Joint Relay and Jammer Selection for Secure

Two-Way Relay Networks

Jingchao Chen

, Rongqing Zhang

, Lingyang Song

, Zhu Han

, Bingli Jiao

School of Electronics Engineering and Computer Science, Peking University, Beijing, China.

Electrical and Computer Engineering Department, University of Houston, Houston, TX, USA.
AbstractIn this paper, we investigate joint relay and jammer
selection in two-way cooperative networks, consisting of two
sources, one eavesdropper, and a number of intermediate nodes,
with secrecy constraints. Specically, the proposed schemes select
two or three intermediate nodes to enhance security against
the malicious eavesdropper. The rst selected node operates
in the conventional relay mode and assists the two sources
to exchange data with each other in the amplify-and-forward
protocol. The second and third selected nodes are used in
different communication phases as jammers in order to create
intentional interference upon the eavesdropper. We nd that in
a scenario where the relay and jamming nodes are randomly
and sparsely distributed, the proposed schemes with cooperative
jamming outperform the conventional non-jamming schemes
within a certain transmitted power range. We also nd that,
in a scenario where the intermediate nodes gather as a close
cluster, the cooperative jamming schemes may be less effective
than their non-jamming counterparts. Therefore, we introduce
a hybrid scheme to switch between jamming and non-jamming
modes. Simulation results validate our theoretical analysis that
the hybrid switching scheme further improves the secrecy rate.
I. INTRODUCTION
Traditionally, security in wireless networks has been mainly
focused on higher layers using cryptographic methods. Pio-
neered by Aaron Wyners work [1], which established fun-
damental results of creating perfectly secure communications
without relying on private keys, physical layer based security
has drawn increasing attention recently. Later work in [2]
[4] studied the secrecy capacity of the Gaussian wiretap
channel, and extended Wyners approach to the transmission of
condential messages over the broadcast and wireless fading
channels. In [5][7], several cooperative jamming schemes
were investigated for different scenarios to increase the secrecy
capacity of networks with secrecy constraints.
Recently, two-way relay channel [8][10] has been well
studied for its potential application to cellular networks and
peer-to-peer networks. In a cooperative network, the efciency
of relay or jammer selection has a great impact on the
performance of the whole system. In [11], a relay selection
scheme was proposed for two-way networks with multiple
relays, which maximized the worse receive signal-to-noise
ratio (SNR) of the two sources. In [12], several relay selection
techniques were proposed in one-way cooperative networks
with secrecy constraints. Although cooperative networks have
received much attention by far, the physical layer security
issues with secrecy constraints in two-way relay networks have
not yet been well investigated.
This paper proposes a scheme which can implement infor-
mation exchange against eavesdroppers in two-way coopera-
tive networks, consisting of two sources, one eavesdropper,
and a number of intermediate nodes, with secrecy constraints.
Specically, an intermediate node is selected to operate in the
conventional amplify-and-forward (AF) relay mode and assists
the sources to deliver data to the corresponding destinations.
Meanwhile, another two intermediate nodes that perform as
jamming nodes are selected and transmit articial interference
in order to degrade the eavesdropper links in the rst and
second phase of data transmission, respectively. The principal
question here is how to select the relay and the jamming nodes
in order to increase information security and protect the source
message against eavesdroppers. Several selection algorithms
are then proposed, aiming at promoting the assistance to the
sources and the interference to the eavesdropper. The analysis
and simulation results reveal that the proposed techniques
with cooperative jamming can improve the secrecy rate of
the system by a large scale within a certain transmitted
power range. However, in some special scenarios, the proposed
jamming schemes are less efcient than the non-jamming ones.
Then we propose a hybrid scheme with intelligent switch
mechanism between the jamming and non-jamming modes,
which can overcome this problem.
The rest of this paper is organized as follows. In Section II
we describe the system model and formulate the problem.
In Section III the selection techniques are presented. In
Section IV, we provide qualitative analysis of the secrecy
performance of different selection schemes in some typical
congurations. Numerical results are shown in Section V, and
the main conclusions are drawn in Section VI.
II. SYSTEM MODEL AND PROBLEM FORMULATION
A. System Model
We assume a simple conguration consisting of two sources
S
1
and S
2
, one eavesdropper E and an intermediate node
set S
in
= {1, 2, ..., K} with K nodes. Fig. 1 schematically
presents the system model. As the intermediate nodes cannot
transmit and receive simultaneously, the communication pro-
cess is performed in two phases. In the broadcasting phase,
S
1
and S
2
transmit their data to the intermediate nodes.
Meanwhile, one node J
1
is selected from S
in
to operate as a
jammer and transmits intentional interference to degrade the
source-eavesdropper links in this phase. Since the jamming
signal is unknown to the rest nodes in S
in
, the interference
will also degrade the performance of the source-relay links, as
shown in Fig. 1. In the second phase, an intermediate node R
is selected to operate as a conventional relay and forwards the
978-1-61284-231-8/11/$26.00 2011 IEEE
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2011 proceedings
R
J1
S1
S2
E
J2
R
S1
S2
E
1st phase 2nd phase
Source nodes
Relay/Jamming nodes
Eavesdroppers
Desired information channels
Eavesdropper links
Jamming channels
Fig. 1. System model for two-way cooperative network.
source messages to the corresponding destinations. A second
jammer J
2
is selected from S
in
, for the same reason as J
1
.
Notice that S
1
and S
2
are not able to mitigate the articial
interference from the jammer, either.
During the two phases, a slow, at, and block Rayleigh
fading environment is assumed, i.e., the channel remains
static for one coherence interval and changes independently
in different coherence intervals with a variance
2
i,j
= d

