Content uploaded by Chen Chen
Author content
All content in this area was uploaded by Chen Chen on May 12, 2021
Content may be subject to copyright.
Energy-Efficient NOMA with QoS-Guaranteed
Power Allocation for Multi-User VLC
Chen Chen1,*, Shu Fu1, Xin Jian1, Min Liu1, and Xiong Deng2
1School of Microelectronics and Communication Engineering, Chongqing University, China
2Department of Electrical Engineering, Eindhoven University of Technology (TU/e), Netherlands
*c.chen@cqu.edu.cn
Abstract—In this paper, we propose an energy-efficient non-
orthogonal multiple access (NOMA) technique for multi-user
visible light communication (VLC) systems, by adopting a low-
complexity quality of service (QoS)-guaranteed power allocation
strategy. From the perspective of energy saving, we design the
energy-efficient NOMA-enabled multi-user VLC system with the
goal to achieve maximal energy efficiency (EE). The closed-form
QoS-guaranteed optimal power allocation strategy is obtained
and the analytical EE of the proposed NOMA-enabled multi-
user VLC system is derived. The analytical and simulation results
show that, for a VLC system with ten users, the average EE can
be improved by 19% when adopting the proposed NOMA in
comparison to NOMA with gain ratio power allocation (GRPA),
and the EE improvement becomes much more significant when
the users have more diverse QoS requirements.
Index Terms—Visible light communication (VLC), multi-user,
non-orthogonal multiple access (NOMA), power allocation.
I. INTRODUCTION
Due to the explosive growth of smart mobile devices (e.g.,
mobile phones and tablets) in our daily life, which demand
data-hungry services such as high definition video streaming,
low-latency gaming, virtual/augmented reality and Internet-of-
things, the mobile data traffic is expected to be exponentially
increased in the near future [1]. The current radio frequency
(RF) systems suffer from the spectrum congestion and hence
might not be able to support the ever-increasing mobile data
traffic. In recent years, the wide application of white light-
emitting diodes (LEDs) for general indoor illumination has
made visible light communication (VLC) a potential candidate
to support next-generation high-speed wireless communica-
tions [2]. Compared with traditional RF systems, VLC systems
enjoy many inherent advantages such as huge and unregulated
spectrum, potentially high data rate, low-cost front-ends, high
security and no electromagnetic interference [3], [4]. There-
fore, VLC has been considered as a promising complementary
technology to traditional RF technologies, which has triggered
great interest in both academia and industry.
In practical VLC systems, white LEDs serve a dual-function
of simultaneous illumination and communication. Generally, a
typical room can be divided into multiple optical attocells and
the LED transmitter of an optical attocell needs to broadcast
signal to all the users within its coverage [5], [6]. Therefore,
an efficient multiple access technique is essential for the
This work was supported by the National Natural Science Foundation of
China under Grant 61901065.
VLC system to simultaneously support multiple users. So far,
several multiple access techniques have been considered for
multi-user VLC systems. In [7] and [8], orthogonal frequency
division multiple access (OFDMA) has been applied to support
multiple users in VLC systems, where the frequency resources
of the VLC system are split and allocated to different users. In
[9] and [10], time division multiple access (TDMA) has been
adopted in multi-user VLC systems, where different users are
allocated with different time slots. Both OFDMA and TDMA
can be categorized as orthogonal multiple access (OMA) tech-
niques. In OMA-based multi-user VLC systems, the mutual
interference between different users can be eliminated by
allocating them with different time/frequency resources. An-
other category of multiple access techniques is non-orthogonal
multiple access (NOMA), which allows users to share the same
time/frequency resources through power domain superposition
coding (SPC) and successive interference cancellation (SIC)
[11]. Due to its efficient resource utilization, NOMA has been
viewed as a promising solution for the fifth generation (5G)
and beyond wireless networks [12].
