264 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO.

1, FIRST QUARTER 2018

Simultaneous Wireless Information and Power

Transfer (SWIPT): Recent Advances and
Future Challenges
Tharindu D. Ponnimbaduge Perera, Student Member, IEEE, Dushantha Nalin K. Jayakody , Member, IEEE,
Shree Krishna Sharma , Member, IEEE, Symeon Chatzinotas, Senior Member, IEEE,
and Jun Li, Senior Member, IEEE

Abstract—Initial efforts on wireless power transfer (WPT) issues presented by simultaneous wireless information and power
have concentrated toward long-distance transmission and high transfer (SWIPT) and WPT assisted technologies. More specif-
power applications. Nonetheless, the lower achievable transmis- ically, in contrast to the existing works, this paper identifies
sion efficiency and potential health concerns arising due to high and provides a detailed description of various potential emerg-
power applications, have caused limitations in their further devel- ing technologies for the fifth generation communications with
opments. Due to tremendous energy consumption growth with SWIPT/WPT. Moreover, we provide some interesting research
ever-increasing connected devices, alternative wireless informa- challenges and recommendations with the objective of stimulating
tion and power transfer techniques have been important not only future research in this emerging domain.
for theoretical research but also for the operational costs saving
and for the sustainable growth of wireless communications. In Index Terms—RF wireless power transfer, 5G communica-
this regard, radio frequency energy harvesting (RF-EH) for a tions, simultaneous wireless information and power transfer,
wireless communications system presents a new paradigm that interference exploitation, RF energy harvesting.
allows wireless nodes to recharge their batteries from the RF
signals instead of fixed power grids and the traditional energy
sources. In this approach, the RF energy is harvested from ambi- I. I NTRODUCTION
ent electromagnetic sources or from the sources that directionally
HE EVENTUAL fate of mobile communication is
transmit RF energy for EH purposes. Notable research activi-
ties and major advances have occurred over the last decade in
this direction. Thus, this paper provides a comprehensive sur-
T expected to be completely different from what we expe-
rience today. Ultra top-notch videos and better widescreen
vey of the state-of-art techniques, based on advances and open resolutions in mobile devices are forcing us either to look for
better sustainable power sources or to perform energy harvest-
Manuscript received April 19, 2017; revised October 23, 2017; accepted ing (EH) from external power sources for the next-generation
November 27, 2017. Date of publication December 15, 2017; date of cur- of wireless communication systems.
rent version February 26, 2018. This work was supported by the framework
of Competitiveness Enhancement Program of the National Research Tomsk The fifth generation (5G) of wireless communication is the
Polytechnic University under Grant TPU CEP_IC_110\2017, in part by FNR, proposed next telecommunication standard over the 4G/IMT-
in part by Luxembourg INTER INWIPNET, in part by the National Natural Advanced standards to meet the demands that communication
Science Foundation of China under Grant 6172780261501238, in part by
the Jiangsu Provincial Science Foundation under Project BK20150786, in networks will face by 2020 [1]. It is expected that the total
part by the Specially Appointed Professor Program in Jiangsu Province number of connected devices could reach up to tens or hun-
in 2015, in part by the Fundamental Research Funds for the Central dreds of billions by the time that the first standard of 5G
Universities under Grant 30916011205, and in part by the COST Action
on Inclusive Radio Communications under Grant CA15104. (Corresponding is expected to be matured due to many new applications
author: Dushantha Nalin K. Jayakody.) in personal communication [1], [2]. As a summarization of
T. D. Ponnimbaduge Perera is with the School of Computer Science and the journey of generations of the cellular network, the first
Robotics, National Research Tomsk Polytechnic University, Tomsk 634050,
Russia (e-mail: [email protected]). generation of communication network referred as first gener-
D. N. K. Jayakody is with the School of Computer Science and Robotics, ation (1G) launched in around 1982 and it was fully analog.
National Research Tomsk Polytechnic University, Tomsk 634050, Russia, Second generation (2G) was launched in 1991 which was
and also with the College of Science and Engineering, Hamad Bin Khalifa
University, Qatar Foundation, Doha, Qatar (e-mail: [email protected]). based on the principles of digital communication. Moreover,
S. K. Sharma is with the Department of Electrical and Computer 2G added cellular data in the form of General Packet Radio
Engineering, University of Western Ontario, London, ON N6A 3K7, Canada Services (GPRS). After almost 10 years, the third genera-
(e-mail: [email protected]).
S. Chatzinotas is with the Interdisciplinary Centre for Security, Reliability tion (3G) networks came out introducing faster data rates than
and Trust, University of Luxembourg, 2665 Luxembourg City, Luxembourg 2G. Roughly 10 years later, the fourth generation communica-
(e-mail: [email protected]). tion (4G) was introduced with even faster data rates and more
J. Li is with the School of Electronic and Optical Engineering, Nanjing
University of Science and Technology, Nanjing 210094, China, also cutting-edge technologies. Mainly, 5G wireless communica-
with the School of Computer Science and Robotics, National Research tion is expected to provide an increase in data rates, bandwidth,
Tomsk Polytechnic University, Tomsk 634050, Russia, and also with the coverage and reliability of connectivity, together with the
National Mobile Communications Research Laboratory, Southeast University,
Nanjing 210094, China (e-mail: [email protected]). enormous reduction in latency and energy consumption [3].
Digital Object Identifier 10.1109/COMST.2017.2783901 Moreover, 5G wireless networks also have integrated new
1553-877X  c 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 265

communication trends such as Internet of Things (IoT) and and smaller in size. We imagine as a future possibility that
machine type communications (MTC) in order to accommo- RF signals will not only become a power source for func-
date more flexible networked social information sharing one tional devices in the network, but their power and information
million connections per square metre [4]. The ultimate respon- transmission aspects will also be combined. Power-line com-
sibility of shaping 5G has been taken by the Group Special munication (PLC) is a kind of communication technology that
Mobile Association (GSMA) with its partners. Development carries both information and power simultaneously through
of green technology and reduction in power consumption wired connections. Thus, the PLC is more appropriate for
by the devices are the two of eight major requirements of in-home applications when compared with other communica-
5G systems identified by the industries and academia [3]. tion methods [12]. However, a number of technical problems
Recently, several attempts have been made to implement self- exist for practical deployment of wireless PLC devices such
sustainable communication systems with EH techniques to as energy efficiency and radio interference. In order to address
support novel concepts such as IoT and 5G wireless networks, the aforementioned issues, SWIPT could be a promising tech-
while maintaining desirable Quality of Service (QoS). nology towards implementing the core idea of PLC in practical
EH is a process of capturing and converting wasted or unim- scenarios. SWIPT can result in notable gains in terms of power
portant energy such as heat, sound and Radio Frequency (RF) consumption, spectral efficiency, interference management and
signals into electricity and putting it to work based on power transmission delay by enabling simultaneous transmission of
requirements. The ambient environment contains high-quality both power and information [7], [8], [13]. A proper design
energy sources compared to currently available batteries and that considers both information and power can provide note-
charged super-capacitors. Lately, several studies have explored worthy energy-efficient wireless communication and lead to
the possibility of using natural energy sources in the EH pro- possible state-of-the-art applications and services for the 5G
cess within the communications network [5], [6]. EH using and next-generation sustainable societies.
natural sources was not as effective as expected due to the
irregular and unforeseeable nature of ambient sources. Also,
major EH techniques are scenario-specific and are applicable A. Review of Related Survey Articles
only in specific environments [7]. Only a few survey and overview articles directly related to
WPT is one of the EH technologies that overcome the SWIPT and EH have been published [3], [7], [11], [13]–[23].
aforementioned limitations [8], where the nodes in the com- A summary of these existing survey articles are listed down in
munications network charge their inbuilt batteries from elec- TABLE I. A comprehensive survey on the next-generation of
tromagnetic radiation. Green energy can be harvested using 5G wireless networks has been done in the work of [3] and a
two methods in the WPT, either from the ambient signals or brief explanation about the energy requirement of 5G has been
from a dedicated fully controlled power source like a base sta- provided with an introduction to EH. Krikidis et al. [7] have
tion. More recent contributions based on WPT have focused on provided a well-organized overview of SWIPT technology,
short distance (near-field) rather than long distance (far-field) including recent advances and future challenges. A rectifier
energy transmission. Near-field and far-field distances depend antenna circuit, which is essential for the implementation of
on the application scenarios. For an example, when exchang- SWIPT technology, is explained in [7]. In [11], a survey is
ing electromagnetic power wirelessly, the near-field distance presented based on current research in radio frequency EH
can only be up to several meters in both indoor and outdoor networks (RF-EHNs). It covers communication protocols, cir-
environments while the far-field distance can be up to several cuit designs and emerging operation designs for different types
kilometres [7]. However, near-field WPT has multiple draw- of RF-EHNs. Applications of smart antenna technologies in
backs such as distance constraints, difficulties in maintaining SWIPT enabled Multiple-Input Multiple-Output (MIMO) and
field strengths within safety levels, high initial cost, difficulties relaying have been discussed in [13] along with the basics of
in resonant inductive tuning, use of high frequency in supply SWIPT and receiver architectures. These works [7], [11], [13]
and the infeasibility of air ionization techniques [9]. Within have provided valuable insights into foundational aspects of
communication networks, the distance from the base station EH and SWIPT techniques along with some recommendations.
to the device is a critical factor in both information and power In the work of [14], Ulukus et al. summarized recent advances
transfer. Therefore, far-field WPT techniques need to be further and related results in EH and WPT enabled wireless networks
improved. The prospect of integrating WPT with communica- from the perspective of signal processing, information and
tion networks creates a need for technology that can transfer communication theory and wireless networking. Moreover,
both information and power simultaneously to the end-devices. Ulukus et al. [14] have discussed a variety of topics including
For this requirement, the concept of SWIPT was first intro- the effects of information theoretic physical layer performance
duced in [8] from a theoretical point of view. Recently, SWIPT to scheduling policies and medium access control protocols.
has attracted significant attention in wireless communication The article [15] provided a review based on SWIPT from
networks [10], [11]. In the era of 5G communication, SWIPT both the technical and theoretical perspectives. The authors
technology can be of fundamental importance for energy and have started with a discussion of different types of SWIPT
information transmission within numerous types of modern scenarios including challenging issues. Then Chen et al. [15]
communications networks. surveyed various effective PHY security techniques and their
With the rapid advancement of the technology, wireless impact on the performance of SWIPT. In addition, they
transceivers and sensors are becoming more power efficient proposed to use Massive MIMO techniques to enhance the
266 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

TABLE I
C OMPARISON OF E XISTING S URVEY A RTICLES

(To be continued)

performance of SWIPT by presenting performance gain via A brief survey on EH and green communication in Full-
numerical results. Finally, a few potential future directions Duplex Cognitive Radio networks (FD-CRNs) has been
have been explained. done in the work of [16]. However, the main focus of
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 267

TABLE I
C ONTINUED

(To be continued)

Amjad et al. [16] had been to provide an in-depth discus- charging compliant communication networks [17]. Moreover,
sion of existing FD-CRN architectures and the related case the authors included existing solutions of providing WPT
studies. A comprehensive survey has been done on wireless through static charger scheduling, mobile charger dispatch and
268 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

TABLE I
C ONTINUED

wireless charger deployment. The latter part of [17] discussed realistic EH networks. In the work of [20], Ozel et al. reviewed
open issues and practical challenges in wireless charging along fundamental limits of EH communications. The authors high-
with some future directions. The work of [18], provides a valu- lighted that every transmitted symbol of an EH system is
able comprehensive survey based on exploiting interference instantaneously constrained by the energy available in the bat-
for wireless EH. The authors discussed the classification of tery. Moreover, partial results for the cases of infinite-sized,
wireless EH systems by exploiting interference based on zero-sized and finite-sized batteries have been reviewed, while
receiver architecture, network topology, antenna dimension building up an awareness in the scheduling and optimization
and interference management. Moreover, the key problems in sides of the available literature. In the work of [21], Visser
interference management and some open research challenges and Vullers presented an overview and the principles of RF-
have been discussed in [18]. EH and power transfer for wireless sensor network (WSN)
Another survey on energy harvesting (EH) communica- applications. In [21], firstly the feasibility of EH and trans-
tions and networks has been carried out in the work of [19]. mission has been discussed leading to the conclusion that RF
Throughout [19], fundamental aspects of energy scheduling, EH and power transmission are more suitable for small-sized
characteristics of different energy sources, research challenges battery-powered sensors in future smart buildings. In addition,
and various topics related to EH have been discussed. In the smart building integration with the RF-EH enabled sensors and
latter part of [19], Ku et al. pointed out possible application the effects of RF radiation on human health were also dis-
systems and future directions for further development of more cussed. Akhtar and Rehmani [22] discussed the way of recent
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 269

technological advancements have given rise to prominent tech- we include recommendations on various aspects such
niques to limit energy constraints in pervasive sensor networks. as scheduling, hardware impairments, channel cod-
Akhtar and Rehmani [22] provided a concise review of cur- ing techniques, investigation of feedback and informa-
rent research on EH in wireless body area networks (WBANs) tion theoretic frameworks to improve the efficiency of
including potential harvestable sources and their characteristics SWIPT enabled communication systems.
and the usability in minimizing energy constraints. Moreover,
the article [22] listed down current formidable challenges and C. Paper Organization
possible future research directions in EH enabled wireless This article is structured as follows: Section II presents an
body sensor networks. While considering WSNs, a compre- overview of RF-EH while Section III provides a comprehen-
hensive review has been done in the work of [23] based on sive survey of WPT. Section IV explains SWIPT technology
energy replenishment using renewable and traditional energy and different receiver architectures used in SWIPT. Then, we
resources for sustainable WSNs. The work [23] also presented comprehensively explain interference exploitation in SWIPT
a survey of alternative techniques that can address the issues of enabled systems in Section V. In Section VI, we introduce
limited power in wireless sensor nodes and specifically focused emerging SWIPT scenarios. This section contains and presents
on EH from renewable sources as well as traditional energy well-structured explanations about SWIPT enabled emerging
resources in WSNs. communication technologies. In Section VII, we focus on rec-
ommendations and future directions of SWIPT. Section VIII
B. Contributions of This Article summarises this survey and finally, the Appendix provides
a table with the definitions of acronyms and notations used
Despite the existing SWIPT related articles highlighted in throughout this paper.
Section I-A, no contributions have been reported in the direc-
tion of providing a comprehensive review of the emerging II. RF E NERGY H ARVESTING
technologies in SWIPT. Therefore, our objective in this paper
is to focus more on SWIPT emerging technologies and its EH is the process of converting energy that is obtained from
applications for the next-generation communication systems. external energy sources such as wind, thermal, kinetic and
Motivated by this aspect, in this paper, we fully investigate var- solar energy into electricity. The environment provides a vari-
ious SWIPT emerging communication technologies with their ety of energy sources, in addition to the energy stored in batter-
advantages, use case scenarios, technical challenges and future ies or super capacitors. There has been considerable growth of
directions. The scope of this survey covers RF-EH, WPT, interest relating to RF EH as the demand of WSNs, the num-
Interference Exploitation, SWIPT, SWIPT enabled emerging ber of radio transmitters and mobile base stations continues to
technologies, key research challenges and future directions increase [24]. RF energy is one form of electromagnetic energy
with the objective of realizing the energy-efficient 5G and which consists of radio waves of magnetic and electrical
beyond. The contributions of this paper are summarized as energy radiating through free space. RF-EH becomes a pro-
follows. pitious solution for wireless networks with a limited lifetime.
1) We provide a comprehensive summary of SWIPT and The power supply of RF-EHN has been maintained at a certain
related concepts such as WPT, RF-EH and interference rate from RF environment built from billions of radio trans-
exploitation with comparative tables and illustrative mitters, mobile phones, mobile base stations and television
figures. Moreover, we introduce one of the uprising broadcast systems around the world [25]. Therefore, the wire-
topics in the domain, Interference Exploitation with a less devices now can harvest energy from RF signals according
basic analysis of interference in two user link. Herein, to the power variations over the time. In this section, we pro-
our objective is to provide foundational knowledge of vide basic foundational knowledge on RF-EH, which helps
SWIPT, as well as some suggestions on productive areas to understand the SWIPT techniques in different emerging
for further research in related to SWIPT concept. technologies explained in the following sections of this paper.
2) We survey most of the SWIPT emerging technologies
from both academic and industrial perspectives based on A. RF-EH Circuit Design
the most recent literature of the particular emerging tech- The schematic of the wireless EH module in the EH devices
nologies such as SWIPT enabled multi-carrier systems, is presented in Fig. 1 [11]. The idea of introducing this circuit
full-duplex SWIPT systems, bistatic scatter radio, design is to provide the knowledge of circuit design required to
symbol level precoding, cooperative relaying, non- understand the features of the RF-EH enabled communication
orthogonal multiple access (NOMA), secure transmis- systems. Comprehensive details regarding the circuit design
sion, mmWave communication, WSNs, MIMO systems, and electronic part of the design is beyond the scope of this
multi-user Multiple-Input Single Output (MISO) and paper.
broadband wireless systems (BWS). At the end of each 1) Antenna Design: Antenna is the device in the RF-EH
sub sections of emerging technologies, we have included circuit that is responsible for capturing RF signals. Antenna
future directions with summary and insights. efficiency is a key factor related to the operating frequency
3) We anticipate the future directions of SWIPT and and helps to ensure the successful operation of the RF-EH
its applications in modern technologies such as the system. High antenna gain while miniaturizing size are the
IoT, MTC and satellite communication. Moreover, main goals of antenna technology, which leads to a trade-off
270 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

TABLE II
E XPERIMENTAL DATA OF RF-EH [26]

antennas’ effective area Ae = λ2 Gr /4π , then Pr can be

given by
Pt Gt
Pr = Ae cos2 φ , (1)
4π R2
where Pt equals to output power of transmitting antenna, Gr
and Gt represent gain of the receiving and transmitting anten-
nas, respectively, wavelength denotes by λ, R is the distance
between transmitter and the receiver and polarization loss
factor represents by cos φ. As in (1), it is proved that the
receiver antenna needs to have high gain in order to harvest
the maximum power possible.