i,j
,
where d
i,j
is the Euclidean distance between node i and node
j, and is the path-loss exponent. The channel gain between
node i and node j is denoted as h
i,j
, which is modeled
as a zero-mean, independent, circularly-symmetric complex
Gaussian random variable with variance
2
i,j
. Furthermore, ad-
ditive white Gaussian noise (AWGN) with zero mean and unit
variance is assumed. P
S
, P
R
and P
J
denote the transmitted
power for the source nodes, the relay node and the jamming
nodes, respectively. The jamming nodes transmit with a lower
power than the relay node [12], and thus their transmitted
power can be dened as P
J
= P
R
/L (with P
S
= P
R
), where
L 1 denotes the power ratio of the relay to the jammer.
In the rst phase, the two sources send information symbols
s
1
and s
2
, respectively, which are randomly chosen from a
constellation set. The intermediate node R and eavesdropper
E thus receive
r =
_
P
S
h
S1,R
s
1
+
_
P
S
h
S2,R
s
2
+
_
P
J
h
J1,R
j
1
+v
R
,
(1)
e
1
=
_
P
S
h
S1,E
s
1
+
_
P
S
h
S2,E
s
2
+
_
P
J
h
J1,E
j
1
+v
E
,
(2)
where v
R
and v
E
denote the noise at R and E, respectively.
In the second phase, relay R amplies its received signal
and forwards it to S
1
and S
2
, i.e., R sends t =

P
R
r, where
=
_
1
1+|hS
1
,R|
2
PS+|hS2,R|
2
PS+|hJ
1
,R|
2
PJ
. Since destination
S
i
knows s
i
(for i = 1, 2), it can cancel the self-interference.
Thus, S
1
, S
2
and eavesdropper E get
x
1
=
_
P
R
_
P
S
h
S2,R
h
R,S1
s
2
+
_
P
J
h
J2,S1
j
2
+w
1
+
_
P
R
_
P
J
h
J1,R
h
R,S1
j
1
+
_
P
R
h
R,S1
v
R
, (3)
x
2
=
_
P
R
_
P
S
h
S1,R
h
R,S2
s
1
+
_
P
J
h
J2,S2
j
2
+w
2
+
_
P
R
_
P
J
h
J1,R
h
R,S2
j
1
+
_
P
R
h
R,S2
v
R
, (4)
e
2
=
_
P
R
_
P
S
(h
S1,R
s
1
+h
S2,R
s
2
) h
R,E
+
_
P
R
_
P
J
h
J1,R
h
R,E
j
1
+
_
P
J
h
J2,E
j
2
+
_
P
R
h
R,E
v
R
+w
E
, (5)
where w
1
, w
2
and w
E
are the noise at S
1
, S
2
, and E,
respectively. Then,
j
which is dened as the overall signal-
to-interference-and-noise-ratio (SINR) of the channel S
i
S
j
(for i, j = 1, 2, i = j) can be calculated as

j
=

Si,Sj

J1,Sj
+
J2,Sj
+
R,Sj
+ 1
, (6)
where
m,n
represents the instantaneous signal-to-noise ratio
(SNR) for the link m n:

Si,Sj
=
2
P
R
P
S
|h
Si,R
|
2
|h
R,Sj
|
2
, (7)

J1,Sj
=
2
P
R
P
J
|h
J1,R
|
2
|h
R,Sj
|
2
, (8)

J2,Sj
= P
J
|h
J2,Sj
|
2
, (9)

R,Sj
=
2
P
R
|h
R,Sj
|
2
. (10)
For simplicity, E applies Maximal Ratio Combining (MRC)
method [13]. And the SINR of link S
i
E can be written as

Ei
=

Si,R,E

Sj,R,E
+
J1,R,E
+
J2,E
+
R,E
+ 1
+

Si,E

Sj,E
+
J1,E
+ 1
, (11)
where

Sj,E
= P
S
|h
Sj,E
|
2
, (12)

Jj,E
= P
J
|h
Jj,E
|
2
, (13)

Sj,R,E
=
2
P
R
P
S
|h
Sj,R
|
2
|h
R,E
|
2
, (14)

J1,R,E
=
2
P
R
P
J
|h
J1,R
|
2
|h
R,E
|
2
, (15)

R,E
=
2
P
R
|h
R,E
|
2
. (16)
B. Problem Formulation
The instantaneous secrecy rate for a relay set S
in
for node
S
i
is [6]
C
Si
(R, J
1
, J
2
) =
_
1
2
log
2
(1 +
i
)
1
2
log
2
_
1 +
Ej
_
_
+
,
(17)
where i, j = 1, 2, i = j, [x]
+
= max {0, x}.
The overall secrecy performance of the system is charac-
terized by the ergodic secrecy rate that is the expectation of
the sum of the two sources secrecy rate, E[C
S
(R, J
1
, J
2
)],
where E[] stands for the expectation operator and
C
S
(R, J
1
, J
2
) = C
S1
(R, J
1
, J
2
) +C
S2
(R, J
1
, J
2
) . (18)
Our objective is to select appropriate nodes R, J
1
, and J
2
in order to maximize the instantaneous (and thus further the
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2011 proceedings
ergodic) secrecy rate for different types of channel feedback.
The considered optimization problem is formulated as
(R

, J

1
, J

2
) = arg max
R,J1,J2Sin
R=J1,J2
C
S
(R, J
1
, J
2
) , s.t.
u
, (19)
where u = 0, 1, R

, J

1
, and J

2
represent the selected relay
and jamming nodes, respectively,
0
= {
k,D

k,E
} and
1
=
{
k,D
, E[
k,E
]} with k S
in
are the feedback knowledge
sets,
0
stands for a global instantaneous knowledge for all
the links and
1
denotes an average channel knowledge for
the eavesdropper links.
C. Selection Techniques without Jamming
In a conventional cooperative network, the relay scheme
does not have a jamming process. We may have the following
solutions.
1) Conventional Selection (CS): The conventional selection
does not take the eavesdropper channels into account and the
relay node is selected according to the instantaneous SNR of
the links between S
1
and S
2
[12]. The conventional selection
is written as
R

= arg max
RSin
{C
S1
(R) +C
S2
(R)}
arg max
RSin
_

CS
1

CS
2
_
= arg max
RSin
_

S1,S2

R,S2
+ 1


S2,S1

R,S1
+ 1
_
arg max
RSin
_

S1,S2

R,S2

S2,S1

R,S1
_
, (20)
with
Si,Sj
and
R,Sj
for (i, j = 0, 1) given by (7) and (10),
respectively. Note that the approximations in (20) are valid
only in high SNR regime.
2) Optimal Selection (OS): This solution takes into account
the relay-eavesdropper links and decides the relay node ac-
cording to the knowledge set
0
[12]. In high SNR regime,
the optimal selection is given as:
R

arg max
RSin
_

OS
1

OS
E2

OS
2

OS
E1
_
, (21)
where

OS
i


Sj,Si

R,Si
,
OS
E
i


Si,E

Sj,E
+

Si,R,E

Sj,R,E
+
R,E
. (22)
III. SELECTION TECHNIQUES FOR JAMMING IN TWO-WAY
RELAY SYSTEMS
In this section, we present several relay selection techniques
based on the optimization problem given by (19).
A. Optimal Selection with Jamming (OSJ)
The optimal selection with jamming assumes a knowledge
set
0
and ensures a maximization of the sum of instantaneous
secrecy rate of node S
1
and node S
2
given in (18), which gives
credit to
(R

