entropy

Article
Energy Harvesting for Physical Layer Security in
Cooperative Networks Based on Compressed Sensing
Shuai Chang 1 , Jialun Li 2 , Xiaomei Fu 1, * ID
and Liang Zhang 1
1 School of Marine Science and Technology, Tianjin University, Tianjin 300072, China;
shuai.chang@tju.edu.cn (S.C.); liangzhang@tju.edu.cn (L.Z.)
2 School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China;
Li_five@tju.edu.cn
* Correspondence: fuxiaomei@tju.edu.cn; Tel.: +86-22-87370655

Received: 20 July 2017; Accepted: 25 August 2017; Published: 1 September 2017

Abstract: Energy harvesting (EH) has attracted a lot of attention in cooperative communication
networks studies for its capability of transferring energy from sources to relays. In this paper,
we study the secrecy capacity of a cooperative compressed sensing amplify and forward (CCS-AF)
wireless network in the presence of eavesdroppers based on an energy harvesting protocol. In this
model, the source nodes send their information to the relays simultaneously, and then the relays
perform EH from the received radio-frequency signals based on the power splitting-based relaying
(PSR) protocol. The energy harvested by the relays will be used to amplify and forward the received
information to the destination. The impacts of some key parameters, such as the power splitting
ratio, energy conversion efficiency, relay location, and the number of relays, on the system secrecy
capacity are analyzed through a group of experiments. Simulation results reveal that under certain
conditions, the proposed EH relaying scheme can achieve higher secrecy capacity than traditional
relaying strategies while consuming equal or even less power.

Keywords: secrecy capacity; cooperative wireless communication; energy harvesting; compressed

sensing; physical layer security

1. Introduction
Physical-layer security approaches can enhance the security of wireless sensor networks (WSNs)
against eavesdroppers by exploiting the physical characteristics of wireless channels, such as noise and
multipath fading [1]. The secrecy capacity is the maximum rate of secret information that can be sent
from the source to the destination in the presence of eavesdroppers. The cooperative relaying strategy
has recently been used as a practical technology to provide transmission secrecy [25], to break through
the limitation on the channel conditions that the main channel (from the source to the destination)
must be better than the eavesdropper channel (from the source to the eavesdropper). Many kinds
of approaches have been proposed to enhance the physical-layer security under cooperative
communication framework conditions, such as amplify-and-forward (AF) [6,7], decode-and-forward
(DF) [8], compress-and-forward (CF) [8], noise-forwarding (NF) [9], etc.
In the past few years, radio frequency (RF) energy harvesting (EH) has becomes a research
hotspot in wireless communications, especially in applications where the battery-limited devices
are difficult to replace or recharge [10]. Simultaneous wireless information and power transfer
(SWIPT) [1113] was proposed to improve the energy utilization efficiency of wireless networks.
Under this scheme, the nodes in wireless networks can be energy self-sufficient by harvesting RF
signals from the surrounding environment.
Cooperative communication is exploited for its capability to further improve the efficiency of
SWIPT systems [14]. As the RF signals can carry information and energy concurrently, thereby in

Entropy 2017, 19, 462; doi:10.3390/e19090462 www.mdpi.com/journal/entropy

Entropy 2017, 19, 462 2 of 11

such wireless networks, the cooperative relays are able to harvest energy and process information
simultaneously [11]. There are two kinds of relaying protocols: power splitting relaying (PSR) protocols
and time switching relaying (TSR) protocols [11,1517]. In a PSR protocol, the received signals are
divided by a power-splitting ratio to operate EH and information processing (AF or DF), and the
EH and information processing are performed simultaneously [10,11,15,16], while, in a TSR protocol,
the EH is performed first, and the information processing is performed in the remainder of the
total transmission period [10,11,15,17,18]. Comparison studies between PSR and TSR are conducted
in [10,15], and the PSR approach is proved to be better than the TSR in EH-based multi-antenna
AF relaying networks. Especially in [15], an eavesdropper is considered in the network, and the
destination transmits noise information to interrupt the eavesdropper. In [16], the EH-AF and
EH-DF schemes with a single EH relay and single eavesdropper are compared based on TSR protocol,
concluding that the EH-DF protocol outperforms the EH-AF protocol in terms of secrecy capacity.
In [18], the authors consider an underlay cognitive radio network (CRN) with a pair of primary
nodes, a couple of secondary nodes, and one eavesdropper. The EH is performed on the secondary
transmitter. Experimental results show that the EH nodes could improve both energy efficiency and
spectral efficiency.
Compressed sensing (CS) technology has recently become an effective solution to improve the
physical layer security in cooperative wireless networks. CS could represent the compressible signals
at a rate below the Nyquist rate, and the information could be retrieved from a small number of linear
measurements [1921]. The application of CS in the field of information-theoretic secrecy has attracted
researchers attention and a series of studies have been conducted. In [22,23], the authors consider the
scenario of one source node, one receiver and one eavesdropper (i.e., a point-to-point scheme). In [22],
the measurement matrix is treated as an encryption key which is unknown to the eavesdroppers. It can
provide computational secrecy with unbounded eavesdropper computation capability. In [23], perfect
secrecy is achievable under the condition that the number of source messages goes to infinity. Different
from these methods, the authors in [24] considered the situation of keyless physical-layer security,
and indicated that the eavesdropper could not decode the information successfully in terms of the
Wolfowitz secrecy. In [25], the CS matrix is used to encode the messages. If the wire-tap channels are
strictly worse than the main channels, the eavesdroppers can learn almost nothing, thereby, in this
situation, it is unnecessary for us to know the channel state information (CSI) of eavesdroppers.
However, the transmitting methods with CS in point-to-point communication scenarios cannot be
directly applied to WSNs, because WSNs possess the property of decentralization [26]. The CS-AF
scheme was first proposed and applied in wireless networks in [27], and the channel capacity of the
CS-AF scheme was investigated in [28]. It is a novel approach in that the channel matrix from sources
to relays is the compressive matrix in CS technology, and thus can help to achieve security. It has been
proved that the recovery probability of the eavesdroppers under a CS-AF scheme can be arbitrarily
small. No more researches considering the channel matrixes as the compressed matrix have been
found, and the study of EH in this kind of networks is still a new area.
In this paper, we study the secrecy capacity of the CCS-AF wireless network based on the EH
protocol. In this model, the sources send their information to the relays simultaneously, and the relays
harvest energy from the radio-frequency signals of sources based on the PSR protocol. Through this
protocol, the energy harvested is used by the cooperative relays to amplify and forward the received
information to the destination. A group of simulation experiments is conducted to analyze the impacts
of some key parameters (such as power splitting ratio, EH efficiency, relay location, and the number of
relays) on the system secrecy capacity. Simulation results reveal that under certain conditions, the EH
relaying scheme in this paper can achieve higher secrecy capacity than traditional relaying strategies
while costing the same or even less power.
The rest of this paper is organized as follows: in Section 2, the CCS-AF model is proposed and
introduced in detail. Then a brief introduction of PSR protocol is presented in Section 3. Simulations
with analysis are presented in Section 4. Finally, conclusions are drawn in Section 5.
Entropy 2017, 19, 462 3 of 11
Entropy 2017, 19, 462 3 of 10

