretic security becomes unachievable. In this case, the E
would succeed in intercepting the transmission of Ri and an intercept event is viewed
to occur. Additionally, the probability of the E intercepting the source message is
referred to as the intercept probability, which is used to measure the performance
of wireless physical-layer security. We focus on exploring the relay selection for
improving the wireless security in terms of decreasing the intercept probability.

2.2 Relay Selection for Secrecy Enhancement

In this section, we propose a relay selection scheme for enhancing the wireless
secrecy against eavesdropping and analyze the intercept probability of the proposed
scheme over Rayleigh fading channels. For comparison purposes, we also present
the conventional direct transmission and random relay selection methods.
Discovering Diverse Content Through
Random Scribd Documents
 BUSH-FIRE ON THE MAKONDE PLATEAU

This ascent, though barren of results in other respects, has

produced one small monument of African art, a drawing of our
climbing caravan, which is here offered to the reader’s inspection.
The native artist has quite correctly indicated the steepness of the
mountain by the vertical line representing the road. The confusion of
circles and curves at the lower end stands for the buildings of the
Mission station:—the foundations of a church vast enough, should it
ever be finished, to hold all the converted heathen of Africa and the
adjacent continents; the ci devant cowhouse, in which the two aged
clergymen have found a primitive refuge after the destruction of their
beautiful buildings by the Majimaji, the boys’ school and the girls’
school—two large bamboo huts in the native style; and the dwellings
of the native teachers and boarders. The curly labyrinth at the upper
end of the line is the top of the mountain with its gneiss blocks. The
two uppermost climbers are the kirongozi or guide and one of our
men, the third is Captain Ewerbeck, and the fourth myself. The
District Commissioner is readily recognizable by the epaulettes with
the two stars denoting his military rank, which belong to the uniform
worn on duty by this class of officials. Of all attributes of the white
man this seems to make the greatest impression on the native mind,
 since, in every drawing in my possession where
officers are represented, their rank is invariably (and
always correctly) indicated by the number of stars. In
the same way the native draughtsman never makes a
mistake with regard to the stripes on the sleeves of
non-commissioned officers, black or white. The
advantages of a well-developed corporation are here
evident! Ewerbeck and Seyfried are about the same
age as myself, and our chest and other measurements
are pretty nearly identical. This I suppose must be
the reason why the inhabitants of Lindi, and later on
those of the interior, have promoted me to the rank
of captain; at Lindi I went by the name of Hoffmani
mpya, “the new captain” (Hauptmann). The drawing
here reproduced is evidence of my promotion, the
artist having bestowed the epaulettes on me as well
as on Ewerbeck. The figures behind us are of no
importance, they are only the rest of our party. Now,
however, comes the psychologically noteworthy
point; I figure in the picture twice over, first
laboriously climbing the mountain, and then in
majestic pose at the top, in the act of photographing
OUR ASCENT the African landscape. You must know that the tripod
OF MTANDI
MOUNTAIN. shown in the drawing is that of my 13 × 18 cm.
DRAWN BY camera, the zig-zags between its legs are the brass
JUMA struts which keep it rigid; the long snake-like line is
the rubber tube for the release of the instantaneous
shutter—of which, as a matter of fact, I could make
no use on account of the mist,—and the photographer is, as above
stated, myself. The men behind me are my personal attendants to
whom the more fragile parts of the apparatus are usually entrusted.
The graphic reproduction of this ascent is no great achievement on
the part of the native intellect, but nevertheless it is a very important
document for the beginnings of art in general and for the African
point of view in particular. To the ethnographer, of all men, the most
apparently insignificant matters are not without importance, and
this is why the prospect of working undisturbed for many months in
these surroundings is such a delight to me.
 Our ascent of Mtandi was concluded, at any rate for the present, by
a ceremonious breakfast, to which the two missionaries had kindly
invited us. Englishmen, as is well known, live extremely well in their
own country; but abroad, too, even in the far interior of a continent,
they know how to make the best of things. I was here impressed with
the fact that Masasi must be a “very nourishing district,” as Wilhelm
Raabe would say. We had no champagne, it is true—Archdeacon
Carnon had set it before us on the previous day, in a huge water-jug,
apologizing for the absence of champagne glasses. We showed him
that we were able to appreciate his hospitality, even in the absence of
such refinements.
The merriest part of our whole Mtandi expedition, however, was
the ride home, with the Mission pupils trotting along beside us. The