, J
1

, J
2

) = arg max
R,J1,J2Sin
R=J1,J2
{C
S
(R, J
1
, J
2
)}
= arg max
R,J1,J2Sin
R=J1,J2
_
1 +
2
1 +
E1

1 +
1
1 +
E2
_
. (23)
B. Optimal Switching (OW)
The original idea of using jamming nodes is to introduce
interference on the eavesdropper links. However, it simul-
taneously degrades the links between the relay R and the
destinations. In some specic situation (e.g. J
2
is close to
one destination), continuous jamming may decreases secrecy
seriously, and acts as a bottleneck for the system. In order
to overcome this problem, we introduce the idea of intelligent
switching between the OSJ and the OS schemes. The threshold
for the involvement of the jammer nodes is
C
S
(R, J
1
, J
2
) > C
OS
S
1
(R) +C
OS
S
2
(R) , (24)
where
C
OS
Si
(R) =
_
1
2
log
2
_
1 +
OS
i
1 +
OS
Ej
__
+
. (25)
For high SNRs, the secrecy rate of the OSJ and OS schemes
can be simplied, and thus, (24) can be written as

E2

E1
>

OS
1

OS
E
2

OS
2

OS
E
1
. (26)
C. Suboptimal Selection with Jamming (SSJ)
With the assumption of
0
, we can get some optimal selec-
tion metrics. In practice, however, only an average knowledge
of these links
1
would be available. As a result. SINRs given
by (11) can be modied as

E
i
=
E[
Si,E
]
E
_

Sj,E

+E[
J1,E
] + 1
+
E
_

Sj,R,E

E
_

Sj,R,E

+E[
J1,R,E
] +E[
J2,E
] +E[
R,E
] + 1
.
(27)
The selection metrics still remain the same as (23), but with
different SINRs given by (27).
IV. PERFORMANCE ANALYSIS
In this section, we analyze the asymptotic performances of
both the proposed jamming and non-jamming schemes in high
transmitted power range.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2011 proceedings
A. Asymptotic Performance for Selection without Jamming
Without loss of generality, we take the the OS scheme for
example. For high transmitted power P
S
, we can get

OS
i
P
S
|h
Si,R
|
2
,
OS
E
i

|h
Si,E
|
2
|h
Sj,E
|
2
+
|h
Si,R
|
2
|h
Sj,R
|
2
. (28)
We can see that
OS
i
grows rapidly as P
S
increases, while

OS
Ei
converges to a value that depends only on the relative
distances between the sources, the eavesdropper, and the
relay. Therefore, the ergodic secrecy rate E[C
S
] also increases
rapidly with the transmitted power P
S
. The slope of the curve
of E[C
S
] versus P
S
(measured by dB) can be calculated as
E(C
S
)
P
S
=
E
_
1
2
log
2
10
P
S
/10
|hS
1
,R|
2
10
P
S
/10
|hS
2
,R|
2

OS
E
1

OS
E
2
_
P
S
=
1
10
log
2
10
0.3322. (29)
For CS, it shares the same asymptotic performance as OS
with a linear increment of approximate 0.3322 slope.
B. Asymptotic Analysis for Selection with Continuous jam-
ming
As the transmitted power P
S
increases to a relatively high
value, it yields

i

L|h
Sj,R
|
2
|h
R,Si
|
2
|h
J1,R
|
2
|h
R,Si
|
2
+ (|h
S1,R
|
2
+|h
S2,R
|
2
) |h
J2,Si
|
2
,

Ei

|h
Si,E
|
2
|h
Sj,E
|
2
+
|h
Si,R
|
2
|h
Sj,R
|
2
. (30)
It is clear that for high P
S
, E[C
S
] converges to a xed
value which is independent of P
S
. Consider the asymptotic
performance of the OS scheme that grows linearly with the
increment of P
S
as described by (29), it is safe to predict
that the ergodic secrecy rate versus transmitted power curves
of jamming and the correspondence non-jamming cases must
intersect with each other at a point P