Recently, the application of NOMA in VLC systems has
attracted great attention. In [13], a gain ratio power alloca-
tion (GRPA) strategy was proposed for NOMA-based VLC
systems. In [14], a theoretical framework was presented to
analyze the performance of NOMA in VLC systems. In [15],
user grouping and power allocation were studied in NOMA-
based multi-cell VLC networks. In [16], the bit-error-rate
performance of NOMA-based VLC systems with noisy and
outdated channel state information was analyzed. In [17],
NOMA was applied to multiple-input multiple-output VLC
systems and a normalized gain difference power allocation
strategy was further proposed. In [18], a SIC-free NOMA
scheme based on constellation partitioning coding was pro-
posed to mitigate error propagation caused by imperfect SIC
in VLC systems. In [19], the impact of random receiver
orientation on the performance of NOMA-based VLC systems
was investigated. These recent works have demonstrated the
superiority of NOMA over OMA in multi-user VLC systems.
The key to implement NOMA is to obtain a proper power
allocation strategy. In most NOMA-based VLC systems, the
achievable rate is generally adopted as the performance metric
and the power allocation strategy is designed to maximize
the system rate. Nevertheless, energy consumption is another
important issue that needs to be considered in practical VLC
systems. Hence, it is of practical significance to design multi-
user VLC systems from the perspective of energy saving.
In this paper, we propose an energy-efficient NOMA tech-
nique for multi-user VLC systems. The proposed NOMA
adopts a quality of service (QoS)-guaranteed power allocation
strategy, which is derived by maximizing the energy efficiency
(EE) of the multi-user VLC system while satisfying the QoS
requirements of all the users. The derived QoS-guaranteed
optimal power allocation strategy is given in the closed form,
which can enable low-complexity power allocation for the
proposed NOMA-enabled multi-user VLC systems. Monte
Carlo simulations are further performed to substantiate the
derived analytical results. The obtained analytical and simula-
tion results clearly demonstrate the advantages of the proposed
NOMA in comparison to OMA and NOMA with GRPA in a
typical multi-user VLC system.
The rest of this paper is organized as follows. Section II
describes the channel model of a typical multi-user VLC sys-
tem. Multi-user VLC using the proposed NOMA technique is
discussed in Section III. The analytical and simulation results
are presented in Section IV. Finally, Section V concludes the
paper.
II. CH AN NE L MOD EL
In practical multi-user VLC systems, the photodiode (PD)
of each user can receive both line-of-sight (LOS) and non-
line-of-sight (NLOS) components of the transmitted optical
signal. Nevertheless, since the NLOS component usually has
much lower electrical power than that of the LOS component,
it is reasonable to neglect the NLOS component during most
channel conditions [20]. For simplicity, we only consider the
LOS component in the following channel model. We assume
that the LED is oriented vertically downwards while the PD
of each user is oriented vertically upwards. Moreover, we
also assume that the overall VLC system has a flat frequency
response and the LED transmitter operates within its linear
dynamic range.
Assuming the VLC system serves totally Kusers and the
white LED has a Lambertian emission pattern, the LOS direct
current (DC) channel gain between the LED and the k-th (k=
1,2,·· · , K) user can be calculated by [3]
hk=
(m+ 1)ρA
2πd2
k
cosm(ψk)gfglcos(φk),0≤φk≤Φ
0, φk>Φ
,
(1)
where m=−ln2/ln(cos(Ψ)) denotes the Lambertian emis-
sion order and Ψis the semi-angle of the LED; ρand Aare
the responsivity and active area of the PD, respectively; dkis
the distance between the LED and the k-th user; ψkand φk
denote the corresponding emission angle and incident angle,
respectively; gfand glrepresent the gains of the optical filter
and lens, respectively. The gain of the optical lens is given by
gl=n2
sin2Φ, where nis the refractive index of the optical lens
and Φis the half-angle field-of-view (FOV) of the PD.
Fig. 1. Conceptual diagram of power domain energy-efficient NOMA.
The additive noise in typical VLC systems usually consists
of thermal and shot noises, which are generally modeled as
real-valued zero-mean additive white Gaussian noises. For
simplicity, it is assumed that the additive noises of all the users
have the same constant noise power spectral density (PSD) N0.
For a signal bandwidth B, the noise power is given by N0B.