little fellows looked warlike enough with their bows and arrows, and
seemed desirous of shouting each other down. I could not at first
make out what they wanted, but on reaching home, that is to say, our
police-post, I soon understood that their object was nothing less than
to offer me the whole of their martial equipment for my ethnographic
collection. But not as a present—giving things away for nothing is not
in the negro’s line, and in this he resembles our German rustics. On
the contrary, these young people demanded fancy prices for the bows
which they had made on purpose to sell them to the mzungu, that
remarkable character who buys all sorts of native rubbish. I
purchased such of their wares as seemed suitable for my objects, and
thought it advisable to prevent disappointment to those whose offers
had been refused by giving each a copper or two out of the famous jar
of which we shall hear again later on. Before doing so, however, I
instituted a pleasing experiment, instructive for myself and highly
enjoyable for the youth of Masasi, in the shape of an archery
competition.
Comparative ethnography has for a long time past busied itself
with the task of classifying and analyzing all the technical and mental
activities of man. Thus some decades ago, the American, Morse,[12]
ascertained that all men who shoot, or ever have shot, with the bow,
have certain definite ways of drawing it. There are about half-a-
dozen distinct methods, which are so distributed over the globe that,
in some places the same release (or “loose” as it is technically called)
is known to be common to the whole of a large area, while elsewhere
the most abrupt contrasts may be observed between contiguous
nations or tribes. It might be supposed that there could be no
possible differences in so simple an action as that of drawing a bow;
but experiment shows otherwise, and this experiment I have made
over and over again in the course of my lectures.
It is a thousand to one that any German (leaving out of
consideration the English and the Belgians, who still practise archery
according to the rules of the game, and can distinguish a good “loose”
from a bad one), when he has taken the bow in his left hand and
grasped the arrow and the string in his right, will hold the notch as it
rests on the string between his thumb and forefinger, and thus only
indirectly draw the string by means of the arrow. This, which is the
“loose” we used on the little toy bows of our boyhood, is the very
worst conceivable, as anyone who understands the other methods
can convince himself by every shot he tries. It is obvious that the
arrow must slip from the fingers if a moderately strong pull is given.
The best proof of the inferiority of this particular “loose” is the fact
that it is very seldom found among those sections of mankind who
still use the bow as a serious and effective weapon, whether in war or
hunting. These handle it after a very different fashion. Only where
the bow is a mere survival, and only used as a toy by children (the
most conservative class in the community), as for instance among
ourselves, this method, quite useless for an effective shot, is
practised simply because no better is known.
If I felt compelled to take the boys at Masasi Mission as a standard
for estimating the culture of the race, I should have to say that here
too the bow is a survival, for nine-tenths of the whole multitude shot
in the same way as our boys at home, but with one difference; we
hold the bow horizontally, the African boys held it vertically, the
arrow lying on the left side of the string between the index and
middle finger. Only one-tenth of the whole number used a different
“loose,” and these, significantly enough, were older boys, who
therefore had evidently taken over with them into their Christianity a
considerable dose of old African conservatism.
My competition was arranged with a view, not so much of
registering the number of hits and misses, as of observing the
method of drawing; but, notwithstanding, I must say that the little
archers acquitted themselves by no means contemptibly. It is true
that the distances were short, and my mark was scarcely a small one,
being a copy of the Tägliche Rundschau; but the greater number sent
their arrows inside the rings I had hastily drawn on this improvised
target. They were proud of their success, too; and when I praised a
good shot it was good to see the triumphant looks that the little black
hero cast round on his admiring companions.
As to the other methods, if I were asked the question in my Leipzig
lecture-room, I should have to answer it at once. As it is, I am
enabled to claim the privilege of the investigator and excuse myself
from giving further information till I have collected sufficient
material by a series of fresh observations. I hope to gratify my
readers’ thirst for knowledge when I have traversed the whole plain
north of the Rovuma, and, encamped on the cool heights of the
Makonde plateau, find leisure to look back and take stock of my
studies. Till then—Au revoir, Messieurs!
MNYASA HUNTER WITH DOG. DRAWN BY SALIM MATCLA
 CHAPTER VI
NATIVE LIFE SEEN FROM THE INSIDE

Masasi, end of July, 1906.

