부산대학교 도서관

The massive Multiple-input Multiple-output (MIMO) and Non-Orthogonal Multiple Access (NOMA) are considered as two key techniques for the coming 5G wireless communications. Through beam selection, the massive MIMO systems can reduce the number of required radio-frequency (RF) chains while without distinct performance degradation, which can enhance energy efficiency of the systems. In order to further improve energy efficiency, in this paper, we propose an energy efficient power allocation design for beamspace based multi-user multiple-input single-output (MISO) NOMA systems. First, according to the beamspace channel state information (CSI) matrix obtained by beam selection, we get the precoding matrix through zero-forcing (ZF) method. Then we formulate the energy efficiency maximization (EEmax) optimization problem with power allocation considerations. Through sequential convex approximation (SCA) and second-order cone programming (SOCP), the original non-convex problem