Review articleJoint user grouping and power allocation in VLC-NOMA system
Introduction
With the rapid development of internet of things, the traditional radio frequency (RF) networks cannot meet the increasing demand of wireless data with the limit of spectrum resources [1]. Recently, optical wireless communication has been presented as a promising solution due to higher optical spectrum and larger bandwidth [2]. Meanwhile, due to large scale deployment of the white light emitting diode (LED) as the next generation green lighting, the visible light communication (VLC), which utilizes the light wave of LED to transfer information in wireless light channel, has been presented as a complementary solution for fifth generation (5G) mobile networks [3].
In VLC systems, the wireless access technologies often used the orthogonal multiple access (OMA) technology [4], such as orthogonal frequency division multiple access (OFDMA), in which one channel can only be allocated one user. However, as number of users grew rapidly, OMA technology cannot support large scale user terminals access. In contrast, NOMA is considered as a promising technical due to superior spectrum efficiency and massive access. Unlike the traditional OMA, NOMA can access more users by exploiting the power multiplexing, in which different users can share time domain and frequency domain resources to improve the spectral efficiency of system and the fairness between users [5]. NOMA has been proved to be very suitable for VLC systems, because the channel state information (CSI) can be estimated more accurately and the signal to noise ratio (SNR) is high in VLC network compared with RF [6], [7]. The combination of NOMA and VLC was firstly proposed by authors in [8]. From then on, the implementation of NOMA in indoor VLC networks has got more and more attention. VLC-NOMA network has better bit error rate performance. In addition, compared with the OMA, NOMA has better user fairness and sum rate performance [9].
On the one hand, user grouping is one of the important issues in VLC-NOMA network. When all users share a subcarrier, it will lead to high computational complexity, decoding delay and bit error rate. It is unrealistic to put all users in a NOMA group. The optimal solution can be found by the exhaustive search (ES) method in which all possible arrange combinations of user grouping can be considered. However, as the number of users increases, the time complexity of ES will increase exponentially. In order to solve this problem, user grouping algorithms are proposed [10], [11], [12], [13]. Authors in [10] proposed a user grouping algorithm based on bipartite matching for heterogeneous VLC-RF relay networks. The Kuhn–Munkras (KM) algorithm is used to find the optimal solution of user grouping for minimizing outage probability. Authors in [11] proposed an optimal user pairing algorithm aiming at maximizing system throughput, and a closed-form global optimal solution in general NOMA networks is derived by deduction method. The above two algorithms are only suitable for even number of users which is not practical. In order to overcome this shortcoming, authors in [12] proposed two user grouping schemes based on [11], which one of the users occupies one subchannel or one of the groups contains three users. In addition, a dynamic user grouping algorithm based on genetic algorithm was proposed in [13], which cannot only solve the problem that the number of users is odd, but also improve the throughput of the system. However, the genetic algorithm is easy to fall into local optimal.
On the other hand, power allocation is also one of the key problems in VLC-NOMA network. There are various power allocation algorithms in VLC-NOMA networks. Fixed power allocation is the simplest power allocation algorithm, in which the power allocation coefficients for each user are fixed. Authors in [14] introduced the gain ratio power allocation (GRPA) algorithm in VLC networks to enhance system throughput and user fairness, in which the power allocation coefficients for each user are calculated according to the CSI of each user, and it is the most popular power allocation algorithm. Authors in [15] proposed a new GRPA algorithm based on the characteristic of indoor VLC channels to enhance system throughput. Authors in [16] proposed a joint PLC-VLC power allocation algorithm, in which the power line communication (PLC) link was the backhaul of VLC networks. Authors in [12] considered a more practical 3D indoor VLC-NOMA networks in which the height of users were unequal. Therefore, more accurate result of user grouping and power allocation algorithm can be obtained. Authors in [17] proposed an optimal multi-factor control power allocation algorithm which considered residual interference generated by the successive interference cancellation (SIC) at the receiver.