C. Typical Applications
Fig. 1. The schematic diagram of the wireless energy harvesting module.
As a result of RF-EH technique, various forms of practical
applications such as WSNs, wireless body networks and wire-
between antenna size and performance [9]. The amount of less charging systems for wireless devices have been enabled
harvested energy can be dramatically increased by correctly by RF-EHNs [28], [29]. In this technique, radio signals are
arranging antennas with the same matching circuit and power used as channels to transfer energy in the form of electromag-
management or operating antennas at different frequencies. netic radiation with a frequency range from 300 GHz to a value
RF antennas can harvest energy from a variety of sources as low as 3 kHz. Moreover, RF energy transfer can be consid-
such as mobile phones (900 − 950 MHz), local area networks ered as a far-field power transfer technique, as the transmitted
(2.4 GHz − 5.8 GHz), Wi-Fi signals and broadcast Ultra High electromagnetic signal cannot retract on the transmit antenna
Frequency (UHF) TV signals. However, this also comes with at a distance of above λ/2π [30]. At the long range, RF
several mitigating factors such as path loss and shadowing. energy can be used in different communication technologies
2) Matching Network: The main function of a match- such as D2D communication, IoT, cooperative communication
ing network is to increase input voltage of the rectifier and cognitive radio networks. All these emerging communi-
and to minimize the transmission loss from the antenna to cation technologies and the use-case scenarios of RF-EH are
the rectifier. When impedance of the loads and impedance discussed in Sections V and VI. Experimental data of RF-EH
at the antenna output are matched together, the maximum from different sources are listed in TABLE II. Design issues in
power transfer can be achieved, and is known as impedance RF-EH networks differ for various emerging communication
matching. systems. Therefore, design issues and antenna architecture are
3) Rectifier: The Main function of the rectifier is to convert discussed under Section IV and VI in relation to emerging
captured RF signals to the DC voltage levels. One of the most technologies.
important challenges of the rectifier design is to generate a
considerable voltage (battery-like voltage) from the received D. Challenges
RF power. RF to DC conversion efficiency is determined by A few cognitive studies related to humans reviewed in [31],
the diode in the circuit of the rectifier. Therefore, the diode depicted that in the presence of RF radiation, there can be a
is recognized as the main component of the rectifier circuit. significant delay in terms of physical response and memory
In general, a higher rectifying efficiency can be achieved by reactions. Moreover, intense exposure to RF radiations can
using a diode with a lower voltage [27]. cause severe heating in biological tissues [32]. Therefore, the
scientific community with the help of government agencies
B. Conditions for Efficient RF WPT must gain more thorough understanding of potential health and
In RF-WPT , RF power at the receiving antenna (Pr ) is safety impacts associated with the usage of WPT/SWIPT in
determined using the Friis free space equation. If receiver public settings. Transfer distance is one of the major concerns
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 271

in RF EH, but due to the inverse-square law, RF energy transfer process and health concerns related to high power applica-
is limited in terms of its coverage range [11]. The directivity tions [9], [10]. Therefore, most WPT research continued in
and the gain of the receive antenna(s) critically affect RF-EH two distinct regions, non-radiative (near-field) and radiative
rate. Thus, developing high gain antenna which can support (far-field). The fields in these regions have distinct character-
a wide range of frequencies is also an important research istics and different WPT techniques have been used to transfer
direction. Impedance matching is one of the major compo- the power.
nents in Rectennas, which determines the amount of harvested
energy delivered to the rectifier. Thus, proper circuit design
techniques are required, which automatically tune the param- A. Near-Field WPT
eters to reduce impedance mismatch. With the rapid advances The near-field or non-radiative region is the area of one
in communication devices, RF-EH module needs to be small wavelength of the transmitting antenna. Power can be trans-
enough to be embedded in low-power devices. However, it is ferred by employing inductive coupling, resonant inductive
challenging to reduce the size of the RF-EH module while coupling, capacitive coupling or air ionization (lightening) in
maintaining high EH efficiency. Further research works are the range of tenths of Watts [9]. In the near-field region,
required in reducing the size of the RF-EH module. The sen- the WPT system requires high power transmission efficiency
sitivity of an information receiver is higher than the receiver (PTE) and can be attained above 80% PTE [9], [52]. Power
used in RF-EH. Thus, the SWIPT scheme cannot be used effi- leaves the transmitter only when there is a receiver or any
ciently in some of the communication scenarios. Therefore, absorbing materials within the wavelength, which depends on
the sensitivity of the RF-EH receiver needs to be improved the shape and size of the antenna on the transmitter. Therefore,
in order to have an efficient SWIPT. RF powered devices power transmitted decreases exponentially with the distance
have a strict power constraint and it is difficult to support between the transmitter and receiver, i.e., meaning if the
algorithms that require high computational power. Therefore, distance between transmitter and receiver is larger than the
all modulation techniques, channel coding, routing protocols wavelength of transmission medium, very little power can be
and receiver operation policies need to be redesigned for RF harvested. Coupled mode theory has been used to analyse the
powered communication networks. characteristics of near-field WPT in [53] and a demonstration
has been conducted by the authors using coupled antennas at
E. Summary and Insights the resonant frequency and 40% PTE was gained at a distance
of 2m. It is also demonstrated in [9] that having a mediat-
In this section, we reviewed RF-EH and the components of
ing resonant antenna in between receiving and transmitting
RF-EH module following future directions. The main idea of
antennas could improve PTE in WPT. Also, it is known that
this section is to provide a foundational knowledge to RF-EH,
coupling distance and the orientation of the antennas have a
which is essential to understand the content of the paper. First,
big impact on optimum source and load impedance [54].
we provided an introduction to RF-EH and its importance
1) Inductive Coupling: In inductive coupling (inductive
in wireless communication. Next, we reviewed the circuit
power transfer, IPT) used in the near-field WPT, electrical
design of RF-EH module and state-of-the-art circuitry imple-
energy is transferred via coils by means of a magnetic field and
mentations. Then we explained typical applications of RF-EH.
high efficiency can be achieved when the coils are very close
Finally, we provided future directions and practical challenges
to each-other [33]. According to Ampere’s law, an oscillating
in RF-EH circuit and techniques. It is shown that the EH
magnetic field has been created, once an Alternating Current
efficiency mainly depends on the power level of received RF
(AC) passes through the transmitter coil. Once that magnetic
signal and the design of RF-EH circuit. The design of RF-EH
field induces a receiving coil, it creates AC in the receiver.
circuit design needs to be modified according to the minimum
Subsequently, a rectifier in the receiver converts it to Direct
EH requirement of the system and the transmission range of
Current (DC) (or drives the load directly). Few applications
RF signal to improve overall efficiency of the system.
work at 50/60 Hz, which is used in inductive coupling to
recharge devices such as electric tooth-brush stands.
III. W IRELESS P OWER T RANSFER (WPT) However, in most applications, AC current is applied only
WPT is an innovative concept that was originally devised after increasing the AC frequency of an electronic oscilla-
by Nikola Tesla in the 1890s. WPT refers to the transmis- tor, since transmission efficiency can be improved with higher
sion of electrical energy from a power source by means of frequency [34]. This technique is used in wireless charging for
electromagnetic fields, to an electrical component or a portion consumer electronic devices such as laptops, cellular phones
of a circuit that consumes electrical power without the aid and other portable devices, and also used to charge electric
of wired interconnections. The WPT system contains a trans- automobiles. IPT has been successfully used in Electronic
mitter connected to the main power source, which transforms Vehicle (EV) systems [35]. In early stages, charging a pad-
main power to a time fluctuating electromagnetic field and one dle has been used to recharge the battery of EV. However,
or more receiver devices to receive and harvest energy from the this system is connector-less, but not wireless.
electromagnetic field as shown in Fig. 2. Two factors, long dis- 2) Resonant Circuits: Later, resonant circuits are added to
tance transmission and high power have been studied in early enhance the inductive power transmission. Therefore, the same
attempts with WPT. Nonetheless, further development of these amount of power transferred using inductive coupling, can be
factors ended due to low efficiency of the power transmission now transferred at a greater distance. Nikola Tesla discovered
272 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 2. RF Wireless Power Transfer - Devices such as cellular phone, sensors, drones, electronic devices in vessels receive and harvest energy from RF
signals and recharge their batteries.

Resonant IPT, which is one of the most well-known current the transmitter generated alternative voltage is applied to the
WPT technologies, enabled by modern electronic components transmitting metal plate and electrostatic induction causes an
such as batteries of electric vehicles [38]. In resonant inductive alternating electrostatic potential on the receiver, metal plates,
power transfer (RIPT), power is transferred by magnetic fields which is the alternative source of current flow in the load cir-
between the transmitter and receiver resonant circuits which cuit. Efficiency of the power transferred depends on the factors
consist of a coil of wire attached to a capacitor or a self- such as frequency, square of the voltage and the capacitance
resonator or a resonator with internal capacitance. There are between transmitter and receiver plates [43]. Moreover, the
several advantages of resonant IPT over IPT including greater amount of power transferred is corresponding to the area of
range, reduced electromagnetic interference, higher frequency smaller plate in the WPT system. Use of capacity coupling
of operation and higher efficiency. Nonetheless, higher oper- in high power applications can be hazardous and can produce
ating frequency in kHz range is one of the main advantages of poisonous ozone gas due to the requirement of high voltage
resonant IPT that can be aided by the current state-of-the art on the electrodes to transmit significant amount of power.
electronic technologies. Wireless power coverage is one of the Therefore, most of the time, the capacitive coupling WPT
possible areas for resonant IPT that can be used to power lights technique has only been used in applications which require
and recharge mobile batteries anywhere in a room without comparatively low power. Nevertheless, capacitive coupling
using wired connections. Another application of resonant IPT has several advantages over IPT, such as limited interference
is public transit systems that supply power across the width of and the alignment required by both the transmitter and receiver
a roadway surface for recharging EVs batteries and to power are less critical. Recently, capacitive coupling has been applied
people mover systems such as auto guided electric buses, to recharge batteries of electric portable devices and are
trains etc. [39]. Wireless powering or recharging small elec- also being used to transfer power between substrate layers
trical devices could drastically reduce the quantity of batteries of an integrated circuit. Moreover, to extend the transmit-
disposed, toxic waste and ground water contamination. ting range, resonance can be used with inductive coupling.
3) Capacitive Coupling: Capacitive coupling is another Initial experiments with both resonant and magnetic cou-
near-field technology used to transfer power by using electric pling were completed by Nikola Tesla at the turn of the 20th
fields between electrodes like metal plates. In this technique, century.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 273

TABLE III
D IFFERENT T ECHNOLOGIES IN WPT

B. Far-Field WPT connections. Nevertheless, with the advancement of communi-

Far-field region describes the area where the distance is cation technologies, there is a need of transferring power using
much larger than the diameter of the transmit antenna and omnidirectional antennas to improve the transmission region.
was discovered immediately after the discovery of high power Therefore, more future studies are needed in the far-field WPT
microwave amplifiers [48]. Far-field power transmission has to improve directivity and energy transmission efficiency in
enabled energy to be transferred over greater distances with- different communication technologies.
out using a wired electrical network and provides motivation
for further research to improve the efficiency, performance D. Summary and Insights
and robustness of WPT applications. The most well-known WPT is a concept that refers to the wireless transmission
forms of electromagnetic radiation used for far-field WPT of energy from the power source to an electrical device intro-
are microwaves and laser beams. According to the Rayleigh duced in the late 1890s by Nikola Tesla. WPT has two distinct
criterion, over distance any kind of radio waves, such as categories, non-radiative (near-field) and radiative (far-field)
microwaves and laser beams will become weaker [51]. Having based on the power transmission regions. In the near-field,
large transmitting antenna in a WPT system compared to the power can be transmitted within an area of one wavelength
radiation wavelength, can tighten the beam and spread shorter of the transmitting antenna which can be up to 1/2 meters.
transmitting distance. When the diameter of a transmitting Inductive coupling, resonant inductive coupling and capaci-
antenna is smaller than the wavelength, this can also cause tive coupling are techniques that can be used in the near-field
excessive losses due to side-lobes. The microwave power beam WPT. Far-field WPT can enable the transmission of RF power
is less vulnerable to atmospheric attenuation originated from over several kilometres. Microwave and laser beams are the
water vapour or dust and more efficient than laser beams. form of electromagnetic radiation used in the far-field power
Microwave power beam is proposed to apply in space applica- transmission. However, these techniques are only supported
tions such as energy transmission from solar power satellites for point-to-point communication. Far-field techniques need
to earth bases in [49]. However, practically it is difficult to considerable refinements to support omnidirectional power
use microwave power beams for most space applications due transmission before handshaking with modern communication
to the requirement of large aperture size. In an investigation of technologies.
solar power satellites in 1978, NASA required a transmitting
antenna having 1 km diameter [50]. However, these sizes can IV. S IMULTANEOUS I NFORMATION
be reduced by having shorter wavelengths, though it can cause AND P OWER T RANSFER (SWIPT)
atmospheric absorption. Moreover, an area of 10 km diame-
SWIPT is a recently developed technique out of various
ter receiving array permits the use of total power levels up to
WPT technologies and it enables the simultaneous transfer
750 megawatts. These can be found in many modern electric
of information and power wirelessly, as shown in Fig. 3.
power plants under human safe power density of 1mW/cm2 .
However, fundamental design changes are required in wire-
A summarized table of different WPT technologies discussed
less communication networks to have an efficient SWIPT.
are listed in TABLE III.
Reception reliability and information transfer rates are con-
ventionally used to assess the performance of the wireless
C. Challenges and Future Work networks [8]. Moreover, the trade-off between information
The major challenge in WPT is the limitation on its trans- rate and harvested energy level becomes an important fac-
mission regions. Maximum energy efficiency that can be tor to evaluate the system performance once users in the
achieved in the far-field usually do not exceed 50% [55]. system perform EH using RF signals [8]. In [8], an ideal
There are existing systems built using microwave signals with receiver is used, which has the ability to perform Information
high gain antennas, which can transfer power over far-field Decoding (ID) and EH simultaneously. Two circuit receivers
regions with 90% of power transmission efficiency [56], [57]. are used in the receiver architecture proposed in [58] to per-
However, these systems need a point to point or line of sight form EH and ID separately. Furthermore, applications of smart
274 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 3. SWIPT via static and mobile base stations (drones and ships). Arrow heads defines the direction of information and power flow. Idle users harvest
energy from RF signals from static and mobile base stations. The active users transmit and receive information from static and mobile base station. Active
users 01, 02 and the mobile base station(drone) engage in SWIPT with static and mobile base stations.

antenna technologies in SWIPT MIMO and relaying were con- separate antennas, which are served by a transmitter with
sidered in [13]. The possibility of further improvements in multiple antennas [58]. Moreover, these two separate antennas
WPT energy efficiency is opened up by the use of these smart observe different channels. This receiver architecture easily
antenna technologies. can be implemented using off-the-rack components for the EH
In general, it is not possible to perform EH and ID oper- and ID receivers. This architecture allows performing both EH
ations on the same received signal in a SWIPT system, as and ID concurrently and independently. Channel state infor-
the EH operation on the RF signal destroys the information mation (CSI) and receiver feedback can be used to optimize
content of the signal. Moreover, a single antenna receiver the trade-off between achievable EH and information rate.
may not be able to facilitate a reliable energy supply, due
to the limited collection of energy. Hence, in order to prac- B. Time Switching Receiver
tically achieve SWIPT, either the received signal must be TS architecture, also known as co-located receiver archi-
divided into two, or separate antennas need to be used for tecture, share the same antenna for EH and for information
both EH and ID. In addition, to generate sufficient power reception. The receiver used in this architecture contains an RF
for reliable devices, a centralized or distributed antenna array energy harvester, information decoder and a switch to change
deployment (MIMO, relaying, etc.) is required [13]. In the the type of receiving antenna in the system. Based on a TS
following section, we provide an overview of several SWIPT sequence, the receiver antenna or antennas change between
enabled receiver architectures, including separate receiver, ID and EH circuit periodically. Furthermore, the TS receiver
Time Switching (TS), Power Splitting (PS) and Antenna also requires accurate information/energy scheduling and time
Switching (AS) architectures as shown in Fig. 4. synchronization. At the time receiver operates in EH mode,
A. Separate Receiver amount of harvested energy by the receiver j from the source
i can be calculated as follows:
In this separate receiver antenna architecture, both ID
and EH circuits are included as two separate receivers with Pi,j = ηPi [hij ]2 , (2)
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 275

Fig. 4. Integrated receiver architecture designs for SWIPT. (a) separated receiver architecture where energy and information decoding have separated receivers.
(b) Time switching architecture: based on a pre-defined time factor, connected antennas will be allocated a time to harvest energy and information decoding.
(c) Power splitting (PS) architecture: the PS receiver divides the received signal into two power streams based on a certain PS ratio. (d) Antenna switching
receiver: antenna switching between harvest energy and information decoding based on a optimization algorithm.

where η equals to the efficiency factor of EH process, Pi the PS receiver j from the source i can be calculated as follows:
denotes the transmit power at source i, and hij represents the
Pj,i = ηPi |hij |2 θj . (4)
channel gain between source i and receiver j. When the same
receiver is changed to ID mode, ID rate can be calculated If the power of signal processing noise is denoted by the σsp
2,

using the following equation [11], where W and σ 2 represent maximum ID rate of receiver j decoded from the source [11]
the transmission bandwidth and noise power, respectively. i can be written as
 
  (1 − θi )Pi h2ij
Pi h2ij Rj,i = W log 1 + . (5)
Ri,j = W log 1 + . (3) σ 2 + σsp
2
σ2
Theoretically, it is recognized that power-splitting achieves the
best trade-off between information rate and amount of RF
Additionally, the transmit signal and TS sequence can be
energy transferred [62].
optimized together for different objectives of system design
based on QoS requirement of energy transfer and the channel
D. Antenna Switching Receiver
statistics.
Low complexity antenna switching between EH and ID can
be used to enable the SWIPT [63]. For example, while a sub-
C. Power Splitting Receiver set of antennas at the receiver work on ID, the rest of the
The PS receiver divides the received signal into two power antennas can work on EH. Compared to time-switching and
streams of different power levels with a certain PS ratio before power-splitting, the antenna switching is comparatively easier
signal processing is performed at the receiver. Afterwards, both and appealing for practical SWIPT architecture designs. The
power streams are sent to an information decoder and energy dual antenna receiver architecture proposed in [64] can also
harvester to make possible simultaneous ID and EH [59], [60]. be adopted. In addition, with a proper antenna switching pro-
Apart from the receiver circuit, no additional change is tocol, it is possible to extend this architecture easily to include
required in the power-splitting design of conventional com- a larger number of antennas. Sometimes antenna switching is
munication systems [61]. In each receiver antenna, the PS also considered as a distinct case of the PS architecture [64].
ratio can be optimized. Furthermore, by varying PS ratios, Moreover, antenna switching architecture can also be used to
the information rate and the harvested energy can be balanced optimize a separate receiver architecture [58].
according to the system requirements. Overall performance
also can be improved by optimizing the combination of the E. Summary and Insights
signal and the PS ratios. Let θ denote the power-splitting coef- In this section, we provided a brief overview of SWIPT
ficient value for receivers. The amount of power harvested at technology along with four different receiver architectures.
276 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