2. System Model
2. System Model
Figure 1 shows the communication network of CCS-AF with eavesdroppers and the scheme
Figure 1 shows the communication network of CCS-AF with eavesdroppers and the scheme of
of the PSR protocol. This network contains N source nodes (S1 , . . . , S N ), M relays ( R1 , . . . , R M ),
the PSR protocol. This network contains source nodes ( , , ), relays ( , , ),
Z eavesdroppers ( E1 , . . . , EZ ) and one destination (D). In the first time slot, the sources transmit
eavesdroppers ( , , ) and one destination ( ). In the first time slot, the sources transmit their
their information to the relays simultaneously with a fixed transmission power PS , thereby, the power
information to the relays simultaneously with a fixed transmission power , thereby, the power
shared by each source node is PS /N. In the second time slot, the relays harvest energy from the RF
shared by each source node is / . In the second time slot, the relays harvest energy from the RF
signals to amplify and forward (AF) the information received to the destination. The channel state
signals to amplify and forward (AF) the information received to the destination. The channel state
information (CSI) is known to all legitimate users.
information (CSI) is known to all legitimate users.

E E3
dSE E1 T

S1 EZ
Energy harvesting at R
E2
Information transmission

S2 R1 Information transmission R to D
S to R
D T/2 T/2
R2
S3
.
. . dRD Energy harvesting receiver
. . YR
Relay
. S Source
Information receiver
RM R Relay
dSR D Destination
SN E Eavesdropper

(a) (b)
Figure 1. Schemes of CCS-AF network and PSR protocol. (a) CCS-AF network; (b) PSR protocol.
Figure 1. Schemes of CCS-AF network and PSR protocol. (a) CCS-AF network; (b) PSR protocol.

The channels in the CCS-AF network are represented as follows. We denote the matrix
The channels
in the CCS-AFchannel
as the source-to-relay network are represented
matrix; the vector as follows.

is We denote the matrix
the relay-to-destination
H R M N as the source-to-relay channel matrix; the vector H R M 1 is the relay-to-destination
channel
SR vector; the matrix is the source-to-eavesdropper
DR channel matrix. The distances
channel vector; the matrix H R Z M is the source-to-eavesdropper channel matrix. The distances
from the sources to relays, from relays to the destination and from sources to eavesdroppers are
SE
from the by
denoted sources , to relays,
, and from relays to the destination and from sources to eavesdroppers are
, respectively.
denoted by dSR , d RD
To measure the, and dSE , respectively.
security level of the communication network, the secrecy capacity is usually
To measure the security
defined as the maximum difference level of the communication
between the mutualnetwork, the secrecy
information capacity
of the main is usually
channels and
defined as the maximum difference between the mutual
eavesdropper channels [21], and could be formulated as Equation (1): information of the main channels and
eavesdropper channels [21], and could be formulated as Equation (1):
= max[ ( ; ) ( ; )] = ( ) (1)
in which CS = max[ Iis
is the secrecy capacity, ( x;the
y) input z)]+ =at(Csources,
I ( x;signal +
D CE ) and are output signals(1)
at
destination and eavesdroppers, respectively. ( ; ) is the mutual information of the sources and
in which CS is the secrecy capacity, x is the input signal at sources, y and z are output signals at
destination; ( ; ) is the mutual information of the sources and eavesdroppers; is the capacity
destination and eavesdroppers, respectively. I ( x; y) is the mutual information of the sources and
for the transmission between sources and destination; and is the capacity at eavesdropper.
destination; I ( x; z) is the mutual information of the sources and eavesdroppers; CD is the capacity for
the Scheme between sources and destination; and CE is the capacity at eavesdropper.
transmission
CCS-AF
CCS-AF first time slot, denote = [ , , ,
Scheme
In the ] as the original signals to be transmitted by the
source nodes. The power constraints of the sources and relays are and , respectively. The
In the first time slot, denote X = [ x1 , x2 , . . . , x N ] as theoriginal signals to be transmitted by the
channel fading between sources and relays is , where [ ] , ~ (0, ) . This is
source nodes. The power constraints of the sources and relays are PS and PR , respectively. The channel
incoherent with the identity matrix and satisfies the restricted isometry property
fading between sources and relays is HSR R M N , where [HSR ]i,j N 0, M1 . This is incoherent