. In a power range below

P

, the jamming scheme outperforms the non-jamming one,

while above this point, the jamming scheme loses its advantage
in providing a higher ergodic secrecy rate.
V. NUMERICAL RESULTS
In this section, we provide computer simulations in or-
der to validate the analysis in the previous section. The
intermediate node cluster consists of K = 8 nodes, all of
which are located in a 1 1 unit square. S
1
, S
2
and E
are located at (X
S1
, Y
S1
) = (0, 1), (X
S2
, Y
S2
) = (1, 1),
and (X
E
, Y
E
) = (0.5, 0), respectively. The relay and jammer
nodes transmit with a relay-jammer power ratio L = 10.
The path-loss exponent is set to = 3. Aside from the
ergodic secrecy rate, some results are also provided in terms of
secrecy outage probability P[C
S
(R

, J

1
, J

2
) < R
S
], where
P[] denotes probability, and R
S
is the target secrecy rate.
0 0.2 0.4 0.6 0.8 1
0
0.2
0.4
0.6
0.8
1
X
Y
User
Eavesdropper
Relay or Jamming Node
E
S1 S2
Fig. 2. The 1 1 simulation environment with K = 8, = 3.
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Ps [dB]
E
r
g
o
d
i
c

S
e
c
r
e
c
y

R
a
t
e

[
B
P
C
U
]
OW
OSJ
SSJ
OS
CS
OSJ(Asym. approx.)
OS(Asym. approx.)
P'
Fig. 3. Ergodic secrecy rate versus transmitted power Ps for different
selection techniques.
In Fig. 2, the intermediate nodes spread randomly within
the square space. From Fig. 3, it is clear that selections with
jamming outperform their non-jamming counterparts within
a certain transmitted power range (less than P

16 dB).
Outside this range (P > P

), the secrecy rate of OSJ converges

to a power-independent value. Whereas the ergodic secrecy
rate of OS continues to grow with a slope of 0.3322. This
validates the analysis in Section IV. In addition, the suboptimal
scheme SSJ performs almost the same as the optimal scheme
OSJ. Furthermore, it can be seen from Fig. 3 that OW provides
better performance than any other selection techniques with or
without continuous jamming.
Within this conguration, we also compare the performance
of different selection techniques measured by secrecy outage
probability, which is shown in Fig. 4. The target secrecy rate
R
S
is set to be 0.2 Bit Per Channel Use (BPCU). It can be
seen that selection schemes with jamming performs better at
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2011 proceedings
0 5 10 15 20 25 30 35 40 45 50
10
-4
10
-3
10
-2
10
-1
10
0
Ps [dB]
S
e
c
r
e
c
y

o
u
t
a
g
e

p
r
o
b
a
b
i
l
i
t
y
CS
SS
OS
SS-MMISR
OS-MMISR
SS-MSISR
OS-MSISR
OSW
P'
Fig. 4. Secrecy outage probability versus transmitted power Ps for different
selection techniques with Rs = 0.2 BPCU.
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25
0
0.5
1
1.5
2
2.5
3
3.5
4
Ps [dB]
E
r
g
o
d
i
c

S
e
c
r
e
c
y

R
a
t
e

[
B
P
C
U
]
0 0.5 1
0
0.5
1
OW
OS
CS
OSJ
SSJ
OS(Asym. Aprox.)
OSJ(Asym. Aprox.)
S1 S2
Relays
E
Fig. 5. Ergodic secrecy rate for a scenario where relays are close to S
1
.
providing lower secrecy outage probability within a certain
transmitted power range (P
S
< P