Every normal human being is a walking demonstration of the theory

of adaptation to environment. I have been in Africa barely two
months, and only as yet a fraction of a month in the interior, and yet
I feel quite at home already. After all, I could scarcely do otherwise.
On the 21st, when we had only lived together a few days, Mr.
Ewerbeck marched away before daybreak, by the light of a lantern
borne before him through the darkness of the tropic night, to attend
to higher duties at Lindi, viz., the reception of the eight delegates
from the Reichstag, now fairly embarked on that desperate
adventure which for many months past has kept our daily press busy
celebrating their heroism.
Nils Knudsen remains behind as the last relic of civilization. His
name alone is sufficient to indicate his Scandinavian origin, and he
is, in fact, a fair-haired descendant of the Vikings. He joined the
expedition so unobtrusively that at first I scarcely noticed the
presence of a third European. While Ewerbeck and I marched
proudly at the head of our long line of followers, Knudsen usually
brought up the rear, and in camp he remained modestly in the
background. Now that we have fixed our headquarters at Masasi, he
has become prominent by virtue of his office; he is supposed to keep
things straight here and exercise some supervision over the native
local authorities. Whether this is necessary, I am at present unable to
judge, having as yet no insight into the difficulties of internal
administration in a large district like Lindi. However, a man who
knows the country as well as Ewerbeck does, would hardly have
taken such a measure without good reason. In the meantime I have
persuaded Knudsen to quit his tent—which, to judge by its venerable
appearance, must have been left behind as too far gone to take away,
by Vasco Da Gama when he landed in this part of Africa—and come
to live with me in the rest-house. Now he is installed, with his scanty
possessions—two old tin trunks, which do not even appear to be full
—on one side of the spacious apartment, while I with my princely
outfit reside on the other. He is, however, abundantly compensated
for the niggardliness with which fortune has treated him by goodness
of heart and fineness of feeling. Knudsen’s life has been adventurous
enough, and recalls to some extent the fate of that English sailor who
was wrecked among the aborigines of South-East Australia, and had
to live as a savage among savages. My fair-haired neighbour did not
fare quite so badly as that; but he has had plenty of time to “go Fanti”
had he been so disposed. So far as I have yet ascertained anything
about his personal affairs, he started life as cabin-boy on board a
merchant vessel, from which he ran away about ten years ago, when
it was anchored in a harbour of Madagascar. He wandered about this
island for some years, and at last found his way across to the
mainland and into the hinterland of Lindi. He says that he never
learnt a trade, but professes to know something of a great many, and
can act on occasion as mason, builder, carpenter, and locksmith.
Indeed he erected all the buildings at the Luisenfelde mines, far
south near the Rovuma, which I may yet be able to visit, and was
general factotum there as long as they continued working. Since then
the municipality of Lindi has appointed him head instructor at the
industrial school, from which post he is at present on leave of
absence.
 THROUGH THE BUSH ON A COLLECTING EXCURSION