III. MULTI -U SE R VL C US IN G ENERGY-E FFIC IE NT NOMA
A. Principle of Energy-Efficient NOMA
Fig. 1 illustrates the conceptual diagram of the proposed
power domain energy-efficient NOMA with two users, where
the QoS requirements for user 1 and user 2 are represented by
QoS 1 and QoS 2, respectively. Differing from conventional
OMA schemes, such as OFDMA and TDMA, which split
time or frequency resources to support multiple users in order
to eliminate mutual interference, the proposed NOMA allows
both users to utilize all the time and frequency resources. It
can be seen that the transmitted data of user 1 and user 2 are
superposed in the power domain and there inevitably exists
mutual interference. To ensure the QoS requirements of both
users, user 1 and user 2 are allocated with powers P1and
P2, respectively. Compared with conventional NOMA which
aims to maximize the system sum rate, the proposed NOMA is
designed to maximize the system EE while satisfying the QoS
requirements of all the users. Therefore, the key to implement
the proposed NOMA is to obtain a proper power allocation
strategy. Moreover, to support a large number of users when
implementing the proposed NOMA, an efficient user pairing
approach is also needed.
B. Energy-Efficient NOMA-Enabled Multi-User VLC
In this subsection, power domain energy-efficient NOMA is
introduced for multi-user VLC systems.
1) System Model: Without loss of generality, we assume
that the VLC system serves K= 2Nusers, which are divided
into Nuser pairs. Fig. 2 shows the schematic diagram of the
proposed NOMA-enabled multi-user VLC. Let si,f and si,n
denote the modulated message signals intended for the far and
near users in the i-th user pair, respectively. The superposed
electrical signal of all Npairs of users to be transmitted by
the LED can be expressed by
x=
N
X
i=1
√pi,f si,f +√pi,nsi,n +IDC ,(2)
Fig. 2. Schematic of energy-efficient NOMA-enabled multi-user VLC.
where pi,f and pi,n are the electrical transmit powers allocated
to the far and near users in the i-th user pair, respectively;
IDC is the DC bias current, which is added to simultaneously
guarantee the non-negativity of the LED driving signal and
ensure sufficient and stable illumination. The total electrical
transmit power allocated to all Npairs of users is obtained by
Pelec =
N
X
i=1
pi,f +pi,n.(3)
After removing the DC term, the received signals of the far
and near users in the i-th user pair can be given by
(yi,f =hi,f (√pi,f si,f +√pi,nsi,n ) + zi,f
yi,n =hi,n(√pi,f si,f +√pi,n si,n) + zi,n
,(4)
where hi,f and hi,n denote the channel gains of the far and
near users in the i-th user pair, respectively, and hi,f ≤hi,n;
zi,f and zi,n are the corresponding additive noises.
To decode the intended message signals for the far and near
users in the i-th user pair, the far user decodes its message
signal directly by treating the intended message signal for the
near user as interference, while the near user needs to decode
the intended message signal for the far user first and then apply
SIC to decode its own message signal without interference. As
a result, (4) can be re-represented as
yi,f =hi,f √pi,f si,f
| {z }
signal
+hi,f √pi,nsi,n
| {z }
interference
+zi,f
yi,n =hi,n√pi,f si,f
| {z }
SIC
+hi,n√pi,n si,n
| {z }
signal
+zi,n
.(5)
Following (5), the achievable rates of the far and near users
in the i-th user pair can be given by
Ri,f =1
2log2 h2
i,f pi,f
h2
i,f pi,n +Pz!
Ri,n =1
2log2 h2
i,npi,n
Pz!,(6)
where the scaling factor 1
2is due to the Hermitian symmetry
[14] and Pz=N0Bis the power of the additive noises. In
addition, the achievable rate for the near user to decode the
far user’s message signal is obtained by
Ri,n→f=1
2log2 h2
i,npi,f
h2
i,npi,n +Pz!.(7)
In practical multi-user VLC systems, different users might
have different QoS requirements. Generally, we can define the
QoS requirement of a specific user as its required achievable
rate per bandwidth, i.e., spectral efficiency [14]. Let e
Ri,f and
e
Ri,n denote the rate requirements of the far and near users in
the i-th user pair, respectively. To meet their rate requirements,
the following conditions need to be satisfied:
Ri,f ≥e
Ri,f
Ri,n ≥e
Ri,n
Ri,n→f≥e
Ri,f
.(8)
According to (8), the power requirements of two users can
be given by
pi,f ≥22
e
Ri,f pi,n +Pz
h2
i,f !
pi,n ≥22
e
Ri,n Pz
h2
i,n
pi,f ≥22
e
Ri,f pi,n +Pz
h2
i,n !