, P

20 dB). Regarding
the hybrid protocols, the OW scheme outperforms the non-
switching selection techniques.
Finally, we relocate the intermediate nodes as a cluster (as
shown in Fig. 5). We can see that continuous jamming schemes
(i.e. OSJ, SSJ) are inefcient here, which converge to less than
0.5 BPCU, conrming our discussion in Section IV. On the
other hand, OW still outperforms all the other selection tech-
niques by a large scale (more than 4 BPCU when P
S
is high),
which demonstrates the OW scheme as an efcient approach to
overcome the negative interference of continuous jamming.
VI. CONCLUSIONS
This paper has discussed relay selection issues in two-way
cooperative networks with secrecy constraints. The proposed
scheme enables an opportunistic selection of one conventional
relay node and two jamming nodes to increase security against
eavesdroppers based on both instantaneous and average knowl-
edge of the eavesdropper channels. The selected relay node
helps to enhance the information transmission between the
two sources via an AF strategy, while the jamming nodes are
used to create intentional interference at the eavesdropper. We
found that jamming is an effective solution within a certain
transmitted power range for scenarios with sparsely spreading
relay/jamming nodes. The proposed hybrid scheme which
switches intelligently between jamming and non-jamming
modes is efcient in providing the highest secrecy rate in
almost the whole transmitted power regime in two-way co-
operative networks.
ACKNOWLEDGMENT
This work was partially supported by the US NSF CNS-
0953377, CNS-0905556, and CNS-0910461, and also by the
National Nature Science Foundation of China under grant
number 60972009 and 61061130561, as well as the Na-
tional Science and Technology Major Project of China un-
der grant number 2009ZX03003-011, 2010ZX03003-003, and
2011ZX03005-002.
REFERENCES
[1] A. D. Wyner, The wire-tap channel, Bell System Technical Journal,
vol. 54, no. 8, pp. 13551387, Oct. 1975.
[2] S. K. Leung-Yan-Cheong and M. E. Hellman, The Gaussian wiretap
channel, IEEE Transactions on Information Theory, vol. 24, no. 4,
pp. 451456, Jul. 1978.
[3] I. Csisz ar and J. K orner, Broadcast channels with condential messages,
IEEE Transactions on Information Theory, vol. 24, no. 3, pp. 339348,
May 1978.
[4] P. Parada and R. Blahut, Secrecy capacity of SIMO and slow fading
channels, in Proceedings of IEEE International Symposium on Informa-
tion Theory, Adelaide, Australia, Sep. 2005.
[5] L. Lai and H. E. Gamal, The relay-eavesdropper channel: Cooperation
for secrecy, IEEE Transactions on Information Theory, vol. 54, no. 9,
pp. 40054019, Sep. 2008.
[6] L. Dong, Z. Han, A. P. Petropulu, and H. V. Poor, Improving wireless
physical layer security via cooperating relays, IEEE Transactions on
Signal Processing, vol. 58, no. 3, pp. 18751888, Mar. 2010.
[7] Jingge Zhu, Jianhua Mo, and Meixia Tao, Cooperative secret commu-
nication with articial noise in symmetric interference channel, IEEE
Communications Letters, vol. 14, no. 10, pp. 885887, Oct. 2010.
[8] B. Rankov and A. Wittneben, Achievable rate regions for the two-
way relay channel, in Proceedings of IEEE International Symposium
on Information Theory, Seattle, USA, Jul. 2006.
[9] B. Rankov and A. Wittneben, Spectral efcient protocols for half-duplex
fading relay channels, IEEE Journal on Selected Areas in Communica-
tions, vol. 25, no. 2, pp. 379389, Feb. 2007.
[10] Lingyang Song, Yonghui Li, Anpeng Huang, Bingli Jiao, and A. V. Vasi-
lakos, Differential modulation for bidirectional relaying with analog
network coding, IEEE Transactions on Signal Processing, vol. 58, no. 7,
pp. 39333938, Jul. 2010.
[11] Yindi Jing,A relay selection scheme for two-way amplify-and-forward
relay networks, in Proceedings of International Conference on Wireless
Communications & Signal Processing, Nov. 2009.
[12] I. Krikidis, J. Thompson, and S. Mclaughlin,Relay selection for secure
cooperative networks with jamming, IEEE Transactions on Wireless
Communications, vol. 8, no. 10, pp. 50035011, Oct. 2009.
[13] T. S. Rappaport, Wireless Communicaitons Principles and Practice, 2nd
ed. Prentice Hall, 2002.
[14] Mugen Peng and Wenbo Wang. Technologies and standards for TD-
SCDMA evolutions to IMT-Advanced, IEEE Communications Magazine,
vol. 47, no. 12, pp. 5058, Dec. 2009.
[15] Lingyang Song and A. G. Burr, Successive interference cancellation
for space-time block codes over time-selective channels, IEEE Commu-
nications Letter, vol. 10, no. 12, pp. 837839, Dec. 2006.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2011 proceedings