Our manner of life here is, of course, essentially different from that
followed on the march. Life on the march is always full of charm,
more especially in a country quite new to one; and mine has so far
been entirely without drawbacks. In African travel-books we find
that almost every expedition begins with a thousand difficulties. The
start is fixed for a certain hour, but no carriers appear, and when at
last the leader of the expedition has, with infinite pains, got his men
together, they have still endless affairs to settle, wives and
sweethearts to take leave of, and what not, and have usually vanished
from the traveller’s ken on the very first evening. In my case
everything went like clockwork from the start. I can blame no one
but myself for the quarter of an hour’s delay in starting from Lindi,
which was caused by my being late for breakfast. On the second
morning the askari could not quite get on with the folding of the
tent, and Moritz with the best will in the world failed to get my
travelling-lamp into its case, which was certainly a very tight fit. But
with these exceptions we have all behaved as if we had been on the
road for months. Anyone who wants a substantial breakfast first
thing in the morning, after the English fashion, should not go
travelling in Africa. I have given directions to wake me at five.
Punctually to the minute, the sentinel calls softly into the tent,
“Amka, bwana” (“Wake up, sir”). I throw both feet over the high
edge of the trough-like camp bed, and jump into my khaki suit. The
water which Kibwana, in the performance of his duties as
housemaid, has thoughtfully placed at the tent door overnight, has
acquired a refreshing coolness in the low temperature of a tropic
night in the dry season. The shadow of the European at his toilet is
sharply outlined on the canvas by the burning lamp, which, however,
does not confine its illumination to its owner, but radiates a circle of
light on the shining brown faces of the carriers and the askari. The
former are busy tying up their loads for the march, while the soldiers
are ready to rush on the tent like a tiger on his prey, so soon as the
white man shall have finished dressing and come out. In the
twinkling of an eye the tent is folded, without a word spoken, or a
superfluous movement; it is division of labour in the best sense of
the word, faultlessly carried out. Meanwhile the traveller goes to his
camp-table, takes a hurried sip of tea, cocoa, or whatever his
favourite beverage may be, eating at the same time a piece of bread
baked by himself, and now stands ready for the march. “Tayari?”
(“Ready?”) his voice rings out over the camp. “Bado” (“Not yet”) is
the invariable answer. It is always the same lazy or awkward
members of the party who utter this word beloved of the African
servant. The beginner lets himself be misled by it at first, but in a few
days he takes no more notice of the “Bado,” but fires off his “Safari!”
(literally “Journey!”) or (as speedily introduced by me), “Los!”[13] at
the band in general, flourishes his walking-stick boldly in the air,
thereby indicating to the two leading askari the direction of the
march, and the day’s work has begun.
I do not know how other tribes are accustomed to behave at the
moment of starting, but my Wanyamwezi are certainly neither to
hold nor to bind on these occasions. With evident difficulty each one
has got his load lifted to head or shoulder, and stands in his place
bending under the weight. At the word of command arises an uproar
which baffles description. All the pent-up energy of their throats
rings out into the silent forest; stout sticks rattle in a wild, irregular
rhythm on the wooden cases, and, alas! also on the tin boxes, which
furnish only too good a resonator. The noise is infernal, but it is a
manifestation of joy and pleasure. We are off! and, once on the
march, the Wanyamwezi are in their element. Before long the chaos
of noise is reduced to some order; these men have an infinitely
delicate sense of rhythm, and so the din gradually resolves itself into
a kind of march sung to a drum accompaniment, whose charm even
the legs of the askari—otherwise too dignified for such childish
goings-on—cannot resist.

READY FOR MARCHING (MASASI)