.(9)
Using hi,f ≤hi,n, we can obtain the power requirements to
satisfy the QoS requirements of both the far and near users in
the i-th user pair:
pi,f ≥22
e
Ri,f 22
e
Ri,n Pz
h2
i,n
+Pz
h2
i,f !
pi,n ≥22
e
Ri,n Pz
h2
i,n
.(10)
Hence, the total electrical transmit power requirements of
all Npairs of users is given by
Pelec ≥
N
X
i=1
22
e
Ri,f 22
e
Ri,n Pz
h2
i,n
+Pz
h2
i,f !+ 22
e
Ri,n Pz
h2
i,n
.(11)
It can be found that the QoS requirements of the far and
near users in the i-th user pair can be guaranteed with mutual
interference by allocating them with proper powers.
2) QoS-Guaranteed Optimal Power Allocation: For the
efficient implementation of the proposed NOMA in multi-
user VLC systems, a QoS-guaranteed optimal power allocation
strategy is proposed. In this work, we design the proposed
NOMA-enabled multi-user VLC systems from the energy
consumption perspective. Therefore, the goal to design the
proposed-enabled multi-user VLC systems is to maximize the
EE through a QoS-guaranteed optimal power allocation strat-
egy. The EE (η) can be defined as the ratio of the achievable
sum rate (R) to the total electrical power consumption (Pelec):
η=R
Pelec
,(12)
where the unit of EE is bits/J/Hz.
For the proposed NOMA-enabled multi-user VLC system
with Npairs of users, the achievable sum rate of the system
can be calculated by
R=
N
X
i=1 e
Ri,f +e
Ri,n.(13)
Considering the fact that each user in the multi-user VLC
system normally has its fixed QoS requirement during a period
of time, the achievable sum rate Rgiven in (13) can be viewed
as a fixed value. Hence, the EE maximization problem can be
transformed into a power minimization problem.
Let Pi={pi,f , pi,n}denote the power allocation set for
the far and near users in the i-th user pair. To obtain a QoS-
guaranteed optimal power allocation strategy, i.e., optimal Pi
with i= 1,2,·· · , N , the power minimization problem of the
multi-user VLC system can be formulated as
min
{P1,··· ,PN}Pelec
s.t. a: (10)
b: Pelec ≤Pmax.
(14)
In (14), constraint “a” is used to guarantee the power require-
ments of all the users so as to meet their QoS requirements
and constraint “b” is that the total electrical transmit power of
the LED should not exceed its maximum value Pmax.
By observing (10), we can find that the optimal solution
for the power minimization problem is that the far and near
users in the i-th user pair are allocated with minimum powers
to satisfy their QoS requirements. As a result, the closed-
form optimal power allocation set Popt
i={popt
i,f , popt
i,n}can be
obtained by
popt
i,f = 22
e
Ri,f 22
e
Ri,n Pz
h2
i,n
+Pz
h2
i,f !
popt
i,n = 22
e
Ri,n Pz
h2
i,n
.(15)
Using (15), the minimum total electrical transmit power of
all the Npairs of users is given by
Pelec,min =
N
X
i=1
22
e
Ri,f 22
e
Ri,n Pz
h2
i,n
+Pz
h2
i,f !+ 22
e
Ri,n Pz
h2
i,n
.
(16)
Substituting (13) and (16) into (17) yields the EE of the multi-
user VLC system using the proposed NOMA:
ηEE =PN
i=1 e
Ri,f +e
Ri,n
PN
i=1 22
e
Ri,f 22
e
Ri,n Pz
h2
i,n +Pz
h2
i,f + 22
e
Ri,n Pz
h2
i,n
.(17)
Based on the above analysis, we can find that the proposed
NOMA can be considered as a special case of NOMA in which
the total electrical transmit power is minimized by adopting the
QoS-guaranteed optimal power allocation strategy. It should
also be noticed that the QoS-guaranteed optimal power allo-
cation strategy is obtained based on the assumption that all the
2Nusers are divided into Nuser pairs. Therefore, efficient
user pairing should be performed before executing the QoS-
guaranteed optimal power allocation strategy in pairs.