TABLE IV
SWIPT R ECEIVER A RCHITECTURES

SWIPT is a recently introduced technology that facilitates both interference avoidance/mitigation techniques help to maintain
information and power transfer concurrently in communica- effective communication between multiple users, these tech-
tion networks. However, due to different sensitivity levels in niques also can cause inefficient use of available wireless
both ID and EH operations, different receiver architectures resources [73]. For an instance, an effective way of utilizing
are needed to facilitate the SWIPT. The received signal has wireless resources can be to allow the nodes in the network
to be split into two distinct parts, one for ID and the other to decide on whether to mitigate or cancel interference.
for EH. Furthermore, we provided the techniques that have
been proposed to achieve this signal splitting into different B. Basic Analysis of Interference in a Two User Link
domains such as time, power and antenna. Summary of the
This section provides a simple analysis of instantaneous
existing SWIPT architectures are listed in TABLE IV. Also
interference in a two-user transmission scheme. A general two-
SWIPT under different emerging technologies is explained in
user transmission link is shown in Fig. 5. We use symbol u1
Section VI with recent literature.
and u2 to represent the symbols of user 01 and user 02 respec-
tively. In this model, we use Binary Phase Shift Keys (BPSK)
V. I NTERFERENCE E XPLOITATION IN SWIPT as the modulation technique for the simplicity, which leads to
A. Interference in Communication Networks u1 = 1, u2 = −1. Moreover, the noise at the receiver has been
Interference in wireless communication is anything that dis- ignored and ρ represents the interference channel and the loss-
rupts or modifies a transmitted signal as it transmits along less channel from the intended transmitter to the receiver. The
the channel between a transmitter and a receiver. In tradi- received signal y can be expressed with these basic parameters
tional wireless communication, interference is treated as a as follows [74]:
limiting factor which prevents achieving substantial quality of y = u1 + u2 .ρ, (6)
experience (QoE) and limits the overall system performance.
Interference adds variance to the transmitted signal which dete- where u2 .ρ represents the interference. In Fig. 6, two dis-
riorates the final result [71], [72]. One of the most important tinct cases of interference are shown, which have the same
design objectives of traditional communication systems is to transmit power which is equal to one. The first case gives
provide low or no interference. Therefore, much great effort is the destructive interference where ρ = 0.5 and it gives
needed to avoid, mitigate or cancel interference. Even though y = 0.5 according to (6). Moreover, destructive interference
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 277

certain conditions to a useful signal that potentially improves

the SINR of the downlink link, resulting in improving the
systems energy efficiency. Using similar set up, constructive
interference has been investigated in the work [81] with the
knowledge of both DI and CSI at the transmitter. Recently,
some interesting beamforming strategies have been proposed
to introduce a hybrid analog-digital precoding, and to exploit
the concept of mutual coupling between the transmitting
antennas by using tunable antenna loads [80].
Fig. 5. A two-user link example with Interference. In the work of [82], it has been shown that the
interference does not limit the capacity of the broadcast chan-
nel. However, many works proposed Successive Interference
Cancellation (SIC) instead. The employment of SIC in SWIPT
enabled bipolar adhoc networks has been studied in [75].
They demonstrated how each receiver in the network can
make the use of SIC to raise WPT without affecting the ID
process. Moreover, results demonstrated that SIC is notably
beneficial for SWIPT enabled communication systems. An
Opportunistic Communication (OC) based SWIPT scheme
in Interference Alignment (IA) networks has been proposed
Fig. 6. Transmission scenario in two-user link with destructive and in [83]. In this work, OC based antenna and user selec-
constructive interference. tion are utilized to achieve SWIPT. Furthermore, this paper
summarized the methods for reducing the complexity of the
OC based AI algorithms. Timotheou et al. [84] investigated
from user 2 has shifted the received symbol of user one
the constructive interference in MISO downlink to improve
towards decision threshold in BPSK constellation denoted by
the performance of both the EH and ID. In addition, they
the dotted line in Fig. 6. Thus, the received power of user
introduced a new precoder design for SWIPT, which signifi-
01 has been reduced and this phenomenon affects the system
cantly reduces the transmit power and applies to the scenarios
performance. However, in the second case, interference from
having lower number of transmit antennas than the num-
the user 02 is constructive [74], with ρ = −0.5. From the (6),
ber of users. Additionally, residual Self-Interference (SI) in
we obtain y = 1.5. Now with the constructive interference,
multi-user MIMO channels has been studied in [85].
received power of user 01 has increased with the help of
the interference from user 02. Thus symbol detection is toler-
ant with constructive interference rather than in the situation D. Future Directions
without interference.
To cope up with the complexity involved in wireless com-
munication systems, low-complexity resource allocation and
C. Interference Exploitation in SWIPT optimized resource sharing are required in the future SWIPT-
Recently, due to identification of new aspects of enabled wireless networks. Therefore, interference manage-
interference, instead of mitigating or avoiding, focus has ment will continue to be a growing challenge in this domain.
been on the potential use of interference in wireless systems. New perspectives on the use of interference are required to
Innovative ways of using interference are now emerging, maintain the expected QoS and to meet quality and technical
since traditional mitigation techniques are no longer optimal. expectations of 5G communication. For instance, by making
Thus, interference exploitation can improve the reliability, the use of simple linear operations to mitigate interference
security and the achievable rate of wireless communica- in future massive MIMO employed networks, interference
tion systems. In modern communication, interference plays a will be available in the network to use for other activities
notable role in both information and power transfer in wireless such as EH, secure communication, etc. without degrading
communication systems [75]. SWIPT has been studied with the overall performance of the system. This could motivate
interference broadly in multi-user systems [72], [76]–[78]. new applications and services in the next-generation wireless
Although interference links are usually harmful for the ID pro- communications.
cess, they can be useful for EH. However, a proper trade-off
between ID and EH is required to facilitate efficient SWIPT
in multi-user systems. The idea of beamforming optimization E. Summary and Insights
has achieved an interesting level of gain by collecting useful In this section, the idea of interference exploitation in
signal power from the scenarios with constructive interference. SWIPT has been presented along with the recent literature.
Alodeh et al. [79], [80] proposed a multi-user symbol- The perspective of interference has changed with the evo-
level precoding approach to address the multi-user interference lution of communication networks. In the present, instead
through jointly utilizing the CSI and Data Information (DI). of mitigating or cancelling interference, the potential use of
The interference among the data streams is converted under interference in communication networks has received recent
278 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 7. Enabling SWIPT techniques for emerging wireless communication areas.

attention from the scientific community. Importance of con- the field. It should be noted that, all these SWIPT emerging
structive interference has been explained by using a two- technologies are used in the Far-Field WPT to transfer power
user interference network. Next, we provided an overview within the communication systems.
of research activities conducted in the area of interference
exploitation and the context of harvesting interfering power to A. Multi-Carrier SWIPT Systems
improve the energy efficiency of the communication system. Orthogonal Frequency Division Multiplexing (OFDM) is a
Although the scientific community has already focused on widely used multi-carrier technique in various wireless stan-
exploiting interference in a number of different scenarios, dards. Several existing papers have studied the performance
the area of interference exploitation is quite broad having of OFDM-based SWIPT systems [60], [64], [66], [86].
the potential in revisiting existing interference cancellation Grover and Sahai [86] have investigated the performance of
techniques in interference limited communication scenarios. SWIPT receivers considering a single user OFDM channel
In future upcoming generations of wireless communication and have provided an upper bound for the rate-energy trade-
networks, power efficiency will be a major concern and will off for the considered system structure. It has been shown
play a major role in system design. In this regard, interference that there exists a trade-off between the achievable rate and
aided EH as RF power and as useful signal power can be the harvested energy, leading to the well known rate-energy
a promising enabling solution in 5G. In addition, the con- trade-off in the considered system set-up. Furthermore, the
cept of constructive interference will change the perspective of contribution in [64] studied SWIPT for BWSs, which employ
the research community on the interference in communication OFDM and beamforming transition in order to create a set
networks. of parallel sub-channels with the objective of simplifying
the resource allocation mechanism. Subsequently, the authors
VI. E MERGING SWIPT S CENARIOS proposed power control mechanisms for SWIPT in the context
In this section, we provide SWIPT enabled emerging tech- of a multi-user multi-antenna OFDM setting, taking circuit-
nologies in different application areas. Fig. 7 and TABLE V power constraints into account. However, in [64], only the
provide various enabling SWIPT techniques for emerging carriers which have been assigned to a particular user are uti-
wireless communication areas such as massive MIMO, bistatic lized for EH purposes for that particular user, assuming a fixed
scatter radio, cooperative NOMA, multi-user MIMO, BWSs, subcarrier allocation.
etc. All these emerging technologies are covered in the fol- Furthermore, the contribution in [60] studied the
lowing paragraphs in conjunction with the latest literature in performance of SWIPT in an OFDM-based multi-user
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 279

TABLE V
E MERGING C OMMUNICATION T ECHNOLOGIES IN SWIPT

(To be continued)

single-antenna system by employing a PS technique at the It has been demonstrated that system energy efficiency can
receiver. Subsequently, authors analysed the following two be enhanced by employing RF EH in the interference lim-
sets of receiver architectures: ited regime. Multi-antenna receivers are beneficial to enhance
i) the first set considered arbitrary PS ratios in order to the system capacity rather than enhancing the system energy
split the received power into continuous sets of power efficiency [60].
streams. Moreover, Zhou et al. [66] have studied the performance
ii) the second set considered fixed PS ratios in order to of SWIPT systems in OFDM-based multi-user wireless
divide the received power level into a discrete set of systems, with broadcast transmission from a fixed access
power streams. point to a set of distributed user terminals. In their analysis,
280 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

TABLE V
C ONTINUED

(To be continued)

authors considered two multiple access schemes, namely, Time two settings, Zhou et al. [66] subsequently addressed the
Division Multiple Access (TDMA) and Orthogonal Frequency problem of weighted sum-rate maximization over all the users
Division Multiple Access (OFDMA). In the first scheme, the by adapting the time/frequency power allocation and either the
TS approach is used in such a way that the information receiver PS or TS ratio considering the constraints of minimum har-
of a particular user operates in the scheduled time slot for that vested energy constraint on each user and also total and/or
user, whereas the energy receiver operates in all other time peak transmit power constraint.
periods. However, in the second approach, the PS approach Future Directions: Most of the existing SWIPT literature
is applied at each receiver, considering all subcarriers share has focused on narrowband receivers for EH and ID pur-
the same PS ratio at each receiver. In the above mentioned poses. In this context, it is interesting to investigate suitable
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 281

TABLE V
C ONTINUED

RF EH techniques in order to harvest energy based on a wide- that a wireless network or node can be aware of during the
band transmitted signal and also to explore wideband SWIPT communication process. Spectral sensing based CR is used
receiver architectures. One of the promising research direc- to identify channels in the radio frequency spectrum. There
tions in this area is to investigate suitable techniques to enable are three main architectural designs for CRN; infrastructure
the utilization of all the unused carriers for EH and the infor- based, ad-hoc and hybrid network [87]. Infrastructure archi-
mation bearing carrier for ID. In this case, the SWIPT receiver tecture contains base station or access points through which all
should be capable of working in multiple frequency bands. device communication and routing operates. There is no cen-
Moreover, it is an interesting research direction to explore RF tral entity for ad-hoc architecture and device communication
EH from multiple sources, which are operating in different using the existing communication protocols. Hybrid architec-
sets of frequencies simultaneously. ture of CRN is a combination of both infrastructure and ad-hoc
architectures.
Integration of RF-EH to CRNs can provide a spectrum and
B. WPT/SWIPT Enabled Cognitive Radio (CR) energy-efficient wireless communication system [26]. In the
The basic requirement of CR is to be aware of surround- work of [26], Lu et al. highlighted main differences between
ing radio environment and take decisions intelligently to use traditional CRNs and RF-EH enabled CRNs. Then the authors
available resources. For example, CR networks contain trans- conducted a study regarding trade-off among spectrum sens-
mitters and receivers that can intelligently recognize idle ing, information transmission and RF-EH. Lee et al. [159]
communication channels. Moreover, CR autonomously adapts proposed a new CR network architecture in which secondary
according the environment conditions by changing operational users harvest RF energy from the active primary transmissions.
parameters such as transmit power, operating frequency, mod- Finally, the authors have proved that this new architec-
ulation and etc. [90]. This intelligent selection optimizes the ture can be applied to WSN powered distributed wireless
use of available RF spectrum while reducing interference to power communications. A SWIPT enabled CRN architec-
other users in the network. There are two different types of ture is shown in Fig. 8. The only difference in the figure
CR, spectrum-sensing based CR and full coordination based compared with the general architecture of CRNs is that the
CR [87]. Full coordination based CR considers all attributes secondary user participates in both information and power
282 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 8. A network architecture of SWIPT enabled cognitive radio network. There are three zones in around the primary base station: energy harvesting,
information transmission and interference zones. Secondary user which located in EH zone can harvest energy and receive information simultaneously. Primary
user in the transmission zone receive information from primary base station. Users can not transmit data if those are in the interference zone.

reception simultaneously. As depicted Fig. 8, secondary user user has to decide whether to switch to another channel for
can harvest energy while communicating with the primary base information transmission or EH based on the availability of
station and could possibly work as a relay in communica- channels. In this direction, future research works should focus
tion between the primary base station and the primary user on suitable decision making methods in order to maximize the
located in the interference zone. Optimal resource allocation performance of the network.
for harvested energy maximization in wideband CRN with
SWIPT has been focused on the work of [88]. Furthermore,
Hu et al. [88], initially formulated the sum of harvested energy C. SWIPT Enabled Full-Duplex Communication
maximization problem and then proposed a new algorithm In general, full-duplex communication means that both
to solve the problem and obtain the optimal power and sub- ends of the transmission link can transmit and receive sig-
channel allocation. A cooperative access scheme for efficient nals simultaneously. Around 1940, the principle of full-duplex
SWIPT transmissions in CRNs has been proposed in [89]. communication was used in radar systems [160]. Use of full-
El Shafie et al. [89] proposed a low-complexity multiple relay duplex communication was not encouraged until recently, due
selection with gain control schemes based on compressive to the bottleneck of SI, which occurs due to the signifi-
sensing principles. cant power difference between the transmitted and received
Future Directions: Spectrum sensing, access, management signals. Due to the tremendous implications of full-duplex
and hand-off are the four main functions to support efficient communication for the network design, researchers are encour-
and intelligent dynamic spectrum access in CR. Detecting aged to find solutions for SI. Enormous efforts and progress,
activities of the primary user is the main function of spec- in both academia and industry, have been made in SI
trum sensing. Spectrum sensing is a challenging problem in Cancellation (SIC) [91], [93], [94]. SIC techniques in full-
SWIPT enabled CRNs with the complexity of received spec- duplex transceivers are classified into two categories called
trum signatures from multiple devices with noise and channel active cancellation and passive suppression [95]. In pas-
impairments in the network. Furthermore, required an opti- sive suppression, before the SI signal is processed by the
mized algorithm, which could find a spectrum opportunity circuit of the receiver, it is suppressed in the propagation
for both information transmission and EH. The main issue domain. Subtracting a processed copy of the transmitted sig-
in spectrum access in CRNs is how to provide access to nal from the received signal, the SI signal is mitigated in
available spectrum while providing fair and efficient sharing active cancellation. Furthermore, active cancellation can be
of the available spectrum while protecting the primary user divided into two categories based on the signal domain, digital
from collision [11]. It is interesting to investigate the suitabil- and analogue cancellation techniques. Full-duplex communi-
ity of radio resource allocation during both information and cation is already recommended to be used in wireless small
power reception. High spectrum utilization for both informa- cells and 5G systems [160], [161]. Moreover, the application
tion transmission and RF-EH can be achieved by executing of full-duplex communication technology to physical layer
channel selection algorithms. The most valuable metrics in security and cognitive radio systems has also been investi-
channel selection, channel occupancy probability and achiev- gated [162]–[164]. In addition, motivation for applying CRs
able EH rate need to be further investigated in SWIPT enabled with the full-duplex as compared with the half-duplex scenario
CRNs. In addition, spectrum hand-off is one of the important has been comprehensively explained in [16].
functionalities in SWIPT enabled CRNs. During the time the Combining full-duplex and SWIPT techniques will
primary user re-uses or releases its channel, the secondary help to improve the network communication performance.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 283

In [97] and [98], SI is converted into an extra source of the overall performance of SWIPT enabled full-duplex systems
energy in full-duplex wireless powered communication. Full- is unknown. Further research on transmission schemes and
duplex SWIPT bi-directional communications, where TS or interference exploitation is required to balance the overall
PS architecture applied to one or more nodes in the system performance of the system in EH and information transmis-
is studied in [67], [68], and [96] to perform ID and EH. sion. Also, effective combination of full-duplex-CRNs with
In [67] with an assumption of perfect SI cancellation, the SWIPT also can be a promising research direction. In full-
authors formulated an optimization problem to minimize the duplex-CRNs, the dedicated sensing and transmission antennas
weighted sum transmit power of the full-duplex system. An require more energy and therefore, there are plenty of future
antenna pair selection scheme has been produced in [68] research opportunities in SWIPT enabled full-duplex-CRNs to
to improve the performance of a full-duplex bi-directional reduce energy consumption [16]. Comparatively, there is less
system. In [96], transmit power and PS ratio were anal- research work done on the MAC layer aspects of full-duplex
ysed with dual antenna nodes. Full-duplex relay topology was than on the physical layer. Therefore, the knowledge of MAC
studied in [97], [161], [165], and [166] with a three-node full- layer operations from an energy consumption viewpoint is not
duplex relay channel and WPT. In [97] and [98], WPT has well established. In this direction, significant research efforts
been proposed where full-duplex nodes harvest energy through must be dedicated to model the behaviour of the full-duplex
the SI channel. SWIPT with full-duplex communication has MAC layer. Moreover, the design of cross-layer algorithms
been studied in [166] with a multi antenna relying system. that can utilize the information available at three layers, i.e.,
The source and the relay transceiver were designed according physical, MAC and network layers, is required.
to mean-square-error criterion. In full-duplex hybrid access
point topology, information transmit to users using uplink and
received energy to users from downlink channel simultane- D. Bistatic Scatter Radio for Energy Harvesting
ously was studied in [45] and [100]–[102]. The work of [99] A scatter radio is defined as a radio which works based
studied throughput of full-duplex relaying in RF-EH systems on the principle of communications by means of reflected
with having the relays equipped with two antennas, one for power [104]. This communication principle is commonly used
information transmission and one for information reception. in Radio Frequency Identification (RFID) systems, which are
Moreover, it was demonstrated that using both relay antennas extensively being used in object tracking and supply chain
for EH during EH time is always beneficial and compared monitoring. In RFID systems, a RFID tag communicates its
to half-duplex relaying architecture, the results of the [99] identification number to a reader and additional information
indicate that full-duplex relaying can improve the system can be stored in its memory in some cases [105]. For short dis-
throughput with optimal time split. References [100] and [101] tances (less than 2-5 m), the tags (also called passive tags) can
investigated a hybrid full-duplex access point, where it broad- derive their received power from the signal received from the
casts energy to a set of users and receives information reader, and the communication distance is determined by the
from another set of users using the technique time division ability of the tag to extract the energy level from the received
multiple access. A resource allocation algorithm was intro- signal. However, for moderate/long distance applications, the
duced in [102] for a SWIPT system which consisted for a tag (also called semi-passive tag) consists of a battery in order
full-duplex base station, single antenna users and multiple EH to operate the backscatter modulator and tag state machine and
antennas. In [103], full-duplex MIMO communication network use the same backscatter mechanism for the communication
has been investigated. In that scenario, up-link users first har- purpose [105].
vest energy from the base station. After harvesting energy, In the commonly used RFID systems, a carrier emitter
up-link users transmit information to the base station, while and a reader are placed in a single reader box, which is
the downlink users receive information from the base station. called a monostatic RFID architecture. In other words, the
In certain communication networks, full-duplex shows a better transmit antenna, which generates the carrier signal and the
throughput than the half-duplex mode. Thus, the integration receive antenna, which is used to demodulate the reflected
of full-duplex with SWIPT can be a promising approach to signal coming from the RFID tag, are components of the
improve the energy efficiency and spectral efficiency of future same equipment [106]. However, there are several disadvan-
wireless communication networks. tages of this monostatic architecture such as high round-trip
Issues and Future Directions: Full-duplex communication path loss and limited communication ranges with the passive
is an enabling technology for 5G as it improves the through- tags. In order to address the aforementioned disadvantages
put and other performance metrics of a wireless system. of monostatic RFID structure, there is an emerging concept
Inherently, it suffers from SI issues. In order to use full-duplex of bistatic architecture in which the carrier signal emitter is
incorporation with SWIPT, a robust SI cancellation algorithm dislocated from the reader where backscattered signals are
enabled low-power transceiver design is needed. This creates received. This architecture is characterized by easier setup with
a new direction for researchers interested in this avenue. A multiple carrier emitters and one centralized reader, and also
test bed design by incineration of non-linearity in the circuit provides long-range scatter radio communication for sensor
design and suitable mathematical modelling will optimize the networks [106].
EH capacity of the rectifier. Interference exploitation in full- Future Directions: One of the main challenges in realiz-
duplex needs to be studied further. Even though SI can be ing SWIPT systems is to find a reliable source for RF EH
exploited as an extra source of energy, its capacity to improve purposes. In this regard, the carrier emitters of bistatic radar
284 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