(RIP) with a high
probability
with as long
the identity matrixas andsatisfies
c log(the/restricted
) [15], where is the number
isometry property of arelays,
(RIP) with is a small
high probability as
constant, is the sparsity, is the number of sources. = is the
long as M cK log( N/K ) [15], where M is the number of relays, c is a small constant, K is the sparsity, transmission
Nmatrix
is theofnumber
the source-to-relay
,
of sources. Hchannels,
= SR HSR the M N represents
theismatrix thematrix
transmission path loss
of the thesource-to-relay
channels. The
element indicates the path loss from the th source node to the th relay and i,j can be calculated
channels,
,
the /matrix SR represents the path loss the channels. The element SR indicates the path
as = , where is the distance between the S and , and is the path loss
Entropy 2017, 19, 462 4 of 11

i,j
loss from the jth source node to the ith relay and can be calculated as SR = d u/2
S R , where dS j Ri is the
j i
distance between the S j and Ri , and is the path loss component. The signals received by the relays
is represented by YR = HX + N0 , where N0 R M1 is an additive white Gaussian noise (AWGN)
vector with variance n20 .
In the second time slot, the selected relays cooperate to amplify and forward the received
signals to the destination. HRD R M indicates the channel between relays and the destination,
then H = RD HRD is the transmission matrix between relays and destination, and RD is the path
loss of the relay-to-destination channels. Therefore, the signal received at the destination could be
represented by Equation (2):

YD = GYR + W0 = G(HX + N0 ) + W0 (2)

r
PR /M
where is a diagonal matrix, the entry ii = Ps 2 in denotes the amplification
N N
j=1 | Aij | +n0
2

coefficient of the ith relay. The noise between relays and destination is W0 R M1 , which is a AWGN
vector with variance w2 0 .
Let = (G)H, then Equation (2) can be rewritten as YD = X + (GN0 + W0 ). As the
channel matrix H satisfies the RIP property and G is a diagonal matrix, therefore, satisfies the
requirements of the RIP property and can be used as the secure measurement matrix to encrypt the
transmitted information. The channel matrix from sources to the destination is the compressive matrix
in CS theory, and also it is the measurement matrix.
The destination recovers the source signals X from YD by solving the convex optimization problem:

min ||X ||l1 s.t. ||YD X ||l2 (3)

where is the upper bound of the noise magnitude and X are the signals reconstructed at the destination.

3. Power Splitting Relaying Protocol

In the power splitting relaying protocol, the total transmission time T from sources to the
destination is divided into two equal parts, i.e., T/2. During the first slot, the relays harvest energy and
process information simultaneously. The RF power P received at each relay is divided by the power
splitting ratio (0 < < 1), which means P is allocated for EH and (1 )P is used for information
processing. In the second slot, the relays amplify and forward the received signal to the destination
using the harvested energy. The channel matrix between the sources and relays is:

H = [h1 h2 hM ]T (4)
 
where hi = hi,1, hi,2, . . . , hi,j , . . . , hi,N , 1 i M, and hi,j (1 j N) indicates the channel between Sj
and Ri .
In the first slot, denoting the signals received by the EH receiver as YRH , then it can be expressed
as Equation (5):

YRH = (HX + N0 ) (5)

The energy harvested by the EH receiver is as follows:

 
T P
ERi = S ||h ||2i 1 + n20 (6)
2 N

Thereby, the power of the harvested energy at Ri could be obtained and shown in Equation (7):
 
PS 2 2
PRi = ||h ||i 1 + n0 (7)
N
Entropy 2017, 19, 462 5 of 11

Denoting he signals at the information receiver as YRI , then it can be computed as Equation (8).
q
YRI = (1 )(HX + N0 ) (8)

Now, the signals received at the eavesdroppers in the first slot can be written as:

YE = HSE X + N0 (9)

In the second slot, the signal transmitted by the relays is XR = YR , where is a diagonal matrix.
The diagonal elements in can be represented as Equation (10):
v
u PRi
ii = t (10)
u  
(1 ) PS
N ||h ||2i 1 + n20

where ii is the amplification coefficient of the ith relay. The noise between relays and destination is
W0 R M1 , which is a AWGN vector with variance w2 0 .Thereby, the signal received at the destination
can be formulated as:
q
YD = GYRI + W0 = (1 ) G (HX + N0 ) + W0 (11)

The secrecy capacity is the difference between the capacity of main channels and the eavesdropper
channels, and could be calculated as the sum of the reliable information received by the destination.
In order to obtain the channel capacity, we transform the M N channel matrix H to a M 1 parallel
channel vector by using singular value decomposition (SVD) approach [16]. Let H = UVH , then
we get YRI = (1 ) UVH X + N0 , where is a diagonal matrix, U C M M and V C M N
p

are both unitary matrixes, and VH is the conjugate transpose of V. Let Y0RI = UH YRI , X0 = VH X,
N00 = UH N0 , then the information received by relays is equivalent to Y0RI = (1 )(X 0 + N00 ).
p

Thereby, Equation (11) can be rewritten as:

q q
Y0D = GY0RI + W0 = (1 ) GX + (1 )GN00 + W0 (12)

According to the property of the unitary matrix, U and V wont change the power of YRI , X,
N0 , which means PX0 = PX , PN0 = PN0 , PY0 = PYR . Thereby, the power matrixes of the received
0 R
information and signal noise at the destination can be easily obtained after SVD, and could be
formulated as Equations (13) and (14):