3) Channel-Based User Pairing: By selecting a pair of
users to perform the proposed NOMA, the computational
complexity of the multi-user VLC system can be substantially
reduced, which results in a hybrid multiple access scheme
consisting of both NOMA and OMA techniques. More specifi-
cally, the proposed NOMA is adopted for the two users within
each user pair, while OMA is applied for different user pairs.
In this work, channel-based user pairing is adopted to divide
all the users into pairs [14], [21]. The key to implement
channel-based user pairing is to pair the two users which have
more distinctive channel conditions. For channel-based user
pairing, all the 2Nusers can be sorted based on their channel
gains in the ascending order:
h1≤ ·· · ≤ hk≤ ··· ≤ h2N,(18)
where hkis given in (1). After that, the sorted 2Nusers are
divided into two groups: the first group G1contains the first
half of the sorted users starting from user 1 to user N; the
second group G2consists of the second half starting from user
N+ 1 to user 2N. Based on the obtained two user groups,
user pairing can be performed in the following manner: Ui=
{G1(i), G2(i)}, i.e., the i-th user pair Uicontains both the
i-th user in G1and the i-th user in G2with i= 1,2,·· · , N .
IV. ANA LYTI CA L AN D SIMULATION RES ULT S
In this section, Monte Carlo simulations are performed to
substantiate the analytical expressions derived above, and the
EE of a typical multi-user VLC system using different multiple
access techniques are evaluated and compared. Specifically,
the following three multiple access techniques are considered:
(i) the proposed NOMA, (ii) NOMA with GRPA [13] and
(iii) OMA. The EE values adopting NOMA with GRPA and
OMA are calculated based on the condition that all the QoS
TABLE I
SIMULATION PARAMETERS
Parameter name Value
Vertical separation 2.5 m
Maximum horizontal separation 3 m
LED semi-angle 70◦
Maximum LED transmit power 1 W
PD responsivity 0.4 A/W
PD active area 1 cm2
PD half-angle FOV 70◦
Optical filter gain 0.9
Optical lens refractive index 1.5
Signal bandwidth 20 MHz
Noise PSD 10−22 A2/Hz
requirements of all the users are satisfied. The key simulation
parameters are summarized in Table I. The vertical separation
between the LED and the receiving plane is 2.5 m. The
maximum horizontal separation between the LED and the
users is 3 m. The LED semi-angle is 70◦and the maximum
LED transmit power is 1 W. The responsivity and the active
area of PD are 0.4 A/W and 1 cm2, respectively. The half-
angle FOV of PD is 70◦. The optical filter gain is 0.9. The
refractive index of optical lens is 1.5. The signal bandwidth
and noise PSD are 20 MHz and 10−22 A2/Hz, respectively.
A. Two-User Case
We first investigate the EE performance of a two-user VLC
system using different multiple access techniques. Fig. 3 shows
the EE versus the horizontal separation of two users with e
Rf=
e
Rn= 1 bit/s/Hz, where the near user is located under the LED
and the horizontal separation between the near user and the
far user is ranging from 0.5 to 3 m. It can be observed that the
simulation results agree well with the analytical results, which
can verify the fidelity of our analytical derivations obtained in
Section III.
For the two-user VLC system using OMA, the EE is grad-
ually reduced with the increase of the horizontal separation.
However, when NOMA with GRPA is adopted, the EE first
increases and then decreases with the increase of the horizontal
separation. A maximum EE of 408.1 bits/J/Hz is obtained at
the horizontal separation of 1 m. As can be seen from Fig. 3,
NOMA with GRPA obtains a lower EE than OMA at the
horizontal separation of 0.5 m, which suggests that the advan-
tage of NOMA cannot be fully exploited if two users have
similar channel conditions. Compared with OMA and NOMA
with GRPA, the proposed NOMA always achieves the highest
EE. More specifically, the proposed NOMA significantly out-
performs NOMA with GRPA when the horizontal separation
is relatively small, and the EE improvement becomes less
significant when the horizontal separation is larger than 2 m.