To improve user fairness, the relevant power allocation algorithms were proposed in [18], [19], [20]. Authors in [18] considered a multi-carrier multi-user downlink scenario, and proposed a new subcarrier and power allocation algorithm based on genetic algorithm to improve system throughput and user fairness. Authors in [19] designed the optimal power allocation algorithm which maximized sum rate and took user fairness as a constraint. Authors in [20] proposed an optimal power allocation algorithm based on differential evolution which achieved a flexible tradeoff between sum rate and user fairness depending on subjective setting of user. To reduce computational complexity, authors in [21] proposed a simplified gain ratio power allocation (S-GRPA) scheme. In the S-GRPA scheme, the CSI used for power allocation is not complete and obtained through look up table method rather than calculation, which reduces the complexity of the power allocation scheme.
The above papers considered the influences of user grouping and power allocation for VLC-NOMA networks, but optimization goals of these papers mainly focused on sum rate or user fairness, ignoring the energy consumption of the system. With the rapid growth of wireless data traffic demand, the energy consumption of wireless networks is growing rapidly, and has been regarded as one of the important indicators in 5G green communication [22].
In order to realize 5G green communication and reduce energy consumption, in this paper, we focus on improvement of system energy efficiency where energy efficiency problem is formulated as a mixed integer nonlinear programming problem (MINLP). It is challenging to find the optimal solution. In order to solve this problem, joint user grouping and power allocation (JUGPA) are proposed aiming at maximizing energy efficiency, including dynamic user grouping and power allocation algorithm. Firstly, we divide users into two sets by the average of channel gains among all users, and then the number of NOMA groups can be obtained. Secondly, for a given NOMA group, we derive the closed-form optimal power allocation expression that maximizes the energy efficiency by Dinkelbach method [23] and Lagrange dual decomposition method. And then, due to the interference-free between different groups, the system energy efficiency can be obtained.
The remainder of this paper is organized as follows. The VLC-NOMA system model is introduced in Section 2. JUGPA algorithm is proposed in Section 3. Then, the simulation results are discussed in Section 4. Finally, conclusions of the paper are narrated in Section 5.
Section snippets
System model and problem formulation
Fig. 1 shows the VLC-NOMA networks model of downlink communication. A LED is located on the ceiling and the communication area of the LED can be considered as a circle according to the Lambert radiation pattern. Assuming N user equipment (UE) are uniformly distributed within the communication area, and are divided into K groups, in which each group uses a solely subcarrier whose bandwidth is B. Since different groups uses different subcarrier, UEs belonging to different groups are
Joint user grouping and power allocation
Problem (8) is a MINLP problem due to C3, which is a challenge to obtain the global optimal solution. In order to solve this problem effectively, we proposed a JUGPA, including user grouping and power allocation. In the user grouping phase, dynamic user grouping algorithm (DUGA) is designed which make full use of the difference of the CSI. Then, based on results of user grouping obtained from DUGA, optimal power allocation based on energy efficiency algorithm (OPAEEA) is designed to maximize
Simulation setup
In this paper, Monte Carlo method is used to simulate JUGPA, the detailed simulation parameters are shown in Table 1.
In order to verify the validity of the JUGPA algorithm, the simulation environment is set as shown in Fig. 1 and Table 1. As comparative algorithm, besides ES algorithm, there are the conventional user grouping algorithm in [10] and random user pairing algorithm, denoted by CUG and RUP respectively. Furthermore, in order to show performance of JUGPA algorithm, four multiple
Conclusions
In this paper, we aim to maximize the system energy efficiency in VLC-NOMA networks. In order to solve energy efficiency optimization problem, JUGPA are proposed, including DUGA and OPAEE. Firstly, we utilize difference of CSI to realize DUGA algorithm with low complexity. Then, we derive a closed-form power allocation expression via Dinkelbach method and Lagrange dual decomposition method, and the system energy efficiency can be obtained. Through extensive simulations, it was shown that the
CRediT authorship contribution statement
Huanlin Liu: Conceptualization, Methodology. Bingchuan Huang: Visualization, Investigation. Jian Yang: Data curation, Writing – original draft, Writing – review & editing. Yong Chen: Supervision. Tong Zhang: Software, Validation. Shuai Yang: Formal analysis. Ke Chen: Software.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This research was supported by the National Natural Science Foundation of China (51977021, 61275077), the Science Foundation Project of Chongqing Science and Technology Commission, China (CSTC2019 jcyj-msxmX0613, CSTC2020 jcyj-msxmX0682), China, and supported by the science and technology research program of Chongqing municipal education commission, China (Grant No. KJQN201900635). We are thankful to the anonymous reviewers for their valuable comments and suggestions on this paper.