(1 ) PS
Pi = | Gii |2 2ii 2i (13)
N

i2 = w2 0 + (1 )n20 | Gii |2 2ii (14)

where i indicates the ith parallel Gaussian channel, i is the ith diagonal entry of .
Then the channel capacity C of the main channel could be computed by Equation (5):

1 M
log2 1 + Pi /i2


C =
2 i =1
M PS
1 2 2
N | Gii | i
= log2 1 +

2
2 i =1 w
0 + 2 | G |2
(15)
n0 ii
(1) 2ii

M PS
1 2 2
N | Gii | i
= log2 1 + 2

2 i =1 w
0 2 2
+ n0 | Gii |
Entropy 2017, 19, 462 6 of 11

in which, n20 and w2 0 are the noise powers of two time slots.
The signal received by eavesdropper Et could be expressed as:

YEt = Bt X + NEt (16)

where Bt is the channel matrix between the sources and Et , NEt is the noise at Et . The power of the
Entropy 2017, 19, 462  2 6 of 10
signals and noise received by Et is Pte = PNS  Bj,t  and E2t , respectively.
The channel capacity of the eavesdropper channel (from S j to Et ) is:
1 ,
, = log 1 +PS  2 ! (17)
e 12 N B j,t
Cj,t = log2 1 + (17)
2 E2t
where . is the path loss between and . The security loss caused by the eavesdroppers is as
where Bj.t is the path loss between S j and Et . The security loss caused by the eavesdroppers is
follows:
as follows:
N
CE == Cje (18)
j =1

whereCje ==max
where max (Cj,1
e ,,
, . . ,. , C
e,
, j,t , ., . . , Cj,Z
e, and e ) indicates the capacity of the channel from
), and, C(1
j,t (1 t Z) indicates the capacity of the channel
fromtoS j to. Thereby, the the
Et . Thereby, secrecysecrecy capacitycapacity of the CCS-AF
of the CCS-AFnetwork can can
network be obtained by Equation
be obtained (1) and
by Equation (1)
represented
and represented as as =CS =C . CE .
Ourscheme
Our schemecan canachieve
achieveperfect perfectsecrecy secrecyunder
underthe thecondition
conditionof
ofbounded
boundedcomputation
computationcapability
capability
ofofeavesdroppers.
eavesdroppers.

4.4.Simulations
Simulationsand
andAnalysis
Analysis
In
Inthis
thissection,
section, we
weconsider
consider aaCCS-AF
CCS-AFwireless
wirelessnetwork
networkwith with1515sources,
sources,four fourrelays,
relays,two two
eavesdroppers
eavesdroppersand andone
onedestination.
destination.TheTheparameters
parametersare areset
setasasfollows:
follows:the
thepath
pathloss
lossofofthethechannels
channels
isis == 4; 2 2
4;the
thenoises
noisesare
areassumed
assumedtotobebeGaussian
Gaussianwith variancesn0 = =w0 ==10
withvariances 10 dbm.
dbm. In In order
ordertoto
simplify the simulation and study the impact of the relays position on the system
simplify the simulation and study the impact of the relays position on the system secrecy capacity, asecrecy capacity,
asimple
simplemodel
modelisisshown
shownininFigure
Figure2,2,where
whereallallnodes
nodes are
are setset
onon a straight
a straight line.
line. AllAll
thethe source
source nodes
nodes are
are considered to be located at the same position, thereby the effect on the secrecy
considered to be located at the same position, thereby the effect on the secrecy capacity caused by thecapacity caused
by the distances
distances amongamong the source
the source nodes isnodes is ignored.
ignored. Theand
The relays relays and eavesdroppers
eavesdroppers are treated arein treated
the same in
the same way. The distance between sources and destination are normalized to
way. The distance between sources and destination are normalized to a unit value, thereby, the a unit value, thereby,
the coordinates
coordinates of the
of the source
source nodes,
nodes, relays,
relays, eavesdroppers,
eavesdroppers, andand destination
destination areare (0, (0), (d, SR
(0, 0), 0),, (0), (d
, 0) , 0)
SEand
and (1, respectively.
(1, 0), 0), respectively.

(0,0) (dSR,0) (dSE,0) (1,0)

SN RM EZ D

Figure 2. Simplified model of CCS-AF network.

Figure 2. Simplified model of CCS-AF network.
In the first simulation, we study how the secrecy capacity changes with . The power splitting
ratioIn theis first simulation,
0.5, the we study how
energy conversion the secrecy
efficiency capacity
is 1, and changes fromdSR
varieswith . The
0 to powerconditions
1. Three splitting
ratio
that the
is 0.5, is 10 W, 10 W and 10 W, are considered,
the energy conversion efficiency is 1, and d SR varies from
respectively. Figure 3 showsthat
0 to 1. Three conditions the
the 4
PS is 10 results. 3
W, 10TheWsecrecy 2
and 10capacity
W, areunder
considered,
simulation each respectively. Figure 3 shows
decreases monotonously andthe simulationto
approaches
results.
0 with The secrecy capacity
the increase of under
. This each be
could PS explained
decreases monotonously
that when and approaches
increases, the EH to 0power
with the
and
increase of
informationSR d . This could be explained that
received by relays decrease rapidly, which when d increases, the EH power and information
SR obeys Equations (6) and (8). The reason for
received
why does bytherelays decrease
network rapidly,
achieves thewhich
higher obeys Equations
secrecy capacity(6)whenand (8). =The10reason
W thanfor why does the
the other two
network
conditions achieves
will be the higher secrecy
analyzed capacity
in the fourth when PS = 103 W than the other two conditions will be
simulation.
analyzed
In thein the
secondfourth simulation.
simulation, we study how the secrecy capacity changes with the power-splitting
ratioIn the second
under simulation,
several we of
conditions study .how the secrecy
The sources power capacity
is = 10 W,
changes with=the 1, power-splitting
and is set to
ratio under several conditions of d . The sources power is P = 10 2 W, = 1, and d is set to4.
0.3, 0.5 and 0.8, respectively. The secrecy
SR rates of the networks are S calculated and plotted SR in Figure
On the one hand, for each situation of , the system secrecy capacity with increases first and
then decreases, and achieves the highest at a certain point.
Entropy 2017, 19, 462 7 of 11