systems, which are essential parts of bistatic scatter radio, can F. WPT/SWIPT on Cooperative Relaying
act as potential sources for RF EH. By exploiting the scat- In wireless networks, the nodes may not be very close to
ter radio emitter’s transmissions, much more ambient energy each other or may hinder each other due to several urban,
can be captured. Moreover, future research work should be climatic or geographical reasons. This results in the line of
focused towards the development of a frequency modulation sight communication being infeasible, therefore idle nodes
(FM) band rectifier. All FM band frequencies and scatter radio between them can be used to create intermediate hops as
emitters frequency should be exploited to increase the acqui- proposed in [111] and [112]. However, it will be still use-
sition and harvesting of more unused energy. Furthermore, ful to use an intermediate idle node even when a direct link in
future research efforts should aim to combine the existing between the source and the destination is present to improve
developed communication systems with the scatter radio sen- the system level performance. Hence, the relaying node can
sors to design a battery-less scatter radio sensors having overhear the source’s transmission and re-transmit its modified
the FM band and emitters as a power source. Finally, an version of it to the destination to assist the overall decod-
ultra large scale sensor networks should be exploited with ing process [113], [114]. In the case where the relay operates
ultra low-cost sensor nodes having capacitors as the power using a battery assisted mode, it will not be very beneficial
sources. for the relay to participate in this cooperative transmission as
it has to spend its own energy. SWIPT or WPT can stand as
a potential candidate for this issue, where the source or an
E. SWIPT With Symbol Level Precoding external access point can transfer energy to the relay [167].
In contrast to the traditional concept that interference is Then, the relay has an extra energy budget to help with coop-
always considered to be harmful, there is an emerging con- erative transmission, without consuming its own energy. There
cept of treating constructive interference among the users as a are two main relaying protocols in practice, 1) decode and for-
source of useful signal energy [107]–[109]. With the knowl- ward or regenerative relaying and 2) amplify and forward or
edge of both the instantaneous CSI and the data symbols at transparent relaying [113], [114].
the Base Station (BS), the received interference can be clas- Over the past decade, interest in cooperative wireless diver-
sified as either constructive or destructive. The destructive sity techniques has rekindled within the research community.
interference deteriorates the detection performance whereas Usage of WPT/SWIPT has also become an interesting avenue
the constructive one moves the received symbols away from for self-powered relaying. A generic example of cooperative
the decision thresholds of the constellation, thus improving the relay schemes is given in Fig. 9. In the first timeslot, an access
signal detection performance [108]. point (or base station), transmits wireless RF energy and infor-
The available precoding for wireless downlink transmission mation simultaneously. The relay nodes (Relay 1 and Relay 2)
can be broadly grouped into: (i) user-level precoding, (ii) group are equipped with two antennas, i.e., one antenna for EH and
level precoding, and (iii) symbol level precoding [109]. The other for ID. The relay harvests energy from the access point
first technique is dependent on the CSI of individual users and uses it for ID and re-transmission signalling to the desti-
and the second technique is dependent on the CSI of each nation using physical layer network coding in the second time
user group. On the other hand, the symbol level precoding slot. Fig. 9 shows the down-link (DL) and the up-link (UL)
technique, also called data-aided precoding, requires CSI of cooperative relay networks. Relays can also harvest energy
knowledge as well as the symbols of the users. Besides the from the RF signals of the UL transmission [168]. In this way,
possibility of using constructive interference as a source of the relay can be a self-powered device. Such a wireless system
a useful information signal, it can also be used as a source can be powered using the presence of WiFi or using ambient
of electrical wireless energy for EH purpose [108]. For this signals.
purpose, symbol level precoding can be employed in order to Cooperative relay algorithm design for resource alloca-
exploit the constructive interference for ID and EH purposes. tion in SWIPT systems includes the following aspects/system
In this context, Timotheou et al. [108] have recently extended design protocols:-(note that some of these approaches have
the conventional SWIPT beamforming techniques in the con- some overlaps though they can be categorized as below).
text of symbol level precoding by exploiting the interference 1) Harvest-then-Cooperate Scheme: In this scheme,
signal as (i) a source of electrical signal for EH, and (ii) a Chen et al. [115] and Rajaram et al. [116] propose a
source of useful information signals [110]. protocol related to first harvest RF energy in the relay
Future Directions: The existing literature on symbol level node and then to use harvested energy for cooperation.
precoding has mostly focused on a single cell scenario. The The source and relay harvest energy from an access
investigation of this approach in practical multi-cell scenarios point or power beacon in the downlink and work coop-
is challenging due to the need for a high level of collaboration eratively in the up-link. The resource allocation and
among base stations in order to acquire the knowledge of sym- performance measuring parameters such as throughput
bols of the users. In this context, future works should focus on and energy consumption are analysed for a three node
the feasibility of SWIPT systems with symbol level precoding scheme in [115] and for multiple access relay scheme
in practical multi-cell systems and multi-user MIMO channels. in [116], and in [117], it is extended to multi-antenna
In addition, research works also should aim towards the syn- scenario.
chronization requirements of the system, practical impairments 2) Harvest-use scheme: In this scheme, the harvested
and the reduction of communication resource overheads. energy cannot be stored in the battery, but should be
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 285

Fig. 9. An energy harvesting and information transmitting relay architecture.

utilised for the operational energy needs [118]. This the power utilization in all relay nodes in the communica-
scheme offers a trade-off between the EH time and tion network for improving the performance of cooperative
communication time from the relay node. The full- communication. Moreover, SWIPT techniques need to be inte-
duplex would appear to be an attractive and promising grated with aforementioned cooperative relay design protocols.
technology for harvest-use assisted cooperative relay Also, researchers need to focus on improving coding gain
communications. by implementing available scheme with powerful channel
3) Self Energy Recycling: In this technique, a part of the codes. Furthermore, the power consumption of available chan-
energy (loop energy) that is used for information trans- nel codes such as polar codes and LDPC codes need to be
mission by the relay is used for harvesting, then reused investigated further in cooperative communication models. The
in addition to the dedicated energy sent by the source. In mobility of relay nodes and its effect on EH also need to be
this scheme, the relay can function in full-duplex mode studied further.
with simultaneous EH and information transmission.
4) Interference Aided Harvesting: An interference aided
RF EH operation is introduced in [119] for cooper- G. SWIPT Enabled Cooperative Non-Orthogonal Multiple
ative wireless relaying systems. In particular, energy- Access (NOMA)
constrained relays harvest energy from the received NOMA is one of the major candidates in the design of radio
information-bearing signal and co-channel interference access techniques for 5G [122]. In NOMA concept, there exist
signals. Then it uses that harvested energy to cover the different solutions that can be classified into two approaches:
energy expenses required to transmit the decoded signal power-domain and code-domain [121].
to the destination. • Power-domain NOMA - In power-domain NOMA con-
Future Directions: Most of the existing literature have cept, the key factor is to maintain significant differences
focused on including RF-EH and WPT to power up relays in power levels of the signals. Therefore, it is possible to
in the network. New EH scheme together with informa- separate high level signals and cancel them out to retain
tion transmission needs to be implemented by optimizing a low level signal in the receiver. Hence, NOMA exploits
286 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

the path loss differences in the groups of users, though it performance of the system. Complexity of 5G and require-
requires extra processing power in the receiver. ments of SWIPT need to be considered in receiver design,
• Code-domain NOMA - Code-domain NOMA achieves without increasing the complexity of the receiver. In addi-
multiplexing in the code domain. Moreover, code domain tion, challenges in NOMA such as resource allocation, channel
shares the available resources such as time and frequency estimation, receiver design etc. need to be addressed before
and utilize user specific spreading sequences. There NOMA can begin handshaking with SWIPT technology.
are different sub-classes of code-domain NOMA, such
as low-density spreading CDMA, low-density spreading
OFDM and sparse code multiple access [121]. H. Secure WPT/SWIPT Transmission
1) Cooperative NOMA: Considering a downlink scheme As already indicated, in SWIPT, the RF signal is used to
with two user groups, the first group (near-user) enjoying harvest the energy at the receiver end. In order to expedite the
good channel conditions, and the second group (far-users) EH process, the transmitter is able to emit a highly boosted
experiencing poor channel conditions, applying cooperative signal. This may likewise lead to an expanded defenceless-
transmission to NOMA can possibly improve the reliability ness against eavesdropping because of a higher potential for
of remote users. According to the proposed method in [123], data spillage when the receiver is malevolent. In line with
neighbouring-users are used as relays in the network to this, a new paradigm of QoS matter is raised in communica-
improve remote-user channel conditions. The advantage of tions systems with SWIPT frameworks, which is imperative
this approach is that the consecutive interference cancella- to incorporate.
tion is used by the neighbouring-users and, consequently, In particular, privacy and authentication have progressively
the neighbouring-users know the information of the remote become major areas for wireless communications. Also, phys-
users. In [124], the use of SWIPT assisted NOMA is con- ical layer security has emerged as a new layer of defence
sidered, where SWIPT was applied to the neighbouring-users, to ensure perfect communications secrecy besides the cryp-
to improve the reliability of the remote users, without drain- tography. Cooperative relay secure transmission with PHY
ing the battery life of neighbouring-users. Thus, cooperative layer security improvements have received significant atten-
NOMA and SWIPT communication concepts are integrated tion [129]–[130]. In [128], the secrecy rate is maximized with
together and a new protocol called cooperative SWIPT enabled the help of SWIPT-enabled AF relays, limited by the EH
NOMA protocol is proposed [124]. It was reported that power constraints of individual relays by jointly optimizing
ln(SNR)/SNR2 is the decay rate of the outage probability of far the cooperative beamforming relays.
users while 1/SNR2 is for conventional cooperative networks By employing a separated receiver mode, i.e., one partic-
(here, SNR indicates the signal-to-noise ratio). As a result ular receiver assigned for confidential ID and the rest of the
of this integration (SWIPT and NOMA), careful selection of receivers assigned for EH, some works [169], [171] focus on
network parameters, such as transmission rate or PS coefficient SWIPT assisted PHY layer security. In [172], artificial noise is
can lead to an acceptable system performance using harvested used to interfere with the eavesdropper for secrecy information
energy to power the relay transmission, even without the users nodes as well as to operate as the primary source of EH for
exploiting their own device batteries. non-ID receivers. This method is not suitable for co-located
2) MISO-NOMA: Applications of MISO-NOMA systems antennas where the user simultaneously receives secrecy ID
have been investigated in [125] and [126]. The work [125] and performs EH. In [131] [132], the aggregate harvested
studied the effect of the quasi-degradation concept on power of all users was optimized while satisfying secrecy rate
MISO-NOMA downlink transmission. A QoS optimization requirements of individual users over an OFDMA. In [133],
problem, considering a two user MISO-NOMA system given the secrecy performance analysis using secrecy outage and
a pair of target interference levels was investigated in [126]. secrecy capacity metrics has been studied for single-input
Do et al. [127] investigated the performance of transmit multiple-output SWIPT systems. In [134], SWIPT assisted
antenna selection schemes in a two-user MISO-NOMA coop- secrecy issues are addressed in non-perfect CSI for the same
erative transmission with SWIPT enabled at near users, used system set-up.
hybrid TS and PS SWIPT architectures to power near user Future Directions: Both power transfer efficiency and infor-
decode and forward relaying operation. mation security are equally important in SWIPT systems.
Future Directions: Most of the research works focused on These two separate objectives in SWIPT could bring new
power-domain NOMA with SWIPT. Therefore, more research challenges for the designing phase of physical layer secu-
activities are needed regarding SWIPT with code-Domain rity in SWIPT enabled communication networks. To be more
NOMA. Further research need to be done in SWIPT enabled specific, both power and information signals may compete
NOMA with other communication technologies such as with each other for limited resources in networks and devices.
mmWave, MIMO and etc. In SWIPT enabled NOMA schemes, Therefore, security issues and possible solutions can vary in
theoretical analysis is required to provide some introduction to different types of SWIPT enabled communication networks,
design the system parameters according to the specific appli- with specific characteristics within the networks. Security
cation requirements. Moreover, overall performance of the issues in each and every emerging SWIPT technologies need
NOMA system is critically based on interference cancellation to be further investigated with possible solutions. A proper
capability and complexity at the receiver. Nonetheless, these security framework for SWIPT has not yet been developed.
factors need to be re-analysed with SWIPT to improve overall Since SWIPT has recently been introduced, a proper security
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 287

Fig. 10. SWIPT enabled wireless sensor network system. Here we depict the dual energy harvesting policies based on the solar and RF power transfer
for WSN.

framework or measures can uplift the possibility of integrating As the availability of large spectrum resources at
SWIPT with existing communication technologies. higher frequencies (above 6GHz) is vital, Millimeter Wave
(mmWave) communication presents as a key candidate for
I. SWIPT Assisted mmWave Communications future 5G communications. We depict smart city application
Low-frequency bearing signals offer less power transfer effi- of mmWave enabled SWIPT based architecture in Fig. 12. In
ciency and ultimately affect the QoE of the user. As in (1), it general, mmWave has been identified as a promising avenue
is suggested that using lower frequencies provides an advan- for WPT due to the following reasons: very high frequencies,
tage. Nevertheless, this assumes a fixed antenna gain and thus, narrow beam, large array gains, and a dense network of
physically large antennas are needed for lower frequencies. mmWave base stations. As with the new IoT trend, many
However, the physical area of the transmitter and/or receiver low powered connected devices are expected to be deployed,
is limited in hand-held devices and sensors in IoT applica- and the mmWave RF signal can be used to harvest energy
tions. Therefore, assuming that the antenna size is larger than substantially.
the wavelength, the aperture would be proportional to actual Pathloss: The mmWave spectrum range is usually referred
physical cross-sectional area [173]. For example, the transmit- between 30 GHz and 300 GHz since the wavelength for these
ter or receiver could be resonant antennas or antenna arrays. frequencies is in the order of less than 10mm [138]. The uti-
With the help of these assumptions, the received power can lization of a higher spectrum above 6 GHz has been reduced
be given by [173] due to the fact that only longer wavelengths can diffract around
Pt At obstacles and terrestrial terrain more smoothly [139]. Also,
Pr = 2 2 Ar , (7) higher frequencies including mmWaves are usually impaired
R λ
where Pt and Pr represent transmitter power and received with strong reflections, refractions and scattering [139]. RF
power, Ar and At represent area of the receiving and trans- propagation in the earth’s atmosphere suffers from minor
mitting antennas, respectively. The wavelength denoted by λ energy loss due to the atmospheric absorption [140]. In gen-
and R is the distance between transmitter and the receiver. This eral, mmWave spectrum signals undergo higher atmospheric
suggests that using higher frequencies with the size limitation absorption than the signals using lower frequencies [139].
in transmitter or receiver is beneficial. Nevertheless, for short distance communication, atmospheric
288 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 11. A clustered SWIPT enabled wireless sensor network consisting multiple clusters of sensors and a sink node which is responsible for collecting data
from each sensor. The head node works as both an information and power transferor while relay node receiver harvest the RF energy from the head node of
the cluster and recharge the batteries.

and rain absorption are not limiting factors in mmWave In [142], a prototype is developed for 24 GHz rectifying
spectrum since around 250 meters, these absorptions are as antenna. This demonstrates the feasibility of wireless EH and
less as 4dB in 61GHZ links and 0.5dB in 24GHz [139]. information transmission by the use of the Millimeter-wave.
An EH scheme demonstrating harvest and use strategy is Moreover, an additional advantage of the mmWave spectrum
applied in [141] where the user harvests energy from the corre- is that high antenna directivity can be achieved using small
sponding base station through mmWave signals, then uses the form factor printed antennas [143]. This will allow the miti-
harvested energy to transmit information messages. The over- gation of interference, which can be used in SWIPT as an extra
all (energy-and-information) coverage probability is improved source of energy. In addition, transmitter will focus the entire
by use of an optimizing technique over the PS ratio in order to power towards the intended receiver, which also can improve
receive a signal (optimally portion) between the EH and the EH efficiency in SWIPT assisted mmWave communication.
ID modules using mmWave assisted SWIPT in [78]. Here, Future Directions: Unfortunately, as high-frequency bands
the authors maximize the network-wide energy coverage for a inherently suffer from poor penetration and diffraction, they
given user population by optimizing the antenna geometry. present as research challenges in using mmWave for WPCN
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 289

Fig. 12. An mmWave enabled wireless power transfer scheme for smart city.

and SWIPT enabled wireless networks. Also, a limited sink node responsible for collecting data from sensors in clus-
amount of work has considered mmWave for SWIPT/WPCN. ters. In this scenario to enable SWIPT, the head node of each
There is a huge gap in the research on mmWave assisted cluster works as both an information and power transferrer.
SWIPT in information processing and hardware develop- Relay node receiver harvests the RF energy from the head
ment. Many of current setups cannot be used directly for node of the cluster and recharges the batteries. Furthermore,
higher frequencies, i.e., above 6GHz due to the path loss Guo et al. [174] proposed a distributed iteration algorithm with
and health concerns. Further research activities needed in closed-form transmission power, PS ratio and relay selection
modelling path loss problem in SWIPT assisted mmWave by utilizing decomposition. Tong et al. [146] focused on the
communication. development of network deployment and its routing strategy.
The idea is to reduce the total recharging cost to enhance
the lifespan of WSNs. An optimization problem is formulated
J. SWIPT/WPT Enabled Wireless Sensor Networks using joint network deployment and routing in [24] by having
The WSN has come to significant attention with the new an assumption that the sensors can be continuously recharged
embarkation of the IoT. Some of the devices are small in before their power source is exhausted and have complete
nature but, generally, placed in hazardous or remote areas knowledge of CSI. A few prototype implementations of sen-
where human access is limited. Therefore, replacing the batter- sor nodes using RF EH are investigated in [24] and [147]. By
ies or supplying a stable power source is an issue for WSNs. reaping the benefits offered through SWIPT, low power WSNs
SWIPT enabled WSN is shown in Fig. 10. Peer et al. [144] deployed in the health care sector can recharge power sources
have proposed a hybrid TS and PS spectrum sharing proto- while they are in operation and this certainly improves the
col for EH wireless sensor nodes. In WSNs, power usage QoE [28].
is usually divided into three parts as in [69]: sensing, data Some works have suggested the use of a wireless
processing, and communication. As compared to sensing and energy charger application for WSNs, as it supports mobil-
data processing, communication is a costly functionality in ity [148], [149]. A practical real time wireless recharging
a typical sensor node [145]. Hence, local data processing is protocol for dynamic wireless recharging in sensor networks
important in order to minimize power utilization of WSNs. is proposed in [150]. The real-time recharging framework
As the failure of any node can significantly cause the struc- supports single or multiple mobile vehicles. Similarly, a
ture to be re-routed, causing inefficiency in terms of reliability collaborative wireless charging scheme has been presented
improving the lifespan of WSNs are important stems for in [151]. Recharging selected nodes in remote area with
future IoT. Unmanned Aerial Vehicles (UAVs) delivers a solution that
Clustered WSNs with SWIPT: Applying SWIPT to re-charge presently can recharge a single node. Johnson et al. [152] sim-
low energy relay nodes in cooperative clustered WSNs has ulated a limited recharging system and presented the evidence
been considered in [174]. As shown in Fig. 11, they con- of the effectiveness of recharging a single node. However, this
sider a WSN consisting of multiple clusters of sensors and a does not describe a robust system model and work is limited
290 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

to simulations. Prototype based implementation for such EH

architecture is provided in [153].
Future Directions: A substantial amount of work con-
centrates on the aspect of wireless EH and relatively less
work concentrates on SWIPT enabled WSNs. Interference
exploitation based EH techniques have not yet been largely
investigated. Physical layer redesign of SWIPT enabled WSNs
is a timely requirement to advance research within a solid the-
oretical framework. More research activities need to be carried
out in selecting a proper energy storage mechanism since it is
directly related to efficiency and long operational life time of Fig. 13. A basic MIMO broadcast system. Here a set of antennas are allocated
sensors. Also, robust communication protocols are vital to effi- for the information decoding and another set is assigned for the RF-EH.
ciently exploit energy sources and must be investigated further
to offer an ultimate self-sustainable solution.
In the seminal work of [58], SWIPT in massive MIMO enabled
multi-way relay networks (MWRNs) and energy constraining
K. SWIPT Enabled MIMO Systems amplify-and-forward relays have been investigated by utilizing
MIMO is an antenna technology for wireless communica- TS and PS techniques. Once the quantity of relay antennas
tion which has been comprehensively researched during the grows in the network, sum rate expressions and harvested
last two decades. It is used in numerous wireless networks, energy have been derived from a considerable amount. The
since it can notably improve the reliability and capacity of effect of co-channel interference (CCI) on the performance
wireless networks. Initial research works focused on point metric was also investigated in [154], and the result proved
to point MIMO links which use two devices with multiple that the CCI can be utilized in the relay for improving EH.
antennas to communicate with each other. Now focus has Integration of SWIPT and MWRNs with massive MIMO can
shifted to multi-user MIMO systems, where the base sta- be a significantly improved trade-off between the performance
tion having multiple antennas concurrently serves a group of of energy constraint AF relaying and the harvested energy. The
single-antenna users [70]. use of SWIPT in Massive MIMO enabled wireless networks
1) MIMO SWIPT: In the MIMO network, all receivers/user has a potential to attain a substantial improvement in energy
terminals are battery limited devices, and batteries need to be efficiency. Furthermore, it also works well with many appli-
recharged to extend the network lifetime. EH is performed by cations associated with millimetre wave communications and
the individual receivers through dedicated power transmission heterogeneous networks. Shi et al. [176] have investigated
from the transmitters in the network. Most of the work that secure beamforming design for two user MIMO broadcast
tried to integrate SWIPT in MIMO wireless networks assumes system for both information and energy transmission. For the
that there are two defined groups of users to be served, one design, authors have proposed semidefinite relaxation based
for receiving information and another for receiving power to best suitable solutions, for both the single and full stream cases
recharge their power sources. To have a better understanding, a with channels satisfying positive semidefiniteness. Meanwhile,
three-device (one transmitter and two receivers) MIMO wire- with an arbitrary number of streams, the inexact block coor-
less broadcast system is considered as shown in Fig. 13. Using dinate descent (IBCD) algorithm has been proposed for the
the signal sent by a transmitter, the energy receiver harvests general case. According to [176], adaptation of this work can
energy while the other receiver performs ID. Two different sce- be the optimization of the IBDC algorithm to more scenarios
narios in the MIMO broadcast system have been investigated in secure beamforming with multiple information and energy
in [70], namely separate and co-located ID and EH receivers. receivers.
In the first scenario (separate receivers), the best transmis- Future Directions: Due to the path loss, signal strength
sion strategy was designed to attain non-identical trade-offs is reduced by a significant amount. Therefore, the energy
in SWIPT. Those trade-offs are characterized by the boundary efficiency of MIMO SWIPT systems might not be as high
of the rate-energy region. For the second scenario (co-located as expected for long distance power transfer. To overcome
receivers), the previous solution can also be applied to the this bottleneck, more research works are needed with integra-
unique MIMO channel from the transmitter to the EH and tion of advanced green EH technologies with MIMO resource
ID receivers. Furthermore, two SWIPT techniques, TS and allocation. In addition, more research activities need to be
power switching, were investigated for the case of co-located focused on communication security in MIMO SWIPT due to
receivers [70]. the vulnerability of information leakage.
2) Massive MIMO SWIPT: Massive MIMO is a scaled
version of MIMO technology containing thousands of anten-
nas and terminals. Massive MIMO provides the new extent L. SWIPT: A Two User MISO Interference Channel
of efficiency and throughput by the antennas attached to the A single antenna is used at both source and destination
base station, focusing the reception and transmission of signal in typical wireless communication networks, which causes
energy into a small region of space [175]. Moreover, partial issues in the communication link with multipath effects. Issues
focussing also can be improved with the increase of antennas. caused by multipath wave propagation can be reduced by
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 291