B. Multi-User Case
In the next, we evaluate and compare the EE performance
of the VLC system with multiple users using different mul-
tiple access techniques. In this multi-user case, the users are
randomly distributed within the receiving plane and the QoS
Fig. 3. EE vs. horizontal separation of two users with
e
Rf=
e
Rn= 1 bit/s/Hz.
requirement of each user is randomly selected from a given
QoS set e
Rwhere the unit of its elements is bits/s/Hz. In order
to obtain a stable EE value under random user locations and
QoS requirements, the average EE value is adopted over 10000
independent trials.
Figs. 4 and 5 show the average EE versus the number of
users with the QoS sets e
R= 1 and e
R={1, 2}, respectively.
As we can see, the simulation results are consistent with the
analytical results. For e
R= 1, the average EE of the multi-user
VLC system using OMA gradually decreases when more users
are served in the system. The same trends can be found when
using NOMA with GRPA and the proposed NOMA. It can
also be seen that NOMA with GRPA achieves substantially
improved average EE than OMA, which demonstrates the
superiority of NOMA against OMA. Nevertheless, when the
proposed NOMA with QoS-guaranteed optimal power alloca-
tion is applied, the average EE can be further enhanced. When
the number of users is 10, the average EEs using NOMA
with GRPA and the proposed NOMA are about 190 and 209
bits/J/Hz, suggesting a 10% improvement of EE. Moreover, for
e
R={1, 2}, the average EEs using NOMA with GRPA and the
proposed NOMA for ten users are about 74 and 88 bits/J/Hz,
which indicates a EE improvement of 19%. Hence, the EE
improvement by using the proposed NOMA in comparison to
NOMA with GRPA becomes much more significant when the
users have more diverse QoS requirements.
V. CONCLUSION
In this paper, we have proposed and investigated a power
domain energy-efficient NOMA technique for multi-user VLC
systems. In the proposed NOMA, the QoS requirements of
different users are guaranteed by allocating them with proper
powers, by adopting a QoS-guaranteed optimal power alloca-
tion strategy to minimize the total electrical transmit power and
hence maximize the EE of the multi-user VLC system. The
obtained Monte Carlo simulation results successfully verify
Fig. 4. Average EE vs. number of users with QoS set
e
R= 1.
Fig. 5. Average EE vs. number of users with QoS set
e
R={1, 2}.
the fidelity of the analytical derivations. Our results show that
the EE of the multi-user VLC system can be greatly improved
by using the proposed NOMA in comparison to NOMA with
GRPA, especially when diverse QoS are required by the users.
REFERENCES
[1] J. G. Andrews, S. Buzzi, W. Choi, S. V. Hanly, A. Lozano, A. C. Soong,
and J. C. Zhang, “What will 5G be?” IEEE J. Sel. Areas Commun.,
vol. 32, no. 6, pp. 1065–1082, Jun. 2014.
[2] L. Grobe, A. Paraskevopoulos, J. Hilt, D. Schulz, F. Lassak, F. Hartlieb,
C. Kottke, V. Jungnickel, and K.-D. Langer, “High-speed visible light
communication systems,” IEEE Commun. Mag., vol. 51, no. 12, pp.
60–66, Dec. 2013.
[3] T. Komine and M. Nakagawa, “Fundamental analysis for visible-light
communication system using LED lights,” IEEE Trans. Consum. Elec-
tron., vol. 50, no. 1, pp. 100–107, Feb. 2004.
[4] H. Haas, “Visible light communication,” in Proc. Opt. Fiber Commun.
Conf. (OFC), Mar. 2015, paper Tu2G.5.
[5] H. Haas, L. Yin, Y. Wang, and C. Chen, “What is LiFi?” J. Lightw.
Technol., vol. 34, no. 6, pp. 1533–1544, Mar. 2016.
[6] C. Chen, W.-D. Zhong, H. Yang, S. Zhang, and P. Du, “Reduction of
SINR fluctuation in indoor multi-cell VLC systems using optimized
angle diversity receiver,” J. Lightw. Technol., vol. 36, no. 17, pp. 3603–
3610, Sep. 2018.