0.3, 0.5 and 0.8, respectively. The secrecy rates of the networks are calculated and plotted in Figure 4.
On the one hand, for each situation of dSR , the system secrecy capacity with increases first and then
decreases, and achieves the highest at a certain point.
Entropy 2017, 19, 462 7 of 10
Entropy 2017, 19, 462 7 of 10
30

30 Ps=10-4
25 Ps=10-3
Ps=10-4
25 Ps=10-2
Ps=10-3

20
Ps=10-2
bps/Hz

20
bps/Hz

15
capacity

15
capacity

10
Secrecy

10
5
Secrecy

5
0

0
-5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
-5 dSR
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
dSR
Figure 3. Secrecy capacity versus
Figure 3. Secrecy capacity versus dSR when 4, =
when M == 4, = 1, and =
1,and = 0.5.
0.5.
Figure 3. Secrecy capacity versus when = 4, = 1, and = 0.5.
This is because when is small, the relays get little power for energy harvesting, which makes
This is is
This because
the transmission when
power
because when is small,
at relays thethe
relays
un-enough
is small, getget
and
relays little
results power
inpower
little lowerforfor
energy
secrecy
energy harvesting,
capacity. which makes
In contrary,
harvesting, which whenthe
makes
transmission
the power
is transmission at
too large, the power relays
power at un-enough
forrelays
the informationand results
un-enoughtransmission in
and results in lower
will secrecy
be too
lower capacity.
little and
secrecy In
muchIn
capacity. contrary,
harvested when
when
contrary,power
is will
tooislarge,
be the
toowasted. power
This
large, the for
could
power the information
forwell transmission
explain that transmission
the information will
the secrecy capacity be too little
achieves
will be too and much
the much
little and highestharvested
when power
harvested power
takes
will bebe
awill wasted.
moderate
wasted. This
value. could
The
This well
reason
could well explain
for that
why the
explain the
thesecrecy
model
that capacity
capacity achieves
with smaller
secrecy could get
achieves thehigher
the highest
highest whenrate
secrecy
when takes
has a
takes
moderate
been value.
explained
a moderate The reason
in the
value. Thefirst for why
simulation.
reason the model with smaller
for why the model with smaller d SR could get higher secrecy rate has
could get higher secrecy rate has been
explained in the first
been explained simulation.
in the first simulation.
14

14
12

12

10
bps/Hz

10
bps/Hz

8
capacity

8
6
capacity
Secrecy

6
4
Secrecy

dSR=0.3
4
2 ddSR=0.5
=0.3
SR
ddSR=0.8
=0.5
2 SR
0 dSR=0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0 power-splitting ratio ()
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
power-splitting ratio ()
Figure 4. Secrecy capacity versus with various values of .
Figure4.4. Secrecy
Figure Secrecy capacity versus with various
versus with various values of
of dSR..
In the third simulation, we study the change of the secrecy capacity with the energy conversion
efficiency
In the .thirdvaries from 0 to
simulation, we1,studyis set
theto 0.5, and
change of the
the other
secrecy parameters are set
capacity with thethe same conversion
energy as the first
In the third simulation, we study the change of the secrecy capacity with the energy conversion
simulation.
efficiency .Thevaries
simulation
from 0results
to 1, are is shown in and
set to 0.5, Figure
the5.other
It is parameters
obvious thatare forset
each
the same , the secrecy
as the first
efficiency
capacity . varies fromprotocol
ofthe 0 to 1, is set to 0.5, and, the
andother parameters a are set the same asisthe first
simulation. The EH-AF
simulation resultsincreases
are shown within Figure 5.approaches
It is obvious tothat constant
for eachwhen high
, the secrecy
simulation.
enough. The simulation results are shown in Figure 5. It is obvious that for each d , the
SR could secrecy
capacity This
of theis easy
EH-AF to explain,
protocolhigher energy
increases withconversion efficiency means
, and approaches higher energy
to a constant when be
is high
capacity
obtained of the EH-AF
without
enough. This protocol
improving
is easy to explain, increases
thehigher with
powerenergy ,
splittingand approaches
ratio. Thereby
conversion to a constant
more
efficiency energy
means when is
can energy
higher high
be obtained enough.
could be to
This is easythe
transmit
obtained to received
explain,
without higher energy
information
improving the to conversion
the
power splittingefficiency
destination. We can
ratio. means higher
also see
Thereby energy
thatenergy
more the could
secrecy
can berates beincrease
obtained
obtained to
slowly
transmit with when information
the received is higher thanto the0.7,destination.
0.5 and 0.1We under
can the
alsoconditions
see that the that
secrecyisrates
0.3, increase
0.5 and
0.8, respectively.
slowly with when This phenomenon
is higher thanhas 0.7,
an important
0.5 and 0.1 instructive
under themeaning in engineering,
conditions that as it0.5
is 0.3, means
and
we
0.8,do not have toThis
respectively. seekphenomenon
EH devices with has anhigh with an
important unnecessarily
instructive meaninghighineconomic
engineering,cost.as it means
we do not have to seek EH devices with high with an unnecessarily high economic cost.
Entropy 2017, 19, 462 8 of 11

without improving the power splitting ratio. Thereby more energy can be obtained to transmit the
received information to the destination. We can also see that the secrecy rates increase slowly with
when is higher than 0.7, 0.5 and 0.1 under the conditions that dSR is 0.3, 0.5 and 0.8, respectively.
This phenomenon has an important instructive meaning in engineering, as it means we do not have to
seek EH devices with high with an unnecessarily high economic cost.
Entropy 2017, 19, 462 8 of 10