MISO technology, having two or more antennas with the trans- Moreover, SWIPT enabled D2D communication in large scale
mission of multiple signals at the source. In MISO, the finest cognitive networks has been studied in [158]. They proposed
optimal transmission strategy for the best scenario has been a new WPT model with battery free design at the energy
studied where the two receivers in the communication system constraint D2D transmitter. In addition, when processing the
simultaneously harvest energy and decode information [65] to WPT, the impact of small scale fading is also considered in
implement self-sustained wireless communication networks. It this proposed model. SWIPT assisted D2D communication
is also worthwhile to note that cross-link signals are useful in network is shown in Fig. 14.
improving EH of the receivers in spite of the fact that it lim- Future Directions: A limited amount of research has
its the achievable sum rate [65]. Thereafter, considering the been conducted on SWIPT enabled D2D communication.
current limitation of circuit technology, two practical schemes Integration of D2D communication with the WPCN would
are proposed based on TDMA, where the receiver performs be a major advance towards achieving SWIPT in D2D com-
ID or EH at each time slot. munication. Several issues such as power control, resource
1) The first scheme named A, splits each transmission allocation, relay node association, relay node selection and
interval into two time slots, in which one slot is used to management, mode switching, etc. will need to be addressed
perform EH by both receivers and afterwards to perform in order to improve the power transfer efficiency of SWIPT
ID in the next time slot. enabled D2D communication networks. We predict that this
2) The transmission time of the second scheme, named as will evolve into one of the key technologies in the 5G
B, divides into two time slots, as in scheme A, with the communication for the years to come.
difference that, in each time slot, one receiver performs
ID whereas the other receiver performs EH operation
simultaneously. N. SWIPT Enabled Broadband Wireless Systems
The achievable sum rate of these proposed TDMA schemes A A wireless broadband is a wireless network technology that
and B in ideal conditions was studied via simulations [65]. It addresses the last mile major problem in telecommunications.
was proved that the ideal scheme, which uses ideal receivers, It provides high-speed Internet and data services via a wire-
is sometimes not the best proposition with reference to sum less local area or a wide area network (WLAN or WWAN).
rate maximization. In an interference limited system, TDMA The local multipoint distribution system and the multichannel
of scheme A offers a better sum rate as compared to the ideal multipoint distribution service are fixed wireless broadband
scheme. When one of the receivers requires comparatively technologies for broadband microwave wireless transmission
higher energy than the other, TDMA of scheme B provides heading from a local antenna to the destinations. Integration
a higher sum rate than TDMA of Scheme A. Overall simu- of SWIPT with Microwave Power Transfer (MPT) within a
lation results and the conclusion are very different from the broadband wireless system is a promising technique for the
previously reported results [58] in the absence of the ideal need of convenient energy supply in wireless networks [64].
receiver, which always performs better. The broadband wireless system with limited interference con-
tains multi-antenna base stations for information transfer and
power transfer to free the wireless devices from the limitation
M. SWIPT Assisted Device-to-Device Communication of restricted battery capacities. Far-field MTP has been used
Device-to-Device (D2D) communication provides direct to support SWIPT with the understanding of the QoS. In this
communication between the mobile users without or par- approach, using orthogonal OFDM, the broadband channel is
tial network-infrastructure utilization. D2D communication split into multiple orthogonal sub-channels, where the base
improves network throughput, spectrum efficiency and energy- station in the communication network transmits/receives one
efficiency, while supporting various location-based and peer- data stream per sub-channel created. Depending on the trans-
to-peer applications and services. It plays an important role in mission policy, each stream is encoded at a fixed rate or with
improving the network capacity and resource utilization [155]. one of the variable rates, according to the reported received
An additional feature of this technology is that the coverage SNRs. It is noted that without beamforming, OFDM alone can
can be significantly improved by the use of a relay node. decouple only an information transfer link, not an MPT link.
The relay node enhances the D2D network performance, while Based on OFDMA, this work considers both single-user and
a mobile user equipment (UE) communicates with the other multi-user systems, where the mobile device is assigned either
user. Relay user equipment relay (UER) node works as an to all sub-channels or to a single sub-channel. Depending on
intermediate node [155], [156]. the direction of information transfer, downlink or the uplink,
Due to energy constraint in transmitters, integration of two practical courses for SWIPT are considered. In the first
SWIPT with D2D communication is identified as a progres- scenario, SWIPT with downlink information transfer uses an
sive research area. Availability of sufficient energy in devices OFDM signal which is transmitted by the base station in the
is important to ensure the power demands of D2D applications. broadband wireless network, for both information and power
This approach became ineffective due to the use of UERs own transfer. The second approach deals with uplink information
power for data transmissions of other UEs . As a solution transfer, where information and power transfer proceeds in
for this bottleneck in [157], a D2D communication enabled opposite directions. Downlink MPT depends on the transmis-
EH cellular network where the UER harvests energy from a sion of power tones categorized as unmodulated, while uplink
base station and uses it for D2D communication is considered. data signals are OFDM modulated.
292 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Fig. 14. SWIPT assisted D2D communication network where the users harvest energy from a base station and use harvested energy for D2D communication.

SWIPT enabled BWS contains dual antenna to support the based SWIPT systems. Moreover, different SWIPT techniques
architecture in Fig. 16, which can be reconfigured depending and different multiple access schemes used in multi-carrier
on the direction of information transfer [64]. This architecture SWIPT systems were explained. Then we provided possible
contains a transceiver based on downlink or uplink informa- future directions in this context including the use of wideband
tion transfer, which either demodulates and decodes or encodes receivers, investigated suitable techniques to utilize unused
and modulates received data, and a component of harvested carriers for EH and explored the use of different sources for
energy converts an input signal to DC power. Once the archi- RF-EH in multi-carrier SWIPT systems. Next, we explained
tecture is configured for downlink, the outputs of the antenna WPT/SWIPT enabled CRNs. Enabling SWIPT in CRNs can
are logically combined to improve the power of received sig- improve the spectral and energy efficiency of communication
nals. Thereafter, the combiner output is splitted into two by systems. A number of different CRN architectures with RF-
the power splitter and is sent to both the transceiver and the EH have been proposed in the literature. Moreover, cooperative
energy harvester. Following this, the architecture configured access schemes have been used to improve the efficiency of
for uplink by having attached two antennas separately to sup- SWIPT transmissions in CRNs. Future research directions of
port the full-duplex operation for both information and power SWIPT enabled CRNs should focus on the main four func-
transfer in the transceiver and the energy harvester. tions of CRNs, i.e., spectrum sensing, access, management and
hand-off to support efficient and dynamic spectrum access in
SWIPT enabled CRNs.
O. Summary and Insights Next, we reviewed the existing works of SWIPT enabled
In this section, we reviewed existing literature related to full-duplex communication. In the past, full-duplex communi-
the use of SWIPT in emerging communication technologies cation was not encouraged to use in communication systems
and provided future directions. The increasing attention to the due to the effect of SI. However, with the development of
energy efficiency of 5G has led research community to an different SI cancellation techniques and the possible usages
extensive range of research opportunities related to the inte- of SI in communication systems, research community moti-
gration of SWIPT with modern communication technologies. vates to use full-duplex in modern communication systems.
First, we reviewed multi-carrier SWIPT systems with the lat- Principally, in SWIPT enabled full-duplex, SI can be con-
est literature. Most of the existing works have studied OFDM verted into an extra source of energy. Moreover, both the
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 293

Fig. 15. SWIPT assisted solar powered satellite to initiate wireless power transmission and information transmission to ground and a mobile base station
(drones or unmanned aerial vehicle) via microwave beams.

TS and PS SWIPT receiver architectures have been used in comprehensive review on SWIPT enabled cooperative com-
recent research related to SWIPT with FD. More importantly, munication. SWIPT stands as a enabling solution for the
some of the works applied TS and PS randomly to the nodes requirement of extra energy in relay nodes to participate
in the communication systems to improve system efficiency. in relay operations. Moreover, we included the aspects of
However, further research activities are required before SWIPT cooperative relay algorithm design for resource allocation in
enabling full-duplex handshakes with modern communication SWIPT systems including, Harvest-then-Cooperate, Harvest-
systems. Furthermore, the use of SI related to interference use, self-energy recycling and interference aided harvesting.
exploitation need to be analysed and the transceiver design Most of the existing literature focused on using RF-EH
needs to be changed accordingly in SWIPT enabled FD and WPT to power up the relays. However, new SWIPT
communication systems. An effective combination of full- schemes need to be integrated with cooperative design pro-
duplex-CRNs with SWIPT can be consider as an important tocols. Investigating cooperative communication models and
future direction. Then we reviewed Bistatic Scatter Radio for the effect of channel codes can be a promising research
EH with proper explanation of literature available along with direction. Next subsection included, SWIPT enabled coop-
some future directions. However,the major research direction erative NOMA. More attention came to NOMA with iden-
is to find a reliable energy source for RF-EH in Bistatic Scatter tifying as one of the major candidates in 5G. Although
Radio. there are two different domains of NOMA, all existing lit-
In the next subsection, we reviewed SWIPT with symbol erature on SWIPT assisted NOMA considered power-domain
level precoding. In this domain, constructive interference can NOMA instead of the code-domain NOMA. Moreover, coop-
be used for ID and EH purposes in symbol level precoding. erative schemes have been used in SWIPT assisted NOMA
A few works have used SWIPT beamforming technique as to improve the reliability of remote users. Future research
a source for EH and a source of useful information sig- work in SWIPT assisted NOMA should further investigate
nals. Investigating the feasibility of SWIPT systems with SWIPT enabled NOMA using code-domain NOMA and with
symbol level precoding in multi-cell systems can be one other communication technologies such as mmWave and
of the promising research directions. Next, we provided a MIMO. In addition, channel estimations and receiver design
294 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

need to be addressed before integrating NOMA with SWIPT to meet some QoS criteria under SWIPT. Due to practical
technology. limitations of the EH circuit, it is not feasible to have both EH
Subsequently, we reviewed the existing works in the area and ID from the received signal. Therefore, optimized resource
of secure WPT/SWIPT transmission. In SWIPT systems, both allocation algorithms are needed to increase the performance
information security and power transfer efficiency are equally of SWIPT/EH system. Boshkovska et al. [177] designed a
important. Highly boosted signal emitted by the transmitter resource allocation algorithm for SWIPT for a realistic non-
can lead to an expanded defencelessness against eavesdrop- linear EH model and showed significant improvement in the
pers. In the literature, separated receiver modes have been system performance. Therefore, existing scheduling algorithms
used as a solution, where one receiver is allocated for con- need to be optimized or new scheduling algorithms need to
fidential ID while the other one is assigned for EH. However, be created to achieve better performance in SWIPT enabled
security issues in each and every emerging SWIPT enabled communication networks.
technologies need to be further investigated. SWIPT assisted
mmWave communication was reviewed in the next subsec- B. Hardware Impairments
tion. The mmWave communication has gained more attention
Most research works related to SWIPT have not yet focused
after being selected as a key candidate for 5G. Very high
on hardware impairments in the SWIPT enabled commu-
frequency, narrow beam, large array gain and a dense network
nication networks. In wireless communication, there occur
with mmWave base station make mmWave as a prominent can-
several hardware impairments such as in-phase/ quadrature-
didate for WPT. However, only a limited amount of works have
phase (I/Q) imbalances, high-power amplifier non-linearity
considered mmWave for SWIPT and there is a huge gap in
and oscillator phase noise due to the involved low quality
information processing and hardware development in SWIPT
hardware transceiver components, and they may significantly
assisted mmWave communication.
degrade the performance of various wireless systems. The
Next, we reviewed SWIPT enabled WSNs, SWIPT enabled
effect of realistic relay transceiver structures on the outage
WSNs TS and PS SWIPT architecture in EH wireless sensor
probability and throughput of WPT has been analysed and
nodes. As compared to sensing, data processing, communi-
confirmed in [178] based on two-way decode-and-forward
cation is a costly functionality in SWIPT enabled WSNs.
(DF) cognitive network in the presence of transceiver impair-
Thus, harvested power in nodes need to be used effectively
ment. Moreover, DF in EH systems with multiple antennas
to avoid catastrophic failure in the system. Furthermore, it is
are also studied in [179] under the presence of transmitter
very important to identify potential energy sources for the EH
hardware impairments. However, hardware impairments can
process in WSNs. Moreover, SWIPT enabled WSNs will lead
have different effects on different SWIPT enabled commu-
the concept of IoT into a different level in the research com-
nication systems. In this regard, it is an interesting research
munity. Next, we reviewed SWIPT enabled MIMO systems
direction to exploit the effect of hardware impairments in
that can notably improve the reliability and capacity of wire-
different SWIPT enabled wireless communication systems.
less networks. Initial research efforts focused on point to point
In addition, we strongly recommend that future research on
MIMO links and now it has shifted to multi-user MIMO sce-
SWIPT and SWIPT emerging technologies need to pay extra
narios. SWIPT integration can be more beneficial since all
attention to hardware impairments since this can be one of
receivers in MIMO are battery limited devices. SWIPT in
the direct reasons for the system performance degradation of
massive MIMO enabled multi-way relay networks used both
SWIPT enabled communication systems. Hopefully, the out-
TS and PS techniques and with the growth of the number
come of such research can be used to select appropriate quality
of antennas in the systems improve the sum-rate expression
hardware for future SWIPT enabled communication systems.
and the harvested energy in a considerable amount. However,
more research activities required in long-distance power trans-
fer to identify enabling solutions for path loss. Security aspects C. Investigation of CSI Feedback
of SWIPT enabled MIMO need to be further investigated Previous improvements in modulation, coding and schedul-
before applying into real-world communication. Finally, we ing have led to the wireless communication networks deployed
have reviewed existing literature in SWIPT enabled D2D in the present. Different channel adaptive techniques tend to
communication and BWSs with possible future directions. be used in next-generation wireless networks. Thus, the trans-
mitter requires some information regarding wireless channel
VII. R ECOMMENDATION FOR F UTURE D IRECTIONS conditions referred to as CSI at the transmitter. In SWIPT
In the previous sections, we have highlighted some of the enabled next-generation communication networks, CSI could
technical issues in SWIPT enabled communication network help to improve efficiency of both EH and ID operations. Since
systems. SWIPT presents a number of common technical chal- the value of the feedback varies with the communication sce-
lenges in different SWIPT enabled emerging communication nario, investigating various types of feedback to improve the
technologies. In the following, we suggest some of the future efficiency of SWIPT enabled communication systems could
research directions for common issues in SWIPT. be a promising research direction. The selections of the best
suitable feedback mechanism for SWIPT systems also can
A. Resource Scheduling in SWIPT be a difficult choice to make since different communication
As previously mentioned, scheduling is a critical factor in systems have different characteristics and requirements. For
improving the system performance of SWIPT networks and instance, some capable terminals at the transmitter could afford
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 295

Fig. 16. Dual mode SWIPT supported mobile architecture: This architecture contains a transceiver based on downlink or uplink information transfer, which
either demodulates and decodes or encodes and modulates received data, and a component of harvested energy converts an input signal to DC power.