[7] J.-Y. Sung, C.-H. Yeh, C.-W. Chow, W.-F. Lin, and Y. Liu, “Orthogonal
frequency-division multiplexing access (OFDMA) based wireless visible
light communication (VLC) system,” Opt. Commun., vol. 355, pp. 261–
268, Nov. 2015.
[8] M. Hammouda, A. M. Vegni, H. Haas, and J. Peissig, “Resource allo-
cation and interference management in OFDMA-based VLC networks,”
Phys. Commun., vol. 31, pp. 169–180, Dec. 2018.
[9] S.-M. Kim, M.-W. Baek, and S. H. Nahm, “Visible light communication
using TDMA optical beamforming,” EURASIP Journal on Wireless
Communications and Networking, vol. 2017, no. 1, p. 56, Mar. 2017.
[10] A. M. Abdelhady, O. Amin, A. Chaaban, B. Shihada, and M.-S.
Alouini, “Downlink resource allocation for dynamic TDMA-based VLC
systems,” IEEE Trans. Wireless Commun., vol. 18, no. 1, pp. 108–120,
Jan. 2019.
[11] Z. Ding, X. Lei, G. K. Karagiannidis, R. Schober, J. Yuan, and
V. K. Bhargava, “A survey on non-orthogonal multiple access for 5G
networks: Research challenges and future trends,” IEEE J. Sel. Areas
Commun., vol. 35, no. 10, pp. 2181–2195, Oct. 2017.
[12] Y. Liu, Z. Qin, M. Elkashlan, Z. Ding, A. Nallanathan, and L. Hanzo,
“Nonorthogonal Multiple Access for 5G and Beyond,” Proc. of the IEEE,
vol. 105, no. 12, pp. 2347–2381, Dec. 2017.
[13] H. Marshoud, V. M. Kapinas, G. K. Karagiannidis, and S. Muhaidat,
“Non-orthogonal multiple access for visible light communications,”
IEEE Photon. Technol. Lett, vol. 28, no. 1, pp. 51–54, Jan. 2016.
[14] L. Yin, W. O. Popoola, X. Wu, and H. Haas, “Performance evaluation of
non-orthogonal multiple access in visible light communication,” IEEE
Trans. Commun., vol. 64, no. 12, pp. 5162–5175, Dec. 2016.
[15] X. Zhang, Q. Gao, C. Gong, and Z. Xu, “User grouping and power
allocation for NOMA visible light communication multi-cell networks,”
IEEE Commun. Lett., vol. 21, no. 4, pp. 777–780, Apr. 2017.
[16] H. Marshoud, P. C. Sofotasios, S. Muhaidat, G. K. Karagiannidis,
and B. S. Sharif, “On the performance of visible light communication
systems with non-orthogonal multiple access,” IEEE Trans. Wireless
Commun., vol. 16, no. 10, pp. 6350–6364, Oct. 2017.
[17] C. Chen, W.-D. Zhong, H. Yang, and P. Du, “On the performance
of MIMO-NOMA-based visible light communication systems,” IEEE
Photon. Technol. Lett., vol. 30, no. 4, pp. 307–310, Feb. 2018.
[18] C. Chen, W.-D. Zhong, H. Yang, P. Du, and Y. Yang, “Flexible-rate
SIC-free NOMA for downlink VLC based on constellation partitioning
coding,” IEEE Wireless Commun. Lett., vol. 8, no. 2, pp. 568–571, Apr.
2019.
[19] Y. Yapıcı and I. G¨
uvenc¸, “NOMA for VLC downlink transmission with
random receiver orientation,” IEEE Trans. Commun., vol. 67, no. 8, pp.
5558–5573, Aug. 2019.
[20] L. Zeng, D. C. O’Brien, H. Le Minh, G. E. Faulkner, K. Lee, D. Jung,
Y. Oh, and E. T. Won, “High data rate multiple input multiple output
(MIMO) optical wireless communications using white LED lighting,”
IEEE J. Sel. Areas Commun., vol. 27, no. 9, Dec. 2009.
[21] M. B. Janjua, D. B. da Costa, and H. Arslan, “User pairing and
power allocation strategies for 3D VLC-NOMA systems,” IEEE Wireless
Commun. Lett., 2020.
