Optimization of Simultaneous Wireless Information
and Power Transfer in Cloud Radio Access
Networks
Yingna Ma, Mugen Peng, Zhongyuan Zhao, Zheng Zhou
Wireless Signal Processing and Network Laboratory
Key Laboratory of Universal Wireless Communication, Ministry of Education
Beijing University of Posts and Telecommunications, Beijing, 100876, China
Email:mayingna@bupt.edu.cn
Abstract—In this paper, we focus on optimization of simul-
taneous wireless information and power transfer (SWIPT) in
uplink cloud radio access network (C-RAN). The key idea is
to design transceiver architecture and implement the SWIPT
strategy at remote radio heads (RRHs). The minimum mean-
square-error (MMSE) is considered as the performance metrics
with transmit power constraint and energy constraint. Toward
the goal of MMSE, the precoders and the detectors are iteratively
updated in proposed scheme. However, because of the energy
constraint, the precoding optimization problem with a specified
detection matrix become non-convex. To solve this problem, a new
precoding design based on the Lagrangian dual relaxation (LDR)
is developed. Meanwhile, conditions to make sure the optimiza-
tion problem has its physical interpretation and LDR become
convex are determined. Finally, simulation results demonstrate
that the proposed transceiver design can significantly improve
the system performance.
Index Terms—Simultaneous wireless information and power
transfer, uplink cloud radio access networks, transceiver design,
energy constraint, Lagrangian dual relaxation.
I. INTRODUCTION
Cloud radio access network (C-RAN) is proposed recently
and its name comes from four characters: centralized process-
ing, collaborative radio, real-time cloud computing and clean
system [1]. C-RAN could be one of direction of future mobile
infrastructure, but it faces some challenges. For example,
its large-scale remote radio heads (RRHs) give rise to the
increasing of transmission power consumption.
Energy harvesting, which uses signals to carry energy, is
an effective approach to prolonging the lifetime of energy-
constrained devices. In wireless networks [2], simultaneous
wireless information and power transfer (SWIPT) technol-
ogy has attracted increasing attention. Compared with the
traditional energy sources such as solar and wind, radio
frequency (RF) signals are better sources for applications with
substantial power consumptions because it is more convenient,
controllable.
For the SWIPT technology, although a unified study is not
accomplished, there have been notable researches in some
literatures. From the work [2], the case of separated receivers
was considered, and authors derived the optimal transmission
strategy to achieve different tradeoffs for the boundary of a
rate-energy (R-E) region. On the other hand, in [3], a cognitive
radio inspired asymmetric network coding (CR-AsNC) scheme
with precoding optimization in MIMO systems was proposed.
An iterative algorithm based on alternating optimization ap-
proach was taken to minimize the MSE of systems. For
a cloud radio access networks (C-RAN) system, however,
every RRH needs to receive the information from many users,
and transfers the information to the central processing office
and the processing power house (i.e. BBU pool). RRH is
the battery-limited appliance that needs continual power to
support its work. Therefore, SWIPT at RRHs is essential and
a transceiver architecture should be designed.
In this paper, we extend the MAC phase part of the work
in [3] and consider the transceiver architecture design under
energy constraint for SWIPT at RRHs in an uplink C-RAN
where multiple users transmit message to multiple RRHs
which have two antennas (one antenna for information trans-
mission and the other for energy harvesting). First, minimizing
mean-square-error (MSE) is chosen as the optimization metric
under transmission power constraint and the energy constraint.
Second, to handle this concave problem with a nonconvex
constraint, we develop a new transceiver design by considering
the Lagrangian dual relaxation (LDR), and from the energy
constraint, we deduce the energy harvesting range which
makes sure the system has physical interpretation. Finally,
comparing with the energy beamforming precoding design
solution proposed in [2], our proposed iterative transceiver
design makes uplink system performance better.
The rest of this paper is organized as follows. In section II,
the system model is introduced. The transceiver optimization
with the energy constraint based on the criterion of MSE
is presented in Section III. Section IV discusses simulation
results. Finally, the conclusion is presented in section V.
II. S
YSTEM MODEL
Consider an uplink transmission scenario in cloud radio
access networks (C-RANs), where 𝐾 users transmit their own
message to 𝑀 remote radio heads (RRHs). As shown in Fig.
1, each RRH is equipped with two antennas to implement
the SWIPT strategy, and the information transmission and the
power transfer processes can be accomplished independently
by using different antennas at each RRH, i.e., the antennas
978-1-5090-1698-3/16/$31.00 ©2016 IEEE