Entropy 2017, 19, 462 8 of 10

14
dSR=0.3
14 dSR=0.5
12
dSR=0.3
dSR=0.8
12 dSR=0.5

10
dSR=0.8
bps/Hz

10
8
bps/Hz
capacity

8
6
capacity
Secrecy

6
4
Secrecy

4
2

2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
energy conversion efficiency ()
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure5.5.Secrecy
Figure Secrecy capacity energy with
capacity versus
versus conversion efficiency () when
different
with different dSR when M == 4,
4, and = 0.5.
and = 0.5.
Figure 5. Secrecy capacity versus with different when = 4, and = 0.5.
The effect of relay number on the system secrecy capacity is studied in the fourth simulation,
andThethree
effect of relay are
scenarios number on the system
considered, where secrecy is setcapacity
to 10 W, is studied
10 W,inand 10
the fourth W,simulation,
respectively.and
The effect of relay number on the system
PS isnumber secrecy
4 capacity
3 is studied
2 in the fourth simulation,
= 0.5,
three scenarios
= 0.5, are = 1,considered,
= 0.5, and where
the relay set to 10varies 10 3W,
W,from to and 10 simulation
13. The result isdSR
W, respectively. shown
and three scenarios are considered, where is set to 10 W, 10 W, and 10 W, respectively.
=in1,Figure
= 0.5, and the relay number
6. It could be seen that when varies is large enough, the model that contains more relays will 6.
from 3 to 13. The simulation result is shown in Figure
= 0.5, = 1, = 0.5, and the relay number varies from 3 to 13. The simulation result is shown
It could be aseen
presents that when
significant PS is large
advantage. enough,
As shown the model
in Figure 6, whenthatthecontains more relays
relay number > 6, will presents a
the higher
in Figure 6. It could be seen that when is large enough, the model that contains more relays will
significant advantage.
is, the higher secrecyAs shownwill
capacity in Figure 6, when
be obtained. theisrelay
This number
because M > 6, of
the number higher PS is, the
theeavesdroppers higher
is fixed,
presents a significant advantage. As shown in Figure 6, when the relay number > 6, the higher
secrecy
thoughcapacity
higherwill will
be obtained.
improve This is because
the channel the number
capacity of eavesdroppers
of eavesdroppers, but theisincreased
fixed, though
relayshigher
will
is, the higher secrecy capacity will be obtained. This is because the number of eavesdroppers is fixed,
obtain
PS though more secrecy
will improve capacity
the will
channel for the of
capacity main channels under
eavesdroppers, butenough energy. Thereby
the increased relays willhigher secrecy
obtain more
higher improve the channel capacity of eavesdroppers, but the increased relays will
capacity
secrecy couldfor
capacity bethe
obtained.
main It could under
channels be seenenough
in Figure 6 that
energy. when the
Thereby higherrelaysecrecy
number = 4, could
capacity the
obtain more secrecy capacity for the main channels under enough energy. Thereby higher secrecy
network achieves
be capacity
obtained. It could the
be highest
seen in secrecy6capacity
Figure that when when the = 10number
relay W, which
M = could
4, the explain
network why the
achieves
could be obtained. It could be seen in Figure 6 that when the relay number = 4, the
network
highest achieves
thenetwork secrecy the highest secrecy
PS =rate when
3 W, which = 10 W explainin the first simulation.
achievescapacity when
the highest secrecy 10
capacity when could = 10 W, which why the network
could explainachieves
why the the
highest secrecy rate when P = 10 3 W in the first simulation.
network achieves the highest S
12
secrecy rate when = 10 W in the first simulation.
Ps=10-4
12
Ps=10-3
10 Ps=10-4
Ps=10-2
Ps=10-3

10
Ps=10-2
bps/Hz

8
Corresponding to the

part circled in Figure 1.

bps/Hz

8
capacity

Corresponding to the
6
part circled in Figure 1.
capacity

6
Secrecy

4
Secrecy

4
2

2
0
3 4 5 6 7 8 9 10 11 12 13
number of relay (M)
0
3 4 5 6 7 8 9 10 11 12 13
Figure 6. Secrecy capacity versus number
withof different
relay (M) when = 1, and = 0.5.

Figure 6. Secrecy capacity versus with different when = 1, and = 0.5.

Figure
In the last 6. Secrecy
simulation, wecapacity versus
study the M with
change differentcapacity
of secrecy = 1, and
PS whenwith = 0.5.
. Totest the efficiency of
the PSR protocol sufficiently, the AF protocols in [28] is taken as a comparison. In both approaches,
In the last simulation, we study the change of secrecy capacity with . To test the efficiency of
the varies from 10 W to 10 W, and the simulation is operated under three conditions where
the PSR protocol sufficiently, the AF protocols in [28] is taken as a comparison. In both approaches,
is 0.3, 0.5, and 0.8, respectively. The total power in the AF protocol is = 10 W, and the energy
the varies from 10 W to 10 W, and the simulation is operated under three conditions where
conversion efficiency in PSR protocol is = 1. Figure 7 presents the simulation results. Under each
is 0.3, 0.5, and 0.8, respectively. The total power in the AF protocol is = 10 W, and the energy
circumstance of , the secrecy capacity of PSR increases with and approaches to a constant. As
conversion efficiency in PSR protocol is = 1. Figure 7 presents the simulation results. Under each
Entropy 2017, 19, 462 9 of 11