CSI feedback, but not the simple IoT terminals. Therefore, of low interconnect complexity and inherent flexibility.
how feedback mechanism affects the selection and effective- Maunder [187] demonstrated that they facilitate high-
ness of the SWIPT strategy needs to be studied further in the throughput flexible channel coding gain at lower implemen-
next-generation communication systems. tation complexities as compared to LDPC and polar codes.
They further explained that the turbo codes offer an additional
D. Channel Coding Techniques benefit of backward compatibility to 3G and 4G standards,
which is favourable for the cellular communications indus-
A proper channel coding technique is needed for SWIPT try in terms of cost savings. This again invites the research
enabled communication systems since the receiving signal may community on channel coding to investigate the best suit-
contain corrupted information such as noise, interference, fad- able coding technique for energy-efficient deployment in 5G
ing etc., and the available energy budget is extremely limited standards.
for self-powered nodes. Polar codes and LDPC codes can be
potential channel coding techniques in SWIPT enabled 5G
E. Information Theoretic Framework
communications. A method for efficiently constructing polar
codes has been presented and analysed in [180], and it has Information theoretic framework assists to quantitatively
been applied in cooperative communications. Polar codes can analyse and model parameters in communication networks.
be decoded by using successive cancellation (SC) or the belief A proper theoretical framework has not yet been developed
propagation (BP) algorithms. However, unlike SC decoders, for SWIPT technology. Since SWIPT has recently been intro-
performance optimizations for BP decoders have not been duced, a theoretical framework for trustworthiness evaluation
very fully explored. An early stopping criteria for polar BP could raise the possibility of merging SWIPT with existing
decoding to reduce energy dissipation and decoding latency communication technologies. In this regard, implementing the-
has been proposed in [181] and [182]. In [183], it has been oretic frameworks for SWIPT systems can be an interesting
shown that polar codes are suitable for DF and compress- and valuable research area.
and-forward relaying in relay channels with the orthogonal
receivers. F. Internet of Things
A low-power decoder design approach for generic quasi- IoT is a novel concept which is based on connecting all
cyclic low-density parity-check (QC-LDPC) codes based on types of electronic devices to the Internet. IoT basically con-
the layered min-sum decoding algorithm has been given nects people, processes, data and every possible things together
in [184] and a low-power high-throughput LDPC decoder in order to fulfill people’s day-to-day needs. In the present,
using non-refresh embedded DRAM has been designed IoT is defined in the broader context comprising of differ-
in [185]. Turbo codes have been used in 3G cellular stan- ent scenarios using different communication strategies such
dard. These codes achieve lower bit error rates at the expense as WSNs, D2D and machine-to-machine communication. One
of high-computational complexity as compared to LDPC and of the main deployment challenges in IoT is to maintain
Polar codes. Therefore, for mobile communication devices, reliable communication with the low cost and power lim-
designing energy-efficient Turbo decoders is of great impor- ited IoT devices. The rapid growth of IoT applications has
tance. Kaza and Chakrabarti [186] presented a MAP-based increased the energy consumption of small devices dramati-
Turbo decoding algorithms with energy-quality trade-offs for cally. Therefore, it is important to address the energy efficiency
various channel models. of small devices. In general, most of the IoT devices are bat-
However, an energy-efficient Polar and LDPC cod- tery operated and sometimes can be located in remote areas
ing scheme for SWIPT enabled communication systems such as devices in a vessel in the middle of the deep ocean,
is required. The investigation of low power consumption military devices etc. [188]. Therefore, charging those batteries
and adaptive coding techniques suitable for SWIPT/WPT can sometimes be economically infeasible. Wireless relaying
dependent wireless communication system is an encourag- has been widely used in IoT wireless systems [189]. However,
ing research direction. The Turbo codes have the benefits relaying operations require extra power from nodes in the
296 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

relay, which sometimes can prevent battery powered devices MTC. Information and power transfer efficiency and informa-
from taking part in the relay. Therefore, RF powered relay- tion security are equally important in SWIPT enabled MTC.
ing is a favourable solution, in which nodes in the relay can However, due to the different characteristics of devices in
harvest energy either from source or externally from external the network, security issues and possible solutions can vary.
RF transmissions. SWIPT integration to IoT devices can be a Therefore, considerations should be given to provide a satis-
promising solution to improve efficiency of IoT. Nevertheless, factory level of security and effective resource management
the integration of SWIPT with IoT opens up new challenges. schemes in SWIPT enabled MTC.
Only the energy used for transmission merely is taken into
account when designing scheduling. However, energy used for
receiving cannot be ignored as compared to that of the trans- H. Satellite Communication
mission [190]. Thus, accurate energy modelling is required. Satellite assisted SWIPT is an interesting application via a
Moreover, more works need to be carried out regarding relay radio frequency beam. As many satellites are solar-powered,
selection in IoT, due to the fact that relay selection criteria for we could use solar EH to power up the satellite and then to
information transmission may be different from energy trans- use it to initiate wireless power transmission to ground stations
fer. The distance between nodes in IoT networks will notably via microwave beams. It could also participate in information
affect overall SWIPT system performance. Most of the exist- transmission [192]. This is designed in a geostationary orbit,
ing energy harvesters work efficiently over short distances. 36,000 km above the Earth’s surface as shown in Fig. 15
Therefore, decisions have to be taken regarding the use of This can be used for a SWIPT application, as the satellite
multi-hop with shorter distance in IoT networks. can transmit information and power for low power consumed
base stations and mobile base stations such as drones etc.
This can also be used to energize UAVs while supporting
G. Machine Type Communications (MTC) the idea of ubiquitous energy. Future research should explore
MTC is a promising technology that enables communica- the development and implementation of such satellite assisted
tion between machine type devices such as devices/sensors, SWIPT based drones, base stations or similar applications and
machines and vehicles, to network entities such as base sta- contribute to the idea of ubiquitous energy becoming a reality.
tions, eNode base stations and access points with minimal
human intervention. A wide range of devices and appli-
cations can be involved in MTC. MTC applications may VIII. C ONCLUSION
include health monitoring devices, transportation terminals, In this paper, we have presented an inclusive survey of
smart parking meters, etc., which must be effectively con- RF-EH, WPT and SWIPT together with associated emerg-
nected via communication links [191]. Nevertheless, some of ing technologies with new opportunities and future directions.
these devices may have limited power capacity within the First, we pointed out the need for sustainable communication
devices. Sometimes failure of a device in the linked network in 5G communications systems. We then reviewed existing
can have huge impact on day-to-day human activities. Thus, articles related to SWIPT/WPT with the objective of high-
while improving communication aspects of the MTC, extra lighting the novelty of our survey paper. Subsequently, we
attention needs to be allocated for power requirement of the provided a comprehensive description of RF-EH and the
devices. SWIPT integration to MTC can be a revolutionary circuit design of EH module together with various related
concept which helps to stabilise the power requirements of technologies. We then provided a detailed description on the
some devices in the network. Energy can be harvested using SWIPT and its antenna architectures. Next, we presented one
the same communication link while devices are in idle mode. of the uprising concepts, interference exploitation in SWIPT.
Nevertheless, integration of SWIPT also can create some This included the identification of new aspects of interference
additional challenges. Some of the challenges are: and the potential use of interference in wireless systems. This
1) privacy concerns of the devices and the users paper, then presented a comprehensive description of SWIPT
2) heterogeneity and mobility management of the devices together with emerging wireless communication technologies
3) QoS requirement such as multi-carrier SWIPT systems, WPT/SWIPT enabled
MTC traffic is different from other communication networks. It CR, full-duplex communications, bistatic scatter radio, sym-
has unique traffic patterns due to its special functionalities and bol level precoding, cooperative relaying, NOMA, mmWave
requirements. These details need to be taken into consideration communications, WSNs, MIMO, D2D and BWSs. Also, we
when designing ID and EH scheduling for the network. This discussed the security aspect of SWIPT transmission as it is
can present another interesting research question in SWIPT of great importance. Moreover, we provided a comprehen-
enabled M2M communication. One of the main characteristics sive survey on up to date research works and open issues
of the MTC, the capability of integrating many different types of each specific emerging technology. Finally, we pointed
of devices, can also present a challenge in the integration of out major common technical challenges in different SWIPT
SWIPT. Available energy, computational power, power storage enabled emerging communication technologies. The biggest
capacity and power requirement of the devices may be very challenge in the next-generation wireless communication is
different and these heterogeneity properties make the design of to integrate the spectrum of several enabling technologies to
EH protocols a very challenging task. Security will be another provide an energy aware resource efficient, cost-effective and
important issue for the successful integration of SWIPT with reliable communication.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 297

TABLE VI
D EFINITIONS OF ACRONYMS AND N OTATIONS

A PPENDIX [6] L. Hou and S. Tan, “A preliminary study of thermal energy harvesting
for industrial wireless sensor networks,” in Proc. 10th Int. Conf. Sens.
TABLE VI provides the definitions of acronyms and nota- Technol. (ICST), 2016, pp. 1–5.
tions used throughout this paper. [7] I. Krikidis et al., “Simultaneous wireless information and power trans-
fer in modern communication systems,” IEEE Commun. Mag., vol. 52,
no. 11, pp. 104–110, Nov. 2014.
R EFERENCES [8] L. R. Varshney, “Transporting information and energy simultaneously,”
in Proc. IEEE Int. Symp. Inf. Theory (ISIT), 2008, pp. 1612–1616.
[1] J. G. Andrews et al., “What will 5G be?” IEEE J. Sel. Areas Commun., [9] I.-J. Yoon, “Wireless power transfer in the radiating near-field region,”
vol. 32, no. 6, pp. 1065–1082, Jun. 2014. in Proc. USNC URSI Radio Sci. Meeting (Joint AP S Symp.), 2015,
[2] A. Osseiran et al., “Scenarios for 5G mobile and wireless communica- p. 344.
tions: The vision of the METIS project,” IEEE Commun. Mag., vol. 52, [10] K. Huang and X. Zhou, “Cutting the last wires for mobile communi-
no. 5, pp. 26–35, May 2014. cations by microwave power transfer,” IEEE Commun. Mag., vol. 53,
[3] M. Agiwal, A. Roy, and N. Saxena, “Next generation 5G wireless no. 6, pp. 86–93, Jun. 2015.
networks: A comprehensive survey,” IEEE Commun. Surveys Tuts., [11] X. Lu, P. Wang, D. Niyato, D. I. Kim, and Z. Han, “Wireless networks
vol. 18, no. 3, pp. 1617–1655, 3rd Quart., 2016. with RF energy harvesting: A contemporary survey,” IEEE Commun.
[4] Q. Wu, G. Y. Li, W. Chen, D. W. K. Ng, and R. Schober, “An overview Surveys Tuts., vol. 17, no. 2, pp. 757–789, 2nd Quart., 2015.
of sustainable green 5G networks,” IEEE Wireless Commun., vol. 24, [12] H. Yu et al., “A cooperative modulation recognition: New paradigm
no. 4, pp. 72–80, Aug. 2017. for power line networks in smart grid,” Phys. Commun., vol. 25,
[5] D. Niyato, E. Hossain, M. M. Rashid, and V. K. Bhargava, “Wireless pp. 268–276, Dec. 2017.
sensor networks with energy harvesting technologies: A game-theoretic [13] Z. Ding et al., “Application of smart antenna technologies in simulta-
approach to optimal energy management,” IEEE Wireless Commun., neous wireless information and power transfer,” IEEE Commun. Mag.,
vol. 14, no. 4, pp. 90–96, Aug. 2007. vol. 53, no. 4, pp. 86–93, Apr. 2015.
298 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

[14] S. Ulukus et al., “Energy harvesting wireless communications: A [37] B. Lee, M. Kiani, and M. Ghovanloo, “A triple-loop inductive power
review of recent advances,” IEEE J. Sel. Areas Commun., vol. 33, no. 3, transmission system for biomedical applications,” IEEE Trans. Biomed.
pp. 360–381, Mar. 2015. Circuits Syst., vol. 10, no. 1, pp. 138–148, Feb. 2016.
[15] X. Chen, D. W. K. Ng, and H.-H. Chen, “Secrecy wireless informa- [38] A. Kurs et al., “Wireless power transfer via strongly coupled magnetic
tion and power transfer: Challenges and opportunities,” IEEE Wireless resonances,” Science, vol. 317, no. 5834, pp. 83–86, 2007.
Commun., vol. 23, no. 2, pp. 54–61, Apr. 2016. [39] G. A. Covic, J. T. Boys, M. L. G. Kissin, and H. G. Lu,
[16] M. Amjad, F. Akhtar, M. H. Rehmani, M. Reisslein, and T. Umer, “A three-phase inductive power transfer system for roadway-powered
“Full-duplex communication in cognitive radio networks: A sur- vehicles,” IEEE Trans. Ind. Electron., vol. 54, no. 6, pp. 3370–3378,
vey,” IEEE Commun. Surveys Tuts., vol. 19, no. 4, pp. 2158–2191, Dec. 2007.
4th Quart., 2017. [40] E. Bou-Balust, A. P. Hu, and E. Alarcon, “Scalability analysis of SIMO
[17] X. Lu, P. Wang, D. Niyato, D. I. Kim, and Z. Han, “Wireless non-radiative resonant wireless power transfer systems based on circuit
charging technologies: Fundamentals, standards, and network appli- models,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 62, no. 10,
cations,” IEEE Commun. Surveys Tuts., vol. 18, no. 2, pp. 1413–1452, pp. 2574–2583, Oct. 2015.
2nd Quart., 2016. [41] M. Chabalko, J. Besnoff, M. Laifenfeld, and D. S. Ricketts,
[18] N. Zhao et al., “Exploiting interference for energy harvesting: “Resonantly coupled wireless power transfer for non-stationary loads
A survey, research issues, and challenges,” IEEE Access, vol. 5, with application in automotive environments,” IEEE Trans. Ind.
pp. 10403–10421, 2017. Electron., vol. 64, no. 1, pp. 91–103, Jan. 2017.
[19] M.-L. Ku, W. Li, Y. Chen, and K. R. Liu, “Advances in energy har- [42] J. Zhang and C. Cheng, “Comparative studies between KVL and BPFT
vesting communications: Past, present, and future challenges,” IEEE in magnetically-coupled resonant wireless power transfer,” IET Power
Commun. Surveys Tuts., vol. 18, no. 2, pp. 1384–1412, 2nd Quart., Electron., vol. 9, no. 10, pp. 2121–2129, Aug. 2016.
2016. [43] A. P. Sample, D. T. Meyer, and J. R. Smith, “Analysis, experimental
[20] O. Ozel, K. Tutuncuoglu, S. Ulukus, and A. Yener, “Fundamental limits results, and range adaptation of magnetically coupled resonators for
of energy harvesting communications,” IEEE Commun. Mag., vol. 53, wireless power transfer,” IEEE Trans. Ind. Electron., vol. 58, no. 2,
no. 4, pp. 126–132, Apr. 2015. pp. 544–554, Feb. 2011.
[21] H. J. Visser and R. J. M. Vullers, “RF energy harvesting and transport [44] J. Dai and D. C. Ludois, “A survey of wireless power transfer and
for wireless sensor network applications: Principles and requirements,” a critical comparison of inductive and capacitive coupling for small
Proc. IEEE, vol. 101, no. 6, pp. 1410–1423, Jun. 2013. gap applications,” IEEE Trans. Power Electron., vol. 30, no. 11,
[22] F. Akhtar and M. H. Rehmani, “Energy harvesting for self-sustainable pp. 6017–6029, Nov. 2015.
wireless body area networks,” IT Prof., vol. 19, no. 2, pp. 32–40, [45] J. Kim, D.-H. Kim, and Y.-J. Park, “Analysis of capacitive impedance
Mar./Apr. 2017. matching networks for simultaneous wireless power transfer to multiple
[23] F. Akhtar and M. H. Rehmani, “Energy replenishment using renew- devices,” IEEE Trans. Ind. Electron., vol. 62, no. 5, pp. 2807–2813,
able and traditional energy resources for sustainable wireless sen- May 2015.
sor networks: A review,” Renew. Sustain. Energy Rev., vol. 45, [46] F. Lu, H. Zhang, H. Hofmann, and C. C. Mi, “An inductive
pp. 769–784, May 2015. and capacitive combined wireless power transfer system with LC-
[24] H. Nishimoto, Y. Kawahara, and T. Asami, “Prototype implementation compensated topology,” IEEE Trans. Power Electron., vol. 31, no. 12,
of ambient RF energy harvesting wireless sensor networks,” in Proc. pp. 8471–8482, Dec. 2016.
IEEE Sensors, 2010, pp. 1282–1287.
[47] J. Dai and D. C. Ludois, “Capacitive power transfer through a confor-
[25] N. Alliance, “CSIT white paper,” Frankfurt, Germany, Next Gener. mal bumper for electric vehicle charging,” IEEE J. Emerg. Sel. Topics
Mobile Netw., White Paper, 2015. Power Electron., vol. 4, no. 3, pp. 1015–1025, Sep. 2016.
[26] X. Lu, P. Wang, D. Niyato, and E. Hossain, “Dynamic spectrum access
[48] A. Massa, G. Oliveri, F. Viani, and P. Rocca, “Array designs
in cognitive radio networks with RF energy harvesting,” IEEE Wireless
for long-distance wireless power transmission: State-of-the-art and
Commun., vol. 21, no. 3, pp. 102–110, Jun. 2014.
innovative solutions,” Proc. IEEE, vol. 101, no. 6, pp. 1464–1481,
[27] J. A. G. Akkermans, M. C. Van Beurden, G. J. N. Doodeman, and
Jun. 2013.
H. J. Visser, “Analytical models for low-power rectenna design,” IEEE
[49] G. A. Landis, “Applications for space power by laser transmission,” in
Antennas Wireless Propag. Lett., vol. 4, no. 1, pp. 187–190, Jun. 2005.
Proc. Spite Int. Soc. Opt. Eng., 1994, pp. 252–255.
[28] X. Zhang et al., “An energy-efficient ASIC for wireless body sensor
networks in medical applications,” IEEE Trans. Biomed. Circuits Syst., [50] G. A. Landis, “RE-evaluating satellite solar power systems for earth,”
vol. 4, no. 1, pp. 11–18, Feb. 2010. in Proc. IEEE 4th World Conf. Photovolt. Energy Convers. Conf. Rec.,
vol. 2, 2006, pp. 1939–1942.
[29] R. C. Johnson, H. A. Ecker, and J. S. Hollis, “Determination of far-field
antenna patterns from near-field measurements,” Proc. IEEE, vol. 61, [51] M. Xia and S. Aissa, “On the efficiency of far-field wireless power
no. 12, pp. 1668–1694, Dec. 1973. transfer,” IEEE Trans. Signal Process., vol. 63, no. 11, pp. 2835–2847,
[30] S. Shrestha, S.-K. Noh, and D.-Y. Choi, “Comparative study Jun. 2015.
of antenna designs for RF energy harvesting,” Int. J. Antennas [52] J. Park, Y. Tak, Y. Kim, Y. Kim, and S. Nam, “Investigation of adaptive
Propag., vol. 2013, Jan. 2013, Art. no. 385260. [Online]. Available: matching methods for near-field wireless power transfer,” IEEE Trans.
https://www.hindawi.com/journals/ijap/2013/385260/ Antennas Propag., vol. 59, no. 5, pp. 1769–1773, May 2011.
[31] R. W. Habash et al., “Recent advances in research on radiofrequency [53] M. Bhutada, V. Singh, and C. Warty, “Transmission of wireless power
fields and health: 2004-2007,” J. Toxicol. Environ. Health B, vol. 12, in two-coil and four-coil systems using coupled mode theory,” in Proc.
no. 4, pp. 250–288, 2009. Aerosp. Conf., 2015, pp. 1–8.
[32] Safe Code 6: Limits of Human Exposure to Radio Frequency [54] J. Lee and S. Nam, “Fundamental aspects of near-field coupling small
Electromagnetic Energy in the Frequency Range From 3 kHz to antennas for wireless power transfer,” IEEE Trans. Antennas Propag.,
300 GHz, document 6, Healthy Environ. Consum. Safety Branch Health vol. 58, no. 11, pp. 3442–3449, Nov. 2010.
Canada, Ottawa, ON, Canada, 2009. [55] S. Y. R. Hui, W. Zhong, and C. K. Lee, “A critical review of recent
[33] R. E. Hamam, A. Karalis, J. D. Joannopoulos, and M. Soljačić, progress in mid-range wireless power transfer,” IEEE Trans. Power
“Efficient weakly-radiative wireless energy transfer: An EIT-like Electron., vol. 29, no. 9, pp. 4500–4511, Sep. 2014.
approach,” Ann. Phys., vol. 324, no. 8, pp. 1783–1795, 2009. [56] W. C. Brown, “The history of power transmission by radio waves,”
[34] S. Ho, J. Wang, W. Fu, and M. Sun, “A comparative study between IEEE Trans. Microw. Theory Techn., vol. 32, no. 9, pp. 1230–1242,
novel witricity and traditional inductive magnetic coupling in wire- Sep. 1984.
less charging,” IEEE Trans. Magn., vol. 47, no. 5, pp. 1522–1525, [57] J. O. McSpadden and J. C. Mankins, “Space solar power programs
May 2011. and microwave wireless power transmission technology,” IEEE Microw.
[35] J. G. Hayes, M. G. Egan, J. M. D. Murphy, S. E. Schulz, and J. T. Hall, Mag., vol. 3, no. 4, pp. 46–57, Dec. 2002.
“Wide-load-range resonant converter supplying the SAE J-1773 electric [58] R. Zhang and C. K. Ho, “MIMO broadcasting for simultaneous wire-
vehicle inductive charging interface,” IEEE Trans. Ind. Appl., vol. 35, less information and power transfer,” IEEE Trans. Wireless Commun.,
no. 4, pp. 884–895, Jul./Aug. 1999. vol. 12, no. 5, pp. 1989–2001, May 2013.
[36] T. Nagashima et al., “Analysis and design of loosely inductive coupled [59] D. W. K. Ng, E. S. Lo, and R. Schober, “Robust beamforming
wireless power transfer system based on class-E2 DC-DC converter for secure communication in systems with wireless information and
for efficiency enhancement,” IEEE Trans. Circuits Syst. I, Reg. Papers, power transfer,” IEEE Trans. Wireless Commun., vol. 13, no. 8,
vol. 62, no. 11, pp. 2781–2791, Nov. 2015. pp. 4599–4615, Aug. 2014.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 299

