Energy-Efficient Power Allocation for Millimeter Wave Beamspace MIMO-NOMA Systems

Penglu Liu, Yong Li, Wei Cheng, Wenjie Zhang, Xiang Gao.
2019 IEEE Access  
Massive multi-input multi-output (MIMO) is envisioned as a key technology for the emerging fifth generation of communication networks (5G). However, considering the energy consumption of the large number of radio frequency (RF) chains, massive MIMO poses a problem to energy efficiency (EE) requirement of 5G. In this paper, we propose an energy-efficient power allocation method for millimeter-wave (mmWave) beamspace MIMO non-orthogonal multiple access (NOMA) systems, where there may be multiple
more » ... sers in each selected beam. First, according to the beam selection (BS) results, we get the precoding matrix through zero-forcing (ZF) beamforming method. Second, we formulate the energy efficiency (EE) maximization optimization problem as a fractional programming. Through sequential convex approximation (SCA) and second-order cone (SOC) transformation, the original optimization problem can be transformed to a convex optimization problem. By using iterative optimization algorithm, we can get the power allocation results. Then, we analyze the convergence of our proposed iterative optimization method and get that the solution in each iteration is a suboptimal solution to the original non-convex optimization problem. Simulation results show that the proposed energy-efficient power allocation scheme has better EE performance comparing with the conventional methods when the transmitted power exceeds the power threshold. INDEX TERMS Beamspace, NOMA, energy efficiency, power allocation, convex optimization. I. INTRODUCTION With the rapid development of mobile Internet and Internet of things (IoT), there will be the prediction of 1000-fold data traffic increase by the year 2020 [1], [2] . Massive multi-input multi-output (MIMO) and non-orthogonal multiple access (NOMA) are two key techniques for the coming fifth generation of communication networks (5G), which work together for satisfying the future large demands of communication service [3]- [6] . Besides the spectral efficiency (SE), massive connectivity for IoT, lower latency and diverse compelling services, energy efficiency (EE) is another key performance indicator (KPI) of 5G, which is more than 100 times the EE of 4G. According to the propagation characteristics of millimeter wave (mmWave), the energy consumption of MIMO-NOMA systems can be reduced by relay [7] and network densification [8], [9] . However, the energy consumption of circuit, which is proportional to the number of radio frequency (RF) chains, can degrade the EE performance
doi:10.1109/access.2019.2935495 fatcat:ho77cw7hnbeatkazzgp4i2pcze