In the last simulation, we study the change of secrecy capacity with PS . To test the efficiency of the
PSR protocol sufficiently, the AF protocols in [28] is taken as a comparison. In both approaches, the PS
varies from 103 W to 102 W, and the simulation is operated under three conditions where dSR is 0.3,
0.5, and 0.8, respectively. The total power in the AF protocol is P = 102 W, and the energy conversion
efficiency in PSR protocol is = 1. Figure 7 presents the simulation results. Under each circumstance
of dSR , the
Entropy 19, 462 capacity of PSR increases with PS and approaches to a constant. As we can see, when
2017,secrecy 9 of 10
dSR = 0.3, the PSR protocol in this paper always outperforms the AF protocol. When dSR = 0.5 and
the PSR = 3 W and 8.5 103 W,
0.8,
When 0.5 and
protocol performs
0.8, thethePSR
better only when
protocol PS is higher
performs 10when
than 2only
the better is higher than
2 10 WThough
respectively. and 8.5 the PSR
10 protocol
W, respectively.
is not so good as thethe
Though AFPSR
protocol in some
protocol stages
is not when as
so good dSRthe
is 0.5
AF
and 0.8, the inferiority
protocol in some stages whenis very small. This means the PSR protocol proposed in this paper could
is 0.5 and 0.8, the inferiority is very small. This means the PSR get a
similar
protocol or proposed
much higher secrecy
in this capacity
paper could by getconsuming
a similar orless
muchpower than secrecy
higher AF, though this advantage
capacity by consumingwill
be weakened
less power than with thethough
AF, increase dSR .
of advantage
this will be weakened with the increase of .

14

12

10
Secrecy capacity bps/Hz

4
PSR-dSR=0.3
2 PSR-dSR=0.5
PSR-dSR=0.8
0
AF-dSR=0.3

-2 AF-dSR=0.5
AF-dSR=0.8
-4
1 2 3 4 5 6 7 8 9 10
Ps -3
x 10

Figure 7.
Figure 7. Secrecy
Secrecy capacity
capacity versus
versus the
the P when
whentotal powerP ==10
totalpower 10 Wand
2 W andM ==4.4.
S

5.5.Conclusions
Conclusions
CS technology
CS technology isis aa novel
novel approach
approach to to enhance
enhance thethe physical-layer
physical-layer security
security ofof cooperative
cooperative
communication networks in the presence of eavesdroppers, and EH can help
communication networks in the presence of eavesdroppers, and EH can help relays work in continuous relays work in
continuous mode in specific environments. In this paper, we analyze the secrecy
mode in specific environments. In this paper, we analyze the secrecy capacity of the CCS-AF wireless capacity of the
CCS-AF wireless
networks with EHnetworks withthe
relays using EHPSR
relays using the
protocol. FivePSR protocol. revealed
simulations Five simulations
how therevealed how the
secrecy capacity
secrecy capacity reacts to the change of some key factors in this network, and verified the
reacts to the change of some key factors in this network, and verified the EH is to be capable of improve EH is to be
capable
the energyof utilization
improve the energy utilization
efficiency. Thus, theyefficiency.
have greatThus, they
guiding have greatfor
significance guiding significance
researchers for
to obtain
researchers to obtain the highest secrecy capacity or design the most appropriate solution
the highest secrecy capacity or design the most appropriate solution in related researches. Our future in related
researches.
work Ouron
will focus future
how towork will focus
improve on how to improve
the physical-layer securitythe physical-layer
of multi-hop security of multi-hop
communication networks
communication
through applyingnetworks through applying EH approaches.
EH approaches.
Acknowledgments:This
Acknowledgments: Thiswork
workisissupported
supportedby
bythe
theNational
NationalNatural
NaturalScience
ScienceFoundation
Foundationof
ofChina
Chinaunder
underGrant
Grant
No.61571323.
No. 61571323.
Author Contributions:Shuai
AuthorContributions: ShuaiChang
ChangandandXiaomei
XiaomeiFuFuconceived
conceivedthe
thestudy;
study;Jialun
JialunLi
Liperformed
performed the
the experiments;
experiments;
Shuai
ShuaiChang
Changand
andJialun
JialunLi
Lianalyzed
analyzedthe
thedata;
data;Shuai
ShuaiChang,
Chang,Jialun
JialunLi,
Li,and
andLiang
LiangZhang
Zhangwrote
wrotethe
thepaper.
paper.
Conflicts of Interest: The authors declare no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Khodakarami, H.; Lahouti, F. Link adaptation for physical layer security over wireless fading channels. IET
Commun. 2012, 6, 353362.
2. Zhuang, W.; Ismail, M. Cooperation in wireless communication networks. IEEE Wirel. Commun. 2012, 19,
1020.
3. Zou, Y.; Wang, X.; Shen, W. Optimal relay selection for physical-layer security in cooperative wireless
networks. IEEE J. Sel. Areas Commun. 2013, 31, 20992111.
Entropy 2017, 19, 462 10 of 11

