Robust beamforming design for multiple-
input–single-output secrecy multicast ing
systems with simultaneous wireless
information and power transmission
ISSN 1751-8628
Received on 29th September 2015
Revised on 4th February 2016
Accepted on 16th March 2016
doi: 10.1049/iet-com.2015.0945
www.ietdl.org
Zhengyu Zhu
1
✉
, Zheng Chu
2
, Zhongyong Wang
1
, Jianhua Cui
3
1
School of Information Engineering, Zhengzhou University, Zhengzhou, People’s Republic of China
2
School of Electrical and Electronic Engineering, Newcastle University, Newcastle upon Tyne NE1 7RU, UK
3
School of Physical and Electronic Information, Luoyang Normal University, Luoyang, People’s Republic of China
✉ E-mail: zhuzhengyu6@gmail.com
Abstract: In this study, the authors study simultaneous wireless information and power transfer for multiuser multiple-
input–single-output secure multicasting channels with imperfect channel state information. First, a robust secure
beamforming design is considered, where the transmit power is minimised subject to the secrecy rate outage
probability constraint for legitimate users and the harvested energy outage probability constraint for energy harvesting
receivers. The original problem is non-convex due to the presence of the probabilistic constraints. By utilising
Bernstein-type inequali ties, the authors transform the outage constraints into the deterministic forms. In order to
identify a l ocal optimal rank-one solution, the authors propose an efficient a pproach based on a constrained concave
convex procedure method to convert the original problem into a sequence of convex programming problems. Finally,
simulation results are provided to validate the performance of the proposed design methods.
1 Introduction
Energy harvesting (EH) techniques have been introduced to extract
energy from the nature [1, 2]. Recently, simultaneous wireless
information and power transfer (SWIPT) has been proposed which
utilises ambient radio frequency (RF) signals for providing power
supplies to wireless networks [3–8]. A typical SWIPT system of
practical interest consists of one transmitter that has constant
power supply and broadcasts wireless signals to a group of
receivers, among which some decode information, typically
referred to as information receivers, while others scavenge energy
from ambient radio signals, named energy receivers.
Secrecy transmission has extracted more and more attentions in
wireless networks, especially focusing on physical-layer security
[9]. On the other hand, security has been recognised as an
important issue for SWIPT systems [10–16]. To address this
challenge, physical-layer security techniques have been applied to
SWIPT to achieve secure communication. In [10], the optimal
information and energy beamforming designs for secrecy
SWIPT systems in multiple-input–single-output (MISO) broadcast
channels have been studied. Moreover, the secure transmission
for SWIPT schemes was investigated in cooperative relay
networks [11], multiple-input–multiple-output broadcasting
systems [12].
It is worth noting that all of these works have considered the
assumption of perfect channel state information (CSI) available at
the transmitter. In practice, it may be difficult to obtain perfect
CSI at the transmitter due to channel estimation and quantisation
errors. Furthermore, as we all know, the perfect CSI methods are
quite sensitive to the channel uncertainties. Some secure SWIPT
systems are constructed under imperfect channel realisation in
[13–16]. However, these literatures only focus on the
norm-bounded channel uncertainty model in secure SWIPT
transmission schemes [13–16]. In [14–16], the rank relaxation is
applied to relax problem and the proof of the rank-two solution of
the transmit covariance matrix is presented. Since it is challenging
to show rank-one solution, the authors only provide a suboptimal
solution by using Gaussian randomisation (GR) method [17].
However, one main disadvantage of the GR method is poor
performance.
In practice, it is always difficult that the channel error bound can
be obtained exactly. In this paper, we consider a system where the
transmitter only obtains the statistical information of channel
uncertainties. We incorporate two different types of statistical
channel uncertainty models: (i) partial channel uncertainties
(PCUs) which assume perfect CSI for legitimate users and
imperfect CSI for eavesdroppers and EH receivers; (ii) full channel
uncertainties (FCUs) which consider imperfect CSI for legitimate
users, eavesdroppers, and EH receivers.
We formulate the robust transmit power minimisation (PM)
problem subject to the secrecy rate outage probability constraint
for legitimate users and the harvested power outage probability
constraint for both channel uncertainty models. Under various
assumptions on the CSI, we seek to design robust transmit
beamforming strategies for MISO SWIPT wiretap channels. The
main contribution of this paper is summarised as follows:
† For the PCU model, the robust PM problem with the outage
constraints is converted into a convex semidefinite programming
(SDP) by employing semidefinite relaxation (SDR) and the
Bernstein-type inequality [18]. In general, the relaxed problem is
not guaranteed to obtain a rank-one solution [
17]. Based on a
constrained concave convex procedure (CCCP)-based algorithm,
we propose an iterative method to recover a local optimal rank-one
beamforming solution to the PM problem.
† For the FCU model, applying the Bernstein-type inequality and
SDR, the PM problem can be transformed into a SDP problem
which adopts the CCCP-based iterative algorithm to achieve a
local optimal rank-one solution.
The rest of this paper is organised as follows. Section 2 presents
the system model. Section 3 provides the statistical channel
uncertainty and problem formulation. The robust design methods
are developed in Section 4. Section 5 compares the computational
complexity of the proposed robust design methods, and Section 6
IET Communications
Research Article
IET Commun., 2016, Vol. 10, Iss. 15, pp. 1979–1985
1979
&
Crown copyright 2016