Oh! the beauty of these early mornings in the tropics! It is now

getting on for six o’clock; the darkness of night has quickly yielded to
the short twilight of dawn; the first bright rays gild the light clouds
floating in the sky, and suddenly the disc of the sun rises in its
wonderful majesty above the horizon. With swift, vigorous strides,
and still in close order, the procession hastens through the dew-
drenched bush, two soldiers in the van, as if in a military expedition;
then, after an interval we Europeans, immediately followed by our
personal servants with guns, travelling-flask and camp-stool. Then
comes the main body of the soldiers followed by the long line of
carriers and the soldiers’ boys, and, lastly, to keep the laggards up to
the mark, and also to help any who have to fall out from exhaustion
or illness, two soldiers bringing up the rear. An admirable figure is
the mnyampara or headman. His position is in a sense purely
honorary, for he receives not a farthing more wages than the lowest
of his subordinates. Perhaps even this expression should not be used;
he is rather primus inter pares. The mnyampara is everywhere. He
is in front when the master sends for him, and he is back at the very
end of the line (which becomes longer with every hour of the march)
if there is a sick man needing his help. In such a case he carries the
man’s load himself, as a matter of course, and brings him safely to
camp. It seems to me that I have made an unusually happy choice in
Pesa mbili. He is young, like the great majority of my men, probably
between 23 and 25, of a deep black complexion, with markedly
negroid features, and a kind of feline glitter in his eyes; he is only of
medium height, but uncommonly strong and muscular; he speaks
shocking Swahili—far worse than my own—and withal he is a
treasure. It is not merely that he is an incomparable singer, whose
pleasant baritone voice never rests whether on the march or in camp,
but he thoroughly understands the organization of camp life, the
distribution of tasks and the direction of his men. The demands
made on such a man by the end of the day’s march are arduous
enough.
The delicious coolness of the morning has long since given place to
a perceptibly high temperature; the white man has exchanged his
light felt hat or still lighter travelling-cap for the heavy tropical
helmet, and the naked bodies of the carriers are coated with a
shining polish. These, who have been longing for the day to get warm
ever since they awoke shivering round the camp fire at four, have
now reached the goal of their desires; they are warm—very warm—
and the white man will do well to march at the head of the caravan,
otherwise he will find opportunities more numerous than agreeable
for studying the subject of “racial odours.” After two hours, or two
hours and a half, comes the first halt. The European shouts for his
camp-stool and sits watching the long string of loads coming up and
being lowered to the ground. A frugal breakfast of a couple of eggs, a
piece of cold meat, or a few bananas, here awaits the traveller, but
the carriers, who started without a meal, steadily fast on. It seems
incomprehensible that these men should be able to march for many
hours with a load of sixty or seventy pounds, while practising such
abstinence, but they are quite content to have it so. In the later hours
of the day, it is true, they begin to flag, their steps become slower and
shorter, and they lag more and more behind the personal “boys” who
have no heavy loads to carry. Yet when they reach camp at last, they
are as merry and cheerful as they were in the early morning. The
same noise—though now with quite different words from the throats
of the singers—overwhelms the European, who has long been seated
at the halting-place. My company seem to be obsessed by the
“Central-Magazin” at Dar es Salam, where they entered my service;
they are celebrating this spacious building in the closing song of their
day’s march.

CAMP AT MASASI

The duties of my followers—whether boys, askari, or porters—are

by no means over when they have reached camp. By the time they
come up, the leader of the expedition has looked round for a place to
pitch his tent, a matter which seems to me to require special gifts.
The fundamental principles to bear in mind are: that it should be
within reach of good drinking water and free from noxious insects,
such as ticks, mosquitoes, and jiggers. The second point, but one by
no means to be overlooked, is the position of the tent-pole with
regard to the course of the sun, and the next the shade of leafy trees,
if that is attainable. I find it simplest to draw the outline of the tent
on the sandy ground, after the spot has been carefully swept,
indicating the place where I want the door to be by a break in the
line. That is quite enough for my corporal in command. Scarcely have
the two unfortunates, whose shoulders are weighed down by my
heavy tent, come up panting and gasping for breath, when the loads
are unrolled, and in a twinkling every warrior has taken up his
position. “One, two, three!” and the two poles are in their places, and
the next moment I hear the blows of the mallet on the tent-pegs.
While this is going on, the two boys, Moritz and Kibwana, are
amusing themselves with my bed. This occupation seems to
represent for them the height of enjoyment, for it seems as if they
would never be done. Neither scolding nor threats can avail to hasten
their movements. It seems as if their usually slow brains had become
absolutely torpid. Mechanically they set up the bedstead;
mechanically they spread the cork mattress and the blankets over it;
in the same dull, apathetic way they finally set up the framework of
the mosquito-net. The soldiers have taken their departure long
before my two gentlemen condescend to carry the bed into the tent.
My carriers meanwhile have found all sorts of work to do. Water
has to be fetched for the whole caravan, and fires to be made, and the
sanitary requirements of the camp provided for; and noon is long
past by the time their turn comes and they can live their own life for
an hour or two. Even now they cannot be said to revel in luxury. This
southern part of the German territory is very poor in game, and in
any case I have no time for shooting, so that meat is almost an
unknown item in my people’s menu. Ugali, always ugali—stiff
porridge of millet, maize or manioc, boiled till it has almost a
vitreous consistency, and then shaped with the spoon used for
stirring into a kind of pudding—forms the staple of their meals day
after day.
 INTERIOR OF A NATIVE HUT IN THE ROVUMA VALLEY