Huanlin Liu received her Ph. D. degree in 2008 from the Chongqing University and the M.S. degree in 2000 from Chongqing University of Posts and Telecommunications, China. Currently he serves as a professor in the School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications. Her research interests include optical network optimization and scheduling algorithm research, all-optical signal information acquiring and processing.
References (27)
- et al.
Service-aware 6G: an intelligent and open network based on the convergence of communication, computing and caching
Digit. Commun. Netw.
(2020) - et al.
Energy and spectral efficiencies trade-off in MIMO-noma system under user-rate fairness and variable user per cluster
Phys. Commun.
(2021) - et al.
A NOMA scheme for visible light communications with single carrier transmission and frequency-domain successive interference cancellation
Optik
(2019) - et al.
A low complexity power allocation scheme for NOMA-based indoor VLC systems
Opt. Commun.
(2020) - et al.
Robust resource allocation for NOMA-assisted digit
Commun. Netw. Heterog. Netw. Digit. Commun. Netw.
(2022) Applications and challenges of intelligent data fusion
J. Chongqing Univ. Post Telecommun.
(2021)- et al.
Non-orthogonal multiple access for hybrid VLC-RF networks with imperfect channel state information
IEEE Trans. Veh. Technol.
(2021) - et al.
New kind of visible light-RFID with ACO-OFDM modulation
J. Chongqing Univ. Post Telecommun.
(2019) - et al.
Green indoor optical wireless communication systems: Pathway towards pervasive deployment
Digit. Commun. Netw.
(2020) - et al.
Max–min fairness-based resource allocation algorithm for heterogeneous VLC/RF networks
J. Chongqing Univ. Post Telecommu.
(2022)
Performance evaluation of non-orthogonal multiple access in visible light communication
IEEE Trans. Commun.
A bipartite matching based user pairing scheme for hybrid VLC-RF NOMA systems
Wu, optimal user pairing for downlink non-orthogonal multiple access (NOMA)
IEEE Wirel. Commun. Lett.
Cited by (0)
Huanlin Liu received her Ph. D. degree in 2008 from the Chongqing University and the M.S. degree in 2000 from Chongqing University of Posts and Telecommunications, China. Currently he serves as a professor in the School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications. Her research interests include optical network optimization and scheduling algorithm research, all-optical signal information acquiring and processing.
Binchuan Huang is currently pursuing the M.E. degree with the Chongqing University of Posts and Telecommunications, China. Her research interests include visible light communication, resource allocation and access point selection for VLC network.
Jian Yang received the B.S. degree in 2019 from University of South China, Hengyang, China. His research focused on visible light communications and new generation wireless communication systems. He is currently working toward the M.S. degree in Chongqing University of Posts and Telecommunications, Chongqing, China.
Yong Chen received his M.S. degree in 1995 from Chongqing University, Chongqing, China, and the Ph.D. degrees in machine and electronic engineering from the Chongqing University, China, in 2003. Currently he serves as a professor in the School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications. His research interests are in automation control, lased communication, sensor detection and signal processing and Logistics Distribution Technology.
Tong Zhang is currently pursuing the M.E. degree with the Chongqing University of Posts and Telecommunications, China. Her research interests include visible light communication, resource allocation and access point selection for VLC network.
Shuai Yang is currently pursuing the M.E. degree with the Chongqing University of Posts and Telecommunications, China. His research interests include visible light communication, resource allocation for VLC network.
Ke Chen is currently pursuing the M.E. degree with the Chongqing University of Posts and Telecommunications, China. His research interests include visible light communication, resource allocation and access point selection for VLC network.