[60] D. W. K. Ng, E. S. Lo, and R. Schober, “Wireless information and [83] N. Zhao, F. R. Yu, and V. C. M. Leung, “Opportunistic communications
power transfer: Energy efficiency optimization in OFDMA systems,” in interference alignment networks with wireless power transfer,” IEEE
IEEE Trans. Wireless Commun., vol. 12, no. 12, pp. 6352–6370, Wireless Commun., vol. 22, no. 1, pp. 88–95, Feb. 2015.
Dec. 2013. [84] S. Timotheou, G. Zheng, C. Masouros, and I. Krikidis, “Exploiting con-
[61] X. Zhou, “Training-based SWIPT: Optimal power splitting at the structive interference for simultaneous wireless information and power
receiver,” IEEE Trans. Veh. Technol., vol. 64, no. 9, pp. 4377–4382, transfer in multiuser downlink systems,” IEEE J. Sel. Areas Commun.,
Sep. 2015. vol. 34, no. 5, pp. 1772–1784, May 2016.
[62] X. Zhou, R. Zhang, and C. K. Ho, “Wireless information and power [85] A. Shojaeifard et al., “Self-interference in full-duplex multi-user
transfer: Architecture design and rate-energy tradeoff,” IEEE Trans. MIMO channels,” IEEE Commun. Lett., vol. 21, no. 4, pp. 841–844,
Commun., vol. 61, no. 11, pp. 4754–4767, Nov. 2013. Apr. 2017.
[63] I. Krikidis, S. Sasaki, S. Timotheou, and Z. Ding, “A low complexity [86] P. Grover and A. Sahai, “Shannon meets tesla: Wireless information
antenna switching for joint wireless information and energy transfer and power transfer,” in Proc. IEEE Int. Symp. Inf. Theory (ISIT), Austin,
in MIMO relay channels,” IEEE Trans. Commun., vol. 62, no. 5, TX, USA, 2010, pp. 2363–2367.
pp. 1577–1587, May 2014. [87] A. Naeem, M. H. Rehmani, Y. Saleem, I. Rashid, and N. Crespi,
[64] K. Huang and E. Larsson, “Simultaneous information and power trans- “Network coding in cognitive radio networks: A comprehensive sur-
fer for broadband wireless systems,” IEEE Trans. Signal Process., vey,” IEEE Commun. Surveys Tuts., vol. 19, no. 3, pp. 1945–1973,
vol. 61, no. 23, pp. 5972–5986, Dec. 2013. 3rd Quart., 2017.
[65] C. Shen, W.-C. Li, and T.-H. Chang, “Simultaneous information and [88] Z. Hu, N. Wei, and Z. Zhang, “Optimal resource allocation for har-
energy transfer: A two-user MISO interference channel case,” in Proc. vested energy maximization in wideband cognitive radio network with
IEEE Glob. Commun. Conf. (GLOBECOM), Anaheim, CA, USA, 2012, SWIPT,” IEEE Access, vol. 5, pp. 23383–23394, 2017.
pp. 3862–3867. [89] A. El Shafie, N. Al-Dhahir, and R. Hamila, “Cooperative access
[66] X. Zhou, R. Zhang, and C. K. Ho, “Wireless information and power schemes for efficient SWIPT transmissions in cognitive radio
transfer in multiuser OFDM systems,” IEEE Trans. Wireless Commun., networks,” in Proc. IEEE Globecom Workshops (GC Wkshps),
vol. 13, no. 4, pp. 2282–2294, Apr. 2014. San Diego, CA, USA, 2015, pp. 1–6.
[67] Z. Hu, C. Yuan, F. Zhu, and F. Gao, “Weighted sum transmit power [90] S. K. Sharma et al., “Cognitive radio techniques under practical imper-
minimization for full-duplex system with SWIPT and self-energy fections: A survey,” IEEE Commun. Surveys Tuts., vol. 17, no. 4,
recycling,” IEEE Access, vol. 4, pp. 4874–4881, 2016. pp. 1858–1884, 4th Quart., 2015.
[68] M. Gao, H. H. Chen, Y. Li, M. Shirvanimoghaddam, and J. Shi, “Full- [91] J. I. Choi, M. Jain, K. Srinivasan, P. Levis, and S. Katti, “Achieving sin-
duplex wireless-powered communication with antenna pair selection,” gle channel, full duplex wireless communication,” in Proc. 16th Annu.
in Proc. IEEE Wireless Commun. Netw. Conf. (WCNC), New Orleans, Int. Conf. Mobile Comput. Netw., Chicago, IL, USA, 2010, pp. 1–12.
LA, USA, 2015, pp. 693–698. [92] D. Jayakody, J. Thompson, S. Chatzinotas, and S. Durani, Wireless
[69] V. C. Gungor and G. P. Hancke, “Industrial wireless sensor networks: Information and Power Transfer: A New Green Communications
Challenges, design principles, and technical approaches,” IEEE Trans. Paradigm. Cham, Switzerland: Springer-Verlag, 2017.
Ind. Electron., vol. 56, no. 10, pp. 4258–4265, Oct. 2009. [93] M. Duarte, C. Dick, and A. Sabharwal, “Experiment-driven char-
[70] L. Lu, G. Y. Li, A. L. Swindlehurst, A. Ashikhmin, and R. Zhang, acterization of full-duplex wireless systems,” IEEE Trans. Wireless
“An overview of massive MIMO: Benefits and challenges,” IEEE J. Commun., vol. 11, no. 12, pp. 4296–4307, Dec. 2012.
Sel. Topics Signal Process., vol. 8, no. 5, pp. 742–758, Oct. 2014.
[94] E. Aryafar, M. A. Khojastepour, K. Sundaresan, S. Rangarajan, and
[71] C. Masouros, T. Ratnarajah, M. Sellathurai, C. B. Papadias, and
M. Chiang, “Midu: Enabling MIMO full duplex,” in Proc. 18th Annu.
A. K. Shukla, “Known interference in the cellular downlink: A
Int. Conf. Mobile Comput. Netw., Istanbul, Turkey, 2012, pp. 257–268.
performance limiting factor or a source of green signal power?” IEEE
[95] E. Ahmed and A. M. Eltawil, “All-digital self-interference cancellation
Commun. Mag., vol. 51, no. 10, pp. 162–171, Oct. 2013.
technique for full-duplex systems,” IEEE Trans. Wireless Commun.,
[72] G. Zheng et al., “Rethinking the role of interference in wireless
vol. 14, no. 7, pp. 3519–3532, Jul. 2015.
networks,” IEEE Commun. Mag., vol. 52, no. 11, pp. 152–158,
Nov. 2014. [96] A. A. Okandeji, M. R. A. Khandaker, and K.-K. Wong, “Wireless
[73] Y. Gu and S. Aïssa, “Interference aided energy harvesting in decode- information and power transfer in full-duplex communication systems,”
and-forward relaying systems,” in Proc. IEEE Int. Conf. Commun. in Proc. IEEE Int. Conf. Commun. (ICC), Kuala Lumpur, Malaysia,
(ICC), Sydney, NSW, Australia, 2014, pp. 5378–5382. 2016, pp. 1–6.
[74] C. Masouros, “Harvesting signal power from constructive interference [97] Y. Zeng and R. Zhang, “Full-duplex wireless-powered relay with
in multiuser downlinks,” in Wireless Information and Power Transfer: self-energy recycling,” IEEE Wireless Commun. Lett., vol. 4, no. 2,
A New Paradigm for Green Communications. New York, NY, USA: pp. 201–204, Apr. 2015.
Springer, 2018, pp. 87–122. [98] M. Maso et al., “Energy-recycling full-duplex radios for next-
[75] C. Psomas and I. Krikidis, “Successive interference cancellation in generation networks,” IEEE J. Sel. Areas Commun., vol. 33, no. 12,
bipolar ad hoc networks with SWIPT,” IEEE Wireless Commun. Lett., pp. 2948–2962, Dec. 2015.
vol. 5, no. 4, pp. 364–367, Aug. 2016. [99] C. Zhong, H. A. Suraweera, G. Zheng, I. Krikidis, and Z. Zhang,
[76] J. Park and B. Clerckx, “Joint wireless information and energy trans- “Wireless information and power transfer with full duplex relaying,”
fer in a k-user MIMO interference channel,” IEEE Trans. Wireless IEEE Trans. Commun., vol. 62, no. 10, pp. 3447–3461, Oct. 2014.
Commun., vol. 13, no. 10, pp. 5781–5796, Oct. 2014. [100] H. Ju and R. Zhang, “Optimal resource allocation in full-duplex
[77] I. Krikidis, “Simultaneous information and energy transfer in large- wireless-powered communication network,” IEEE Trans. Commun.,
scale networks with/without relaying,” IEEE Trans. Commun., vol. 62, vol. 62, no. 10, pp. 3528–3540, Oct. 2014.
no. 3, pp. 900–912, Mar. 2014. [101] X. Kang, C. K. Ho, and S. Sun, “Full-duplex wireless-powered com-
[78] T. A. Khan, A. Alkhateeb, and R. W. Heath, “Millimeter wave munication network with energy causality,” IEEE Trans. Wireless
energy harvesting,” IEEE Trans. Wireless Commun., vol. 15, no. 9, Commun., vol. 14, no. 10, pp. 5539–5551, Oct. 2015.
pp. 6048–6062, Sep. 2016. [102] S. Leng, D. W. K. Ng, N. Zlatanov, and R. Schober, “Multi-objective
[79] M. Alodeh, S. Chatzinotas, and B. Ottersten, “Constructive multiuser resource allocation in full-duplex SWIPT systems,” in Proc. IEEE Int.
interference in symbol level precoding for the MISO downlink chan- Conf. Commun. (ICC), Kuala Lumpur, Malaysia, 2016, pp. 1–7.
nel,” IEEE Trans. Signal Process., vol. 63, no. 9, pp. 2239–2252, [103] V.-D. Nguyen, H. V. Nguyen, G.-M. Kang, H. M. Kim, and O.-S. Shin,
May 2015. “Sum rate maximization for full duplex wireless-powered communica-
[80] M. Alodeh, S. Chatzinotas, and B. Ottersten, “Energy-efficient symbol- tion networks,” in Proc. 24th Eur. Signal Process. Conf. (EUSIPCO),
level precoding in multiuser MISO based on relaxed detection region,” Budapest, Hungary, 2016, pp. 798–802.
IEEE Trans. Wireless Commun., vol. 15, no. 5, pp. 3755–3767, [104] H. Stockman, “Communication by means of reflected power,” Proc.
May 2016. IRE, vol. 36, no. 10, pp. 1196–1204, Oct. 1948.
[81] C. Masouros and G. Zheng, “Exploiting known interference as green [105] G. Vannucci, A. Bletsas, and D. Leigh, “A software-defined radio
signal power for downlink beamforming optimization,” IEEE Trans. system for backscatter sensor networks,” IEEE Trans. Wireless
Signal Process., vol. 63, no. 14, pp. 3628–3640, Jul. 2015. Commun., vol. 7, no. 6, pp. 2170–2179, Jun. 2008.
[82] D. B. da Costa, H. Ding, and J. Ge, “Interference-limited relaying trans- [106] N. Fasarakis-Hilliard, P. N. Alevizos, and A. Bletsas, “Coherent detec-
missions in dual-hop cooperative networks over Nakagami-m fading,” tion and channel coding for bistatic scatter radio sensor networking,”
IEEE Commun. Lett., vol. 15, no. 5, pp. 503–505, May 2011. IEEE Trans. Commun., vol. 63, no. 5, pp. 1798–1810, May 2015.
300 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

[107] D. Kwon, H. S. Kang, and D. K. Kim, “Robust interference [129] L. Dong, Z. Han, A. P. Petropulu, and H. V. Poor, “Improving wire-
exploitation-based precoding scheme with quantized CSIT,” IEEE less physical layer security via cooperating relays,” IEEE Trans. Signal
Commun. Lett., vol. 20, no. 4, pp. 780–783, Apr. 2016. Process., vol. 58, no. 3, pp. 1875–1888, Mar. 2010.
[108] S. Timotheou, G. Zheng, C. Masouros, and I. Krikidis, “Symbol- [130] J. Li, A. P. Petropulu, and S. Weber, “On cooperative relaying schemes
level precoding in MISO broadcast channels for SWIPT systems,” in for wireless physical layer security,” IEEE Trans. Signal Process.,
Proc. 23rd Int. Conf. Telecommun. (ICT), Thessaloniki, Greece, 2016, vol. 59, no. 10, pp. 4985–4997, Oct. 2011.
pp. 1–5. [131] M. Zhang and Y. Liu, “Energy harvesting for physical-layer security
[109] M. Alodeh, S. Chatzinotas, and B. Ottersten, “Constructive interference in OFDMA networks,” IEEE Trans. Inf. Forensics Security, vol. 11,
through symbol level precoding for multi-level modulation,” in Proc. no. 1, pp. 154–162, Jan. 2016.
IEEE Glob. Commun. Conf. (GLOBECOM), San Diego, CA, USA, [132] M. Zhang, Y. Liu, and R. Zhang, “Artificial noise aided secrecy infor-
2015, pp. 1–6. mation and power transfer in OFDMA systems,” IEEE Trans. Wireless
[110] Q. Shi, L. Liu, W. Xu, and R. Zhang, “Joint transmit beamforming Commun., vol. 15, no. 4, pp. 3085–3096, Apr. 2016.
and receive power splitting for MISO SWIPT systems,” IEEE Trans. [133] G. Pan, C. Tang, T. Li, and Y. Chen, “Secrecy performance analysis for
Wireless Commun., vol. 13, no. 6, pp. 3269–3280, Jun. 2014. SIMO simultaneous wireless information and power transfer systems,”
[111] M. Peng, Y. Liu, D. Wei, W. Wang, and H.-H. Chen, “Hierarchical IEEE Trans. Commun., vol. 63, no. 9, pp. 3423–3433, Sep. 2015.
cooperative relay based heterogeneous networks,” IEEE Wireless [134] G. Pan et al., “On secrecy performance of MISO SWIPT systems
Commun., vol. 18, no. 3, pp. 48–56, Jun. 2011. with TAS and imperfect CSI,” IEEE Trans. Commun., vol. 64, no. 9,
[112] B. Zhou, H. Hu, S.-Q. Huang, and H.-H. Chen, “Intracluster device-to- pp. 3831–3843, Sep. 2016.
device relay algorithm with optimal resource utilization,” IEEE Trans. [135] Y. Yang, Q. Li, W.-K. Ma, J. Ge, and P. C. Ching, “Cooperative secure
Veh. Technol., vol. 62, no. 5, pp. 2315–2326, Jun. 2013. beamforming for AF relay networks with multiple eavesdroppers,”
[113] J. N. Laneman and G. W. Wornell, “Energy-efficient antenna sharing IEEE Signal Process. Lett., vol. 20, no. 1, pp. 35–38, Jan. 2013.
and relaying for wireless networks,” in Proc. IEEE Wireless Commun. [136] H. Xing, L. Liu, and R. Zhang, “Secrecy wireless information and
Netw. Conf., vol. 1. Chicago, IL, USA, 2000, pp. 7–12. power transfer in fading wiretap channel,” IEEE Trans. Veh. Technol.,
[114] J. N. Laneman, D. N. Tse, and G. W. Wornell, “Cooperative diversity vol. 65, no. 1, pp. 180–190, Jan. 2016.
in wireless networks: Efficient protocols and outage behavior,” IEEE [137] Y. Liu, “Wireless information and power transfer for multirelay-assisted
Trans. Inf. Theory, vol. 50, no. 12, pp. 3062–3080, Dec. 2004. cooperative communication,” IEEE Commun. Lett., vol. 20, no. 4,
[115] H. Chen, Y. Li, J. L. Rebelatto, B. F. U. Filho, and B. Vucetic, “Harvest- pp. 784–787, Apr. 2016.
then-cooperate: Wireless-powered cooperative communications,” IEEE [138] M. Marcus and B. Pattan, “Millimeter wave propagation: Spectrum
Trans. Signal Process., vol. 63, no. 7, pp. 1700–1711, Apr. 2015. management implications,” IEEE Microw. Mag., vol. 6, no. 2,
[116] A. Rajaram, D. N. K. Jayakody, and V. Skachek, “Store-then-cooperate: pp. 54–62, Jun. 2005.
Energy harvesting scheme in cooperative relay networks,” in Proc.
[139] A. Al-Hourani, S. Chandrasekharan, and S. Kandeepan, “Path loss
Int. Symp. Wireless Commun. Syst. (ISWCS), Poznań, Poland, 2016,
study for millimeter wave device-to-device communications in urban
pp. 445–450.
environment,” in Proc. IEEE Int. Conf. Commun. Workshops (ICC),
[117] W. Huang, H. Chen, Y. Li, and B. Vucetic, “On the performance of Sydney, NSW, Australia, 2014, pp. 102–107.
multi-antenna wireless-powered communications with energy beam-
[140] J. Lillibridge, R. Scharroo, S. Abdalla, and D. Vandemark, “One-and
forming,” IEEE Trans. Veh. Technol., vol. 65, no. 3, pp. 1801–1808,
two-dimensional wind speed models for Ka-band altimetry,” J. Atmos.
Mar. 2016.
Ocean. Technol., vol. 31, no. 3, pp. 630–638, 2014.
[118] I. Krikidis, G. Zheng, and B. Ottersten, “Harvest-use cooper-
ative networks with half/full-duplex relaying,” in Proc. IEEE [141] L. Wang, M. Elkashlan, R. W. Heath, M. Di Renzo, and K.-K. Wong,
Wireless Commun. Netw. Conf. (WCNC), Shanghai, China, 2013, “Millimeter wave power transfer and information transmission,” in
pp. 4256–4260. Proc. IEEE Glob. Commun. Conf. (GLOBECOM), San Diego, CA,
USA, 2015, pp. 1–6.
[119] A. A. Nasir, X. Zhou, S. Durrani, and R. A. Kennedy, “Relaying proto-
cols for wireless energy harvesting and information processing,” IEEE [142] S. Ladan, A. B. Guntupalli, and K. Wu, “A high-efficiency 24 GHz
Trans. Wireless Commun., vol. 12, no. 7, pp. 3622–3636, Jul. 2013. rectenna development towards millimeter-wave energy harvesting and
[120] Z. Ding, I. Krikidis, B. Sharif, and H. V. Poor, “Wireless informa- wireless power transmission,” IEEE Trans. Circuits Syst. I, Reg. Papers,
tion and power transfer in cooperative networks with spatially random vol. 61, no. 12, pp. 3358–3366, Dec. 2014.
relays,” IEEE Trans. Wireless Commun., vol. 13, no. 8, pp. 4440–4453, [143] A. Nesic, Z. Micic, S. Jovanovic, and I. Radnovic, “Millimeter wave
Aug. 2014. printed antenna array with high side lobe suppression,” in Proc. IEEE
[121] S. R. Islam, N. Avazov, O. A. Dobre, and K.-S. Kwak, “Power-domain Antennas Propag. Soc. Int. Symp., Albuquerque, NM, USA, 2006,
non-orthogonal multiple access (NOMA) in 5G systems: Potentials and pp. 3051–3054.
challenges,” IEEE Commun. Surveys Tuts., vol. 19, no. 2, pp. 721–742, [144] M. Peer, N. Jain, and V. A. Bohara, “A hybrid spectrum sharing pro-
2nd Quart., 2017. tocol for energy harvesting wireless sensor nodes,” in Proc. IEEE
[122] Z. Ding et al., “A survey on non-orthogonal multiple access for 5G 17th Int. Workshop Signal Process. Adv. Wireless Commun. (SPAWC),
networks: Research challenges and future trends,” IEEE J. Sel. Areas Edinburgh, U.K., 2016, pp. 1–6.
Commun., vol. 35, no. 10, pp. 2181–2195, Oct. 2017. [145] I. F. Akyildiz, T. Melodia, and K. R. Chowdhury, “A survey on wire-
[123] Z. Ding, M. Peng, and H. V. Poor, “Cooperative non-orthogonal less multimedia sensor networks,” Comput. Netw., vol. 51, no. 4,
multiple access in 5G systems,” IEEE Commun. Lett., vol. 19, no. 8, pp. 921–960, 2007.
pp. 1462–1465, Aug. 2015. [146] B. Tong, Z. Li, G. Wang, and W. Zhang, “How wireless power charging
[124] Y. Liu, Z. Ding, M. Elkashlan, and H. V. Poor, “Cooperative non- technology affects sensor network deployment and routing,” in Proc.
orthogonal multiple access with simultaneous wireless information IEEE 30th Int. Conf. Distrib. Comput. Syst. (ICDCS), Genoa, Italy,
and power transfer,” IEEE J. Sel. Areas Commun., vol. 34, no. 4, 2010, pp. 438–447.
pp. 938–953, Apr. 2016. [147] Z. Popović, E. A. Falkenstein, D. Costinett, and R. Zane, “Low-power
[125] Z. Chen, Z. Ding, X. Dai, and G. K. Karagiannidis, “On the application far-field wireless powering for wireless sensors,” Proc. IEEE, vol. 101,
of quasi-degradation to MISO-NOMA downlink,” IEEE Trans. Signal no. 6, pp. 1397–1409, Jun. 2013.
Process., vol. 64, no. 23, pp. 6174–6189, Dec. 2016. [148] S. Guo, C. Wang, and Y. Yang, “Mobile data gathering with wireless
[126] Z. Chen, Z. Ding, P. Xu, and X. Dai, “Optimal precoding for a energy replenishment in rechargeable sensor networks,” in Proc. IEEE
qos optimization problem in two-user MISO-NOMA downlink,” IEEE INFOCOM, Turin, Italy, 2013, pp. 1932–1940.
Commun. Lett., vol. 20, no. 6, pp. 1263–1266, Jun. 2016. [149] S. Guo, C. Wang, and Y. Yang, “Joint mobile data gathering and energy
[127] N. T. Do, D. B. da Costa, T. Q. Duong, and B. An, “Transmit antenna provisioning in wireless rechargeable sensor networks,” IEEE Trans.
selection schemes for MISO-NOMA cooperative downlink transmis- Mobile Comput., vol. 13, no. 12, pp. 2836–2852, Dec. 2014.
sions with hybrid SWIPT protocol,” in Proc. IEEE Int. Conf. Commun. [150] C. Wang, J. Li, F. Ye, and Y. Yang, “NETWRAP: An NDN based
(ICC), Paris, France, 2017, pp. 1–6. real-timewireless recharging framework for wireless sensor networks,”
[128] H. Xing, K.-K. Wong, A. Nallanathan, and R. Zhang, “Wireless pow- IEEE Trans. Mobile Comput., vol. 13, no. 6, pp. 1283–1297, Jun. 2014.
ered cooperative jamming for secrecy multi-AF relaying networks,” [151] S. Zhang, J. Wu, and S. Lu, “Collaborative mobile charging for sensor
IEEE Trans. Wireless Commun., vol. 15, no. 12, pp. 7971–7984, networks,” in Proc. IEEE 9th Int. Conf. Mobile Adhoc Sensor Syst.
Dec. 2016. (MASS), Las Vegas, NV, USA, 2012, pp. 84–92.
PONNIMBADUGE PERERA et al.: SWIPT: RECENT ADVANCES AND FUTURE CHALLENGES 301