Here at Masasi the tables are turned; my men have a good time,
while I can scarcely get a minute to myself. My escort are quite
magnificently housed, they have moved into the baraza or council-
house to the left of my palatial quarters and fitted it up in the native
way. The negro has no love for a common apartment; he likes to
make a little nest apart for himself. This is quickly done: two or three
horizontal poles are placed as a scaffolding all round the projected
cabin, then a thick layer of long African grass is tied to them, and a
cosy place, cool by day and warm by night, is ready for each one. The
carriers, on the other hand, have built themselves huts in the open
space facing my abode, quite simple and neat, but, to my
astonishment, quite in the Masai style—neither circular hut nor
tembe. The circular hut I shall discuss in full later on, but in case
anyone should not know what a tembe is like, I will here say that the
best notion of it can be got by placing three or four railway carriages
at right angles to one another, so that they form a square or
parallelogram, with the doors inward. This tembe is found
throughout most of the northern and central part of German East
Africa, from Unyamwezi in the west to the coast on the east, and
from the Eyasi and Manyara basin in the north to Uhehe in the
south. The Masai hut, finally, can best be compared with a round-
topped trunk. Though the Masai, as everyone knows, usually stand
well over six feet, their huts, which (quite conformably with the
owners’ mode of life as cattle-breeders par excellence) are neatly and
fragrantly plastered with cowdung, are so low that even a person of
normal stature cannot stand upright in them. My Wanyamwezi,
however, never attempt to stand up in their huts; on the contrary,
they lie about lazily all day long on their heaps of straw.
My activities are all the more strenuous. The tropical day is short,
being only twelve hours from year’s end to year’s end, so that one has
to make the fullest possible use of it. At sunrise, which of course is at
six, everyone is on foot, breakfast is quickly despatched, and then the
day’s work begins. This beginning is curious enough. Everyone who
has commanded an African expedition must have experienced the
persistence of the natives in crediting him with medical skill and
knowledge, and every morning I find a long row of patients waiting
for me. Some of them are my own men, others inhabitants of Masasi
and its neighbourhood. One of my carriers has had a bad time. The
carrier’s load is, in East Africa, usually packed in the American
petroleum case. This is a light but strong wooden box measuring
about twenty-four inches in length by twelve in width and sixteen in
height, and originally intended to hold two tins of “kerosene.” The
tins have usually been divorced from the case, in order to continue a
useful and respected existence as utensils of all work in every Swahili
household; while the case without the tins is used as above stated.
One only of my cases remained true to its original destination, and
travelled with its full complement of oil on the shoulders of the
Mnyamwezi Kazi Ulaya.[14] The honest fellow strides ahead sturdily.
“It is hot,” he thinks. “I am beginning to perspire. Well, that is no
harm; the others are doing the same.... It is really very hot!” he
ejaculates after a while; “even my mafuta ya Ulaya, my European
oil, is beginning to smell.” The smell becomes stronger and the
carrier wetter as the day draws on, and when, at the end of the
march, he sets down his fragrant load, it is with a double feeling of
relief, for the load itself has become inexplicably lighter during the
last six hours. At last the truth dawns on him and his friends, and it
is a matter for thankfulness that none of them possess any matches,
for had one been struck close to Kazi Ulaya, the whole man would
have burst into a blaze, so soaked was he with Mr. Rockefeller’s
stock-in-trade.
 Whether it is to be accounted for by a strong sense of discipline or
by an almost incredible apathy, the fact remains that this man did
not report himself on the first day when he discovered that the tins
were leaking, but calmly took up his burden next morning and
carried it without a murmur to the next stopping place. Though once
more actually swimming in kerosene, Kazi Ulaya’s peace of mind
would not even now have been disturbed but for the fact that
symptoms of eczema had appeared, which made him somewhat
uneasy. He therefore presented himself with the words a native
always uses when something is wrong with him and he asks the help
of the all-powerful white man—“Dawa, bwana” (“Medicine, sir”),
