Paired-relay-selection schemes for two-way
relaying with network coding
ISSN 1751-8628
Received on 12th August 2014
Accepted on 2nd December 2014
doi: 10.1049/iet-com.2014.0726
www.ietdl.org
Yunxiang Jiang
1
, Francis Chung Ming Lau
1,3
✉
, Zeeshan Sattar
1
, Ivan Wang-Hei Ho
1
,
Qing Feng Zhou
2
1
Department of Electronic and Information Engineering, The Hong Kong Polytechnic University, Hong Kong, Hong Kong
2
Communication Engineering Department, School of Computer and Information, HeFei University of Technology (HFUT), Hefei,
People’s Republic of China
3
Shenzhen Research Institute, The Hong Kong Polytechnic University, Hong Kong, Hong Kong
✉ E-mail: encmlau@polyu.edu.hk
Abstract: To exchange information between two sources in a t wo-way relaying network with multiple potential
relays, most researches focus on two-hop relay system with single-relay-selection (SRS) scheme. Comparing with SRS
scheme, the authors first design a paired-relay-selection (PRS) scheme in which a pair of ‘best’ relays broadcast
network-coded information to other nodes (source or relay). They propose an optimal selection algorithm and a
suboptimal al gorithm tha t selects the pair of ‘best’ relays in the PRS scheme and they describe how the n odes
exchange information in a frame consisting of four timeslots. Both the analytical and simulation results show that
when the pathloss exponent is large and/or there is a sufficient number of relays to choose from, using two relay
nodes can provide a lower outage compared with using only one relay node even under the same total transmit power
in uniformly distributed relay networks. In addition, to reduce the overhead of the P RS scheme, they propose an
iterative-PRS (I-PRS) scheme in which the paired relay is selected in an iterative and opportunistic way. Simulation
results show that the I-PRS scheme has nearly the same outage performance as the PRS scheme under time-invariant
channels and significantly outperforms the PRS scheme under time-varying channels.
1 Introduction
Over the past decade, various aspects of wireless cooperative
channels and networks such as information-theoretic capacity [1],
diversity [2], outage performance [3, 4] and cooperative coding
[5] have been investigated. Relay selection (RS) [6]isa
promising scheme to achieve diversity gain and is easy to be
implemented. A lot of work has been done on the issue of RS
and the achievable diversity order of RS [7–9] in cooperative
communication systems.
1.1 Related work
In [7], Zhao et al. have proposed an optimal power allocation
scheme for the amplify-and-forward (AF) protocol in
cooperative communication networks. They have also proposed
a criterion for selecting the ‘best’ relay to participate in a
transmission. They have proved analytically and also with the
help of simulations that the single ‘best’ relay achieves the same
maximum diversity order of (K +1) with K relays. They also
claimed that single-relay-selection (SRS) scheme and two-hop
relay system can achieve the maximum diversity order.
Moreover, the SRS scheme achieves a higher instantaneous
throughput and lower outage probability than the m-relay
scenario. This result, however, is based on the assumption that
all channels have equal gain, which might not be realistic in
practice. In [8], Jing and Jafarkhani have proposed the idea of
multiple-RS. They have introduced the idea of relay ordering
(RO) for selecting more t han one relay in the AF protocol. It
has been proved that multiple-RS usin g RO achieves full
diversity and performs better than s ingle-RS in terms of bit
error rate. Yet, they impose no constraint on the to tal transmit
power of the cooperative relay network. Therefore the total
transmit power increases as the number of cooperative relays
increases. Furthermore, they have assumed that all cooperative
relays work in parallel and are perfectly synchronised. In [9], Yi
and Kim have proposed an AF RO strategy, in which data
transmission is achieved using K relays and (K + 1) hops. An
approximated closed-form end-to-end signal-to-noise ratio (SNR)
in a two-relay case has been derived. I n [10], RS and RO have
been jointly considered in a two-relay system. RS and RO can
complement each other because the former is more spectrally
efficient, whereas the latter is more energy efficient. The results
in [10] have indicated that RO is preferred for s everely
attenuating channels, whereas RS is preferred for less
attenuating channels.
At the same time, network coding is a promising method aiming to
improve the throughput of communication systems [11–13]. In the
two-way relaying channel in [14,
15], which is a special case of
cooperative communication networks, network coding has been
applied. The network consists of two users who communicate with
each other with the help of a single relay, and network coding has
been shown to improve the network capacity, diversity gain and
network efficiency of such a wireless two-way relay network [14,
15]. In [14], a physical-layer network-coding (PNC) method has
been presented to improve the network efficiency further. It has
been demonstrated that the joint use of relay and network coding
not only improves the information transmission efficiency [13] but
also reduces the overall power consumption significantly in
communication systems [14].
In [12, 16, 17 ], RS with network coding for two-wa y
relaying has been studied under a multiple-r elay scenario. A
denoise-and-forward network coding opportunistic relaying scheme
hasbeenproposedin[12], wher eas a decode -and-fo rw ard (DF)
two-way relay ing with network coding and opportunistic RS has
been proposed in [16]. Both schemes have been shown to achieve
the full diversity gain.
IET Communications
Research Article
IET Commun., 2015, Vol. 9, Iss. 6, pp. 888–896
888
&
The Institution of Engineering and Technology 2015