[152] J. Johnson, E. Basha, and C. Detweiler, “Charge selection algorithms [174] S. Guo, F. Wang, Y. Yang, and B. Xiao, “Energy-efficient cooperative
for maximizing sensor network life with UAV-based limited wireless Tfor simultaneous wireless information and power transfer in clus-
recharging,” in Proc. IEEE 8th Int. Conf. Intell. Sensors Sensor Netw. tered wireless sensor networks,” IEEE Trans. Commun., vol. 63, no. 11,
Inf. Process., Melbourne, VIC, Australia, 2013, pp. 159–164. pp. 4405–4417, Nov. 2015.
[153] M. Y. Naderi et al., “Experimental study of concurrent data and [175] M. T. Mushtaq, S. A. Hassan, S. Saleem, and D. N. K. Jayakody,
wireless energy transfer for sensor networks,” in Proc. IEEE Glob. “Impacts of K-fading on the performance of massive MIMO systems,”
Commun. Conf. (GLOBECOM), 2014, pp. 2543–2549. IET Electron. Lett., vol. 54, no. 1, pp. 49–51, Jan. 2017.
[154] G. Amarasuriya and H. V. Poor, “Wireless information and power trans- [176] Q. Shi, W. Xu, J. Wu, E. Song, and Y. Wang, “Secure beamforming
fer in multi-way relay networks with massive MIMO,” in Proc. IEEE for MIMO broadcasting with wireless information and power trans-
Glob. Commun. Conf. (GLOBECOM), San Diego, CA, USA, 2015, fer,” IEEE Trans. Wireless Commun., vol. 14, no. 5, pp. 2841–2853,
pp. 1–7. May 2015.
[155] P. Phunchongharn, E. Hossain, and D. I. Kim, “Resource alloca- [177] E. Boshkovska, R. Morsi, D. W. K. Ng, and R. Schober, “Power alloca-
tion for device-to-device communications underlaying LTE-advanced tion and scheduling for SWIPT systems with non-linear energy harvest-
networks,” IEEE Wireless Commun., vol. 20, no. 4, pp. 91–100, ing model,” in Proc. IEEE Int. Conf. Commun. (ICC), Kuala Lumpur,
Aug. 2013. Malaysia, 2016, pp. 1–6.
[156] S. S. Nam, D. I. Kim, and H.-C. Yang, “Modified dynamic DF for [178] D. K. Nguyen, D. N. K. Jayakody, S. Chatzinotas, J. S. Thompson, and
type-2 UE relays,” in Proc. IEEE Wireless Commun. Netw. Conf. J. Li, “Wireless energy harvesting assisted two-way cognitive relay
(WCNC), Shanghai, China, 2012, pp. 1392–1397. networks: Protocol design and performance analysis,” IEEE Access,
[157] H. H. Yang, J. Lee, and T. Q. S. Quek, “Green device-to-device com- vol. 5, pp. 21447–21460, 2017.
munication with harvesting energy in cellular networks,” in Proc. 6th [179] V. P. Tuan, S. Q. Nguyen, and H. Y. Kong, “Performance analysis
Int. Conf. Wireless Commun. Signal Process. (WCSP), Hefei, China, of energy-harvesting relay selection systems with multiple antennas in
2014, pp. 1–6. presence of transmit hardware impairments,” in Proc. Int. Conf. Adv.
[158] Y. Liu, L. Wang, S. A. R. Zaidi, M. Elkashlan, and T. Q. Duong, Technol. Commun. (ATC), Hanoi, Vietnam, 2016, pp. 126–130.
“Secure D2D communication in large-scale cognitive cellular networks: [180] I. Tal and A. Vardy, “How to construct polar codes,” IEEE Trans. Inf.
A wireless power transfer model,” IEEE Trans. Commun., vol. 64, Theory, vol. 59, no. 10, pp. 6562–6582, Oct. 2013.
no. 1, pp. 329–342, Jan. 2016. [181] H. Kong, C. Xing, S. Zhao, and P. Shi, “Cooperative coding scheme
[159] S. Lee, R. Zhang, and K. Huang, “Opportunistic wireless energy har- using polar codes,” in Proc. 2nd Int. Conf. Comput. Sci. Netw. Technol.
vesting in cognitive radio networks,” IEEE Trans. Wireless Commun., (ICCSNT), Changchun, China, 2012, pp. 602–606.
vol. 12, no. 9, pp. 4788–4799, Sep. 2013. [182] B. Yuan and K. K. Parhi, “Early stopping criteria for energy-efficient
[160] S. K. Sharma et al., “Dynamic spectrum sharing in 5g wire- low-latency belief-propagation polar code decoders,” IEEE Trans.
less networks with full-duplex technology: Recent advances and Signal Process., vol. 62, no. 24, pp. 6496–6506, Dec. 2014.
research challenges,” IEEE Commun. Surveys Tuts., to be published, [183] R. Blasco-Serrano, R. Thobaben, M. Andersson, V. Rathi, and
doi: 10.1109/COMST.2017.2773628. M. Skoglund, “Polar codes for cooperative relaying,” IEEE Trans.
[161] Z. Zhang, K. Long, A. V. Vasilakos, and L. Hanzo, “Full-duplex Commun., vol. 60, no. 11, pp. 3263–3273, Nov. 2012.
wireless communications: Challenges, solutions, and future research [184] K. He, J. Sha, L. Li, and Z. Wang, “Low power decoder design for
directions,” Proc. IEEE, vol. 104, no. 7, pp. 1369–1409, Jul. 2016. QC-LDPC codes,” in Proc. IEEE Int. Symp. Circuits Syst. (ISCAS),
[162] T. Riihonen and R. Wichman, “Energy detection in full-duplex cog- Paris, France, 2010, pp. 3937–3940.
nitive radios under residual self-interference,” in Proc. 9th Int. Conf. [185] Y. S. Park, D. Blaauw, D. Sylvester, and Z. Zhang, “Low-power high-
Cogn. Radio Orient. Wireless Netw. Commun. (CROWNCOM), Oulu, throughput LDPC decoder using non-refresh embedded dram,” IEEE
Finland, 2014, pp. 57–60. J. Solid-State Circuits, vol. 49, no. 3, pp. 783–794, Mar. 2014.
[163] G. Zheng, I. Krikidis, J. Li, A. P. Petropulu, and B. Ottersten, [186] J. Kaza and C. Chakrabarti, “Design and implementation of low-energy
“Improving physical layer secrecy using full-duplex jam- turbo decoders,” IEEE Trans. Very Large Scale Integr. (VLSI) Syst.,
ming receivers,” IEEE Trans. Signal Process., vol. 61, no. 20, vol. 12, no. 9, pp. 968–977, Sep. 2004.
pp. 4962–4974, Oct. 2013. [187] R. G. Maunder, “The 5G channel code contenders,” Southampton,
[164] S. K. Sharma et al., “Two-phase concurrent sensing and transmis- U.K., AccelerComm, White Paper, pp. 1–13, 2016.
sion scheme for full duplex cognitive radio,” in Proc. IEEE 84th Veh. [188] S. K. Sharma, T. E. Bogale, S. Chatzinotas, X. Wang, and L. B. Le,
Technol. Conf. (VTC Fall), Montreal, QC, Canada, 2016, pp. 1–5. “Physical layer aspects of wireless IoT,” in Proc. Int. Symp. Wireless
[165] S. Hu, Z. Ding, and Q. Ni, “Beamforming optimisation in energy Commun. Syst. (ISWCS), Poznań, Poland, 2016, pp. 304–308.
harvesting cooperative full-duplex networks with self-energy recycling [189] G. Reiter, “Wireless connectivity for the Internet of Things,” Texas
protocol,” IET Commun., vol. 10, no. 7, pp. 848–853, May 2016. Instrum., Dallas, TX, USA, White Paper, vol. 433, 2014.
[166] Z. Wen, X. Liu, N. C. Beaulieu, R. Wang, and S. Wang, “Joint source [190] S. Zhou, T. Chen, W. Chen, and Z. Niu, “Outage minimization for a
and relay beamforming design for full-duplex MIMO AF relay SWIPT fading wireless link with energy harvesting transmitter and receiver,”
systems,” IEEE Commun. Lett., vol. 20, no. 2, pp. 320–323, Feb. 2016. IEEE J. Sel. Areas Commun., vol. 33, no. 3, pp. 496–511, Mar. 2015.
[167] D. N. K. Jayakody and M. F. Flanagan, “A soft decode–compress– [191] B. W. Khoueiry and M. R. Soleymani, “A novel machine-to-machine
forward relaying scheme for cooperative wireless networks,” IEEE communication strategy using rateless coding for the Internet of
Trans. Veh. Technol., vol. 65, no. 5, pp. 3033–3041, May 2016. Things,” IEEE Internet Things J., vol. 3, no. 6, pp. 937–950, Dec. 2016,
[168] D. N. Jayakody, J. Li, and M. F. Flanagan, “A soft-network-coded doi: 10.1109/JIOT.2016.2518925.
multilevel forwarding scheme for multiple-access relay systems,” IEEE [192] H. Matsumoto and K. Hashimoto, “Report of the URSI
Trans. Veh. Technol., vol. 65, no. 5, pp. 3430–3439, May 2016. inter-commission working group on SPS and appendices,” Solar Power
[169] O. Taghizadeh, A. Zamani, and R. Mathar, “Physical-layer security for Satellite Syst. Gen. Assembly Sci. Symp. Int. Union Radio Sci., Ghent,
simultaneous information and power transfer in full-duplex multi-user Belgium, White Paper, 2006. [Online]. Available: http://www.ursi.org
networks,” in Proc. 20th Int. ITG Workshop Smart Antennas, Munich,
Germany, 2016, pp. 1–8.
[170] F. Zhu, F. Gao, and M. Yao, “Zero-forcing beamforming for phys-
ical layer security of energy harvesting wireless communications,” Tharindu D. Ponnimbaduge Perera (S’17)
EURASIP J. Wireless Commun. Netw., vol. 2015, no. 1, p. 58, received the B.Sc. degree (First Class Hons.) in
2015. [Online]. Available: https://www.hindawi.com/journals/ijap/ software engineering and the M.Sc. degree in
2013/385260/ GSD from the Department of Computing, Faculty
[171] A. Salem, K. A. Hamdi, and K. M. Rabie, “Physical layer security with of ACES, Sheffield Hallam University, U.K. He
RF energy harvesting in AF multi-antenna relaying networks,” IEEE is currently pursuing the Ph.D. degree in com-
Trans. Commun., vol. 64, no. 7, pp. 3025–3038, Jul. 2016. puter science and wireless communications from
[172] L. Liu, R. Zhang, and K.-C. Chua, “Secrecy wireless informa- Infocomm Laboratory, School of Computer Science
tion and power transfer with MISO beamforming,” in Proc. IEEE and Robotics, National Research Tomsk Polytechnic
Glob. Commun. Conf. (GLOBECOM), Atlanta, GA, USA, 2013, University, Russia, where he is currently a Research
pp. 1831–1836. Engineer. His research interests include simultane-
[173] M. Tabesh, N. Dolatsha, A. Arbabian, and A. M. Niknejad, “A power- ous wireless information and power transfer, interference exploitation in RF
harvesting pad-less millimeter-sized radio,” IEEE J. Solid-State energy harvesting, and channel coding techniques in simultaneous wireless
Circuits, vol. 50, no. 4, pp. 962–977, Apr. 2015. information and power transfer.
302 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 20, NO. 1, FIRST QUARTER 2018

Dushantha Nalin K. Jayakody (M’14) received the Symeon Chatzinotas (S’06–M’09–SM’13) received
B.Eng. degree (First Class Hons. and Merit Position the M.Eng. degree in telecommunications from the
Holder) from Pakistan (under SAARC Scholarship), Aristotle University of Thessaloniki, Thessaloniki,
the M.Sc. degree (First Merit Position Holder) in Greece, in 2003 and the M.Sc. and Ph.D. degrees
electronics and communications engineering from in electronic engineering from the University of
Eastern Mediterranean University, Turkey, (under the Surrey, Surrey, U.K., in 2006 and 2009, respec-
University Fellowship), and the Ph.D. degree in elec- tively. He was involved in numerous Research
tronics and communications engineering from the and Development projects with the Institute of
University College Dublin, Ireland, under Science Informatics Telecommunications, National Center
Foundation Ireland grant. He is currently a Professor for Scientific Research Demokritos, the Institute of
with the School of Computer Science and Robotics, Telematics and Informatics, Center of Research and
National Research Tomsk Polytechnic University, Russia. From 2014 to 2016, Technology Hellas, and the Mobile Communications Research Group, Center
he has held a Research Fellow positions with the University of Tartu, Estonia, of Communication Systems Research, University of Surrey. He is currently
and the University of Bergen, Norway. He has published one book, nearly 80 the Deputy Head of the SIGCOM Research Group, Interdisciplinary Centre
journal papers, book chapters, and conference papers. He has 6 Ph.D. students for Security, Reliability, and Trust, University of Luxembourg, Luxembourg,
(current and past). He was a recipient of the Best Paper Award from the IEEE and a Visiting Professor with the University of Parma, Italy. He has over 250
International Conference on Communication, Management and Information publications and 2000 Google Scholar citations with an H-index of 24. His
Technology, Warsaw, Poland, in 2017. He also serves as a reviewer for var- research interests include multiuser information theory, co-operative/cognitive
ious IEEE Transactions and other journals. He has served as the Session communications, and wireless networks optimization. He was a co-recipient
Chair or the Technical Program Committee Member for various international of the 2014 Distinguished Contributions to Satellite Communications Award,
conferences, such as IEEE PIMRC 2013/2014, IEEE WCNC 2014/2016, and the Satellite and Space Communications Technical Committee, the IEEE
and IEEE VTC 2015/2016. He currently serves as a Lead Guest Editor for Communications Society, and the CROWNCOM 2015 Best Paper Award.
MDPI Information journal (invited) and an Editor of Physical Communication
(Elsevier).

Shree Krishna Sharma (S’12–M’15) received

the M.Sc. degree in information and communica-
tion engineering from the Institute of Engineering,
Pulchowk, Nepal, the M.A. degree in economics
from Tribhuvan University, Nepal, the M.Res.
degree in computing science from Staffordshire
University, Staffordshire, U.K., and the Ph.D. degree
in wireless communications from the University of
Luxembourg, Luxembourg, in 2014, where he was a
Research Associate with the Interdisciplinary Centre
for Security, Reliability and Trust, for two years, and
was involved in EU FP7 CoRaSat Project, EU H2020 SANSA, ESA Project
ASPIM, as well as Luxembourgish national projects Co2Sat, and SeMIGod.
He is currently a Post-Doctoral Fellow with Western University, Canada.
His research interests include resource-efficient design of 5G and beyond
wireless systems, dynamic spectrum sharing, multiantenna signal processing, Jun Li (M’09–SM’16) received the Ph.D. degree
wireless Internet of Things, and adaptive optimization of distributed com- in electronic engineering from Shanghai Jiao Tong
munication, computing, and caching resources. In the past, he was involved University, Shanghai, China, in 2009. In 2009,
with Kathmandu University, Dhulikhel, Nepal, as a Teaching Assistant, and he was with the Department of Research and
he also a Part-Time Lecturer for eight engineering colleges in Nepal. He Innovation, Alcatel Lucent Shanghai Bell as a
was with Nepal Telecom for over four years as a Telecom Engineer in the Research Scientist for five months. From 2009 to
field of information technology and telecommunication. He has authored over 2012, he was a Post-Doctoral Fellow with the School
80 technical papers in refereed international journals, scientific books, and of Electrical Engineering and Telecommunications,
conferences. He was a recipient of the Indian Embassy Scholarship for his University of New South Wales, Australia. From
B.E. study, the Erasmus Mundus Scholarship for his M.Res. study, and the 2012 to 2015, he was a Research Fellow with
AFR Ph.D. grant from the National Research Fund (FNR) of Luxembourg, the the School of Electrical Engineering, University of
Best Paper Award in CROWNCOM 2015 Conference, and the FNR Award for Sydney, Australia. Since 2015, he has been a Professor with the School
Outstanding Ph.D. Thesis 2015 for his Ph.D. thesis from FNR, Luxembourg. of Electronic and Optical Engineering, Nanjing University of Science and
He has been serving as a reviewer for several international journals and confer- Technology, Nanjing, China. His research interests include network informa-
ences, and also as a TPC Member for a number of international conferences, tion theory, channel coding theory, wireless network coding, and cooperative
including IEEE ICC, IEEE PIMRC, IEEE Globecom, and IEEE ISWCS. communications.