and pointed significantly, but with no sign of indignation, to his
condition. A thorough treatment with soap and water seemed
indicated in the first instance, to remove the incrustation of dirt
accumulated in seven days’ marching. It must be said, in justice to
the patient, that this state of things was exceptional and due to
scarcity of water, for Kazi Ulaya’s personal cleanliness was above the
average. I then dressed with lanoline, of which, fortunately, I had
brought a large tin with me. The patient is now gradually getting over
his trouble.
Another case gives a slight idea of the havoc wrought by the jigger.
One of the soldiers’ boys, an immensely tall Maaraba from the
country behind Sudi, comes up every morning to get dawa for a
badly, damaged great toe. Strangely enough, I have at present
neither corrosive sublimate nor iodoform in my medicine chest, the
only substitute being boric acid tabloids. I have to do the best I can
with these, but my patients have, whether they like it or not, got
accustomed to have my weak disinfectant applied at a somewhat
high temperature. In the case of such careless fellows as this
Maaraba, who has to thank his own lazy apathy for the loss of his
toe-nail (which has quite disappeared and is replaced by a large
ulcerated wound), the hot water is after all a well-deserved penalty.
He yells every time like a stuck pig, and swears by all his gods that
from henceforth he will look out for the funsa with the most
unceasing vigilance—for the strengthening of which laudable
resolutions his lord and master, thoroughly annoyed by the childish
behaviour of this giant, bestows on him a couple of vigorous but
kindly meant cuffs.
 As to the health of the Masasi natives, I prefer to offer no opinion
for the present. The insight so far gained through my morning
consultations into the negligence or helplessness of the natives as
regards hygiene, only makes me more determined to study other
districts before pronouncing a judgment. I shall content myself with
saying here that the negro’s power of resisting the deleterious
influences of his treacherous continent is by no means as great as we,
amid the over-refined surroundings of our civilized life, usually
imagine. Infant mortality, in particular, seems to reach a height of
which we can form no idea.
Having seen my patients, the real day’s work begins, and I march
through the country in the character of Diogenes. On the first few
days, I crawled into the native huts armed merely with a box of
matches, which was very romantic, but did not answer my purpose. I
had never before been able to picture to myself what is meant by
Egyptian darkness, but now I know that the epithet is merely used on
the principle of pars pro toto, and that the thing belongs to the whole
continent, and is to be had of the very best quality here in the plain
west of the Makonde plateau. The native huts are entirely devoid of
windows, a feature which may seem to us unprogressive, but which is
in reality the outcome of long experience. The native wants to keep
his house cool, and can only do so by excluding the outside
temperature. For this reason he dislikes opening the front and back
doors of his home at the same time, and makes the thatch project
outward and downward far beyond the walls. My stable-lantern,
carried about the country in broad daylight by Moritz, is a great
amusement to the aborigines, and in truth our proceeding might well
seem eccentric to anyone ignorant of our object. In the darkness of a
hut-interior, however, they find their complete justification. First
comes a polite request from me, or from Mr. Knudsen, to the owner,
for permission to inspect his domain, which is granted with equal
politeness. This is followed by an eager search through the rooms
and compartments of which, to my surprise, the dwellings here are
composed. These are not elegant, such a notion being as yet wholly
foreign to the native consciousness; but they give unimpeachable
testimony to the inmates’ mode of life. In the centre, midway
between the two doors is the kitchen with the hearth and the most
indispensable household implements and stores. The hearth is
simplicity itself: three stones the size of a man’s head, or perhaps
only lumps of earth from an ant-heap, are placed at an angle of 120°
to each other. On these, surrounded by other pots, the great earthen
pot, with the inevitable ugali, rests over the smouldering fire. Lying
about among them are ladles, or spoons, and “spurtles” for stirring
the porridge. Over the fireplace, and well within reach of the smoke,
is a stage constructed out of five or six forked poles. On the cross-
sticks are laid heads of millet in close, uniform rows, and under