References
1. Khodakarami, H.; Lahouti, F. Link adaptation for physical layer security over wireless fading channels.
IET Commun. 2012, 6, 353362. [CrossRef]
2. Zhuang, W.; Ismail, M. Cooperation in wireless communication networks. IEEE Wirel. Commun. 2012, 19,
1020. [CrossRef]
3. Zou, Y.; Wang, X.; Shen, W. Optimal relay selection for physical-layer security in cooperative wireless
networks. IEEE J. Sel. Areas Commun. 2013, 31, 20992111. [CrossRef]
4. Jing, Y.; Jafarkhani, H. Single and multiple relay selection schemes and their achievable diversity orders.
IEEE Trans. Wirel. Commun. 2009, 8, 14141423. [CrossRef]
5. Lai, L.; El Gamal, H. The relayeavesdropper channel: Cooperation for secrecy. IEEE Trans. Inf. Theor. 2008,
54, 40054019. [CrossRef]
6. Zhang, R.; Song, L.; Han, Z.; Jiao, B. Physical layer security for two-way untrusted relaying with friendly
jammers. IEEE Trans. Veh. Technol. 2012, 61, 36933704. [CrossRef]
7. Zhang, J.; Gursoy, M.C. Relay beamforming strategies for physical-layer security. In Proceedings of the
2010 44th Annual Conference on Information Sciences and Systems, Princeton, NJ, USA, 1719 March 2010;
pp. 16.
8. Ju, P.; Song, W.; Zhou, D. Survey on cooperative medium access control protocols. IET Commun. 2013, 7,
893902. [CrossRef]
9. Popovski, P.; Simeone, O. Wireless secrecy in cellular systems with infrastructure-aided cooperation.
IEEE Trans. Inf. Forensics Secur. 2009, 4, 243256. [CrossRef]
10. Nasir, A.A.; Zhou, X.; Durrani, S.; Kennedy, R.A. Relaying protocols for wireless energy harvesting and
information processing. IEEE Trans. Wirel. Commun. 2013, 12, 36223636. [CrossRef]
11. Zhou, X.; Zhang, R.; Ho, C.K. Wireless information and power transfer: Architecture design and rate-energy
tradeoff. IEEE Trans. Commun. 2013, 61, 47544767. [CrossRef]
12. Varshney, L.R. Transporting information and energy simultaneously. In Proceedings of the 2008 IEEE
International Symposium on Information Theory (ISIT), Toronto, ON, Canada, 611 July 2008; pp. 16121616.
13. Zhang, M.; Liu, Y. Energy harvesting for physical-layer security in OFDMA networks. IEEE Trans. Inf.
Forensics Secur. 2016, 11, 154162. [CrossRef]
14. Zhang, R.; Ho, C.K. MIMO broadcasting for simultaneous wireless information and power transfer.
IEEE Trans. Wirel. Commun. 2013, 12, 19892001. [CrossRef]
15. Salem, A.; Hamdi, K.A.; Rabie, K.M. Physical layer security with RF energy harvesting in AF multi-antenna
relaying networks. IEEE Trans. Commun. 2016, 64, 30253038. [CrossRef]
16. Son, P.N.; Kong, H.Y. Cooperative communication with energy-harvesting relays under physical layer
security. IET Commun. 2015, 9, 21312139. [CrossRef]
17. Hoang, T.M.; Duong, T.Q.; Vo, N.S.; Kundu, C. Physical layer security in cooperative energy harvesting
networks with a friendly jammer. IEEE Wirel. Commun. Lett. 2017, 6, 174177. [CrossRef]
18. Lei, H.; Xu, M.; Ansari, I.S.; Pan, G.; Qaraqe, K.A.; Alouini, M.S. On Secure Underlay MIMO Cognitive Radio
Networks with Energy Harvesting and Transmit Antenna Selection. IEEE Trans. Green Commun. Netw. 2017,
1, 192203. [CrossRef]
19. Patterson, S.; Eldar, Y.C.; Keidar, I. Distributed compressed sensing for static and time-varying networks.
IEEE Trans. Signal Process. 2014, 62, 49314946. [CrossRef]
20. Martinez, J.; Mejia, J.; Mederos, B. Compress sensing for wireless sensor networks using gossip pairwise
algorithm and optimization algorithms. In Proceedings of the IEEE Ecuador Technical Chapters Meeting,
Guayaquil, Ecuador, 1214 October 2016; pp. 15.
21. Candes, E.J.; Romberg, J.K.; Tao, T. Stable signal recovery from incomplete and inaccurate measurements.
Commun. Pure Appl. Math. 2006, 59, 12071223. [CrossRef]
22. Rachlin, Y.; Baron, D. The secrecy of compressed sensing measurements. In Proceedings of the 46th Annual
Allerton Conference on Communication, Control, and Computing, Urbana-Champaign, IL, USA,
2326 September 2008; pp. 813817.
23. Mayiami, M.R.; Seyfe, B.; Bafghi, H.G. Perfect secrecy using compressed sensing. Mathematics 2010.
[CrossRef]
Entropy 2017, 19, 462 11 of 11

24. Agrawal, S.; Vishwanath, S. Secrecy using compressive sensing. In Proceedings of the 2011 IEEE Information
Theory Workshop, Paraty, Brazil, 1620 October 2011; pp. 563567.
25. Reeves, G.; Goela, N.; Milosavljevic, N.; Gastpar, M. A compressed sensing wire-tap channel. In Proceedings
of the 2011 IEEE Information Theory Workshop, Paraty, Brazil, 1620 October 2011; pp. 548552.
26. Barcelo-Llado, J.E.; Morell, A.; Seco-Granados, G. Amplify-and-forward compressed sensing as an
energy-efficient solution in wireless sensor networks. IEEE Sens. J. 2014, 14, 17101719. [CrossRef]
27. Barcelo-Llado, J.E.; Morell, A.; Seco-Granados, G. Amplify-and-forward compressed sensing as a physical-layer
secrecy solution in wireless sensor networks. IEEE Trans. Inf. Forensics Secur. 2014, 9, 839850. [CrossRef]
28. Fu, X.; Cui, Y. Multi-source and multi-relay cooperative system based on compressed sensing. Electron. Lett.
2015, 51, 18281830. [CrossRef]

2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).