On Energy Efficiency Optimization in Downlink MIMO-NOMA

Non-orthogonal multiple access (NOMA) is a promising technique for the fifth generation mobile communication system due to its high spectral efficiency (SE). Previous works focus on user scheduling or/and power allocation to achieve the maximization of sum rate. However, the energy efficiency (EE) has not been well studied for NOMA systems. In this paper, we aim to optimize user scheduling and power allocation to maximize the EE of the downlink multiple-input multiple-output (MIMO) NOMA system. There are two scheduling schemes with low complexity and signaling overhead. In the signal space alignment (SSA) based user scheduling scheme, we use the SSA with reference signal space to schedule the users. In the maxcollinearity user scheduling approach, we schedule the users considering maximum collinearity among users. Furthermore, we use the sequential convex approximation (SCA) to transform the original optimal problem into a convex optimization problem, so that an optimal power allocation strategy can be obtained. The results show that our algorithms can achieve higher EE when the transmitting power is relatively low. Besides, the sum rate of the center users is higher using the second scheduling approach.