them, like the sausages in the smoke-room of a German farmhouse,
hang a great number of the largest and finest cobs of maize, by this
time covered with a shining layer of soot. If this does not protect
them from insects, nothing else will; for such is the final end and aim
of the whole process. In the temperate regions of Europe, science
may be concerned with preserving the seed-corn in a state capable of
germination till sowing-time; but here, in tropical Africa, with its all-
penetrating damp, its all-devouring insect and other destroyers, and,
finally, its want of suitable and permanent building material, this
saving of the seed is an art of practical utility. It will be one, and not
the least welcome, of my tasks, to study this art thoroughly in all its
details.
As to the economy of these natives, their struggle with the
recalcitrant nature of the country, and their care for the morrow, I
am waiting to express an opinion till I shall have gained fuller
experience. In the literature dealing with ethnology and national
economy, we have a long series of works devoted to the classification
of mankind according to the forms and stages of their economic life.
It is a matter of course that we occupy the highest stage; all authors
are agreed on one point, that we have taken out a lease of civilization
in all its departments. As to the arrangement of the other races and
nations, no two authors are agreed. The text-books swarm with
barbarous and half-barbarous peoples, with settled and nomadic,
hunter, shepherd, and fisher tribes, migratory and collecting tribes.
One group carries on its economic arts on a basis of tradition,
another on that of innate instinct, finally, we have even an animal
stage of economics. If all these classifications are thrown into a
common receptacle, the result is a dish with many ingredients, but
insipid as a whole. Its main constituent is a profound contempt for
those whom we may call the “nature-peoples.”[15] These books
produce the impression that the negro, for instance, lives direct from
hand to mouth, and in his divine carelessness takes no thought even
for to-day, much less for to-morrow morning.
The reality is quite otherwise, here and elsewhere, but here in an
especial degree. In Northern Germany, the modern intensive style of
farming is characterized by the barns irregularly distributed over the
fields, and in quite recent times by the corn-stacks, both of which,
since the introduction of the movable threshing-machine, have made
the old barn at the homestead well-nigh useless. Here the farming
differs only in degree, not in principle; here, too, miniature barns are
irregularly scattered over the shambas, or gardens; while other food-
stores which surprise us by their number and size are found close to
and in the homestead. If we examine the interior of the house with a
light, we find in all its compartments large earthen jars, hermetically
sealed with clay, containing ground-nuts, peas, beans, and the like,
and neatly-made bark cylinders, about a yard long, also covered with
clay and well caulked, for holding maize, millet and other kinds of
grain. All these receptacles, both outdoor and indoor, are placed to
protect them from insects, rodents and damp, on racks or platforms
of wood and bamboo, from fifteen inches to two feet high, plastered
with clay, and resting on stout, forked poles. The outdoor food-stores
are often of considerable dimensions. They resemble gigantic
mushrooms, with their thatched roofs projecting far beyond the
bamboo or straw structure, which is always plastered with mud
inside and out. Some have a door in their circumference after the
fashion of our cylindrical iron stoves; others have no opening
whatever, and if the owner wishes to take out the contents, he has to
tilt the roof on one side. For this purpose he has to ascend a ladder of
the most primitive construction—a couple of logs, no matter how
crooked, with slips of bamboo lashed across them a yard apart. I
cannot sketch these appliances without a smile, yet, in spite of their
primitive character, they show a certain gift of technical invention.
The keeping of pigeons is to us Europeans a very pleasing feature
in the village economy of these parts. Almost every homestead we
visit has one or more dovecotes, very different from ours, and yet
well suited to their purpose. The simplest form is a single bark
cylinder, made by stripping the bark whole from the section of a
moderately thick tree. The ends are fastened up with sticks or flat
stones, a hole is cut in the middle for letting the birds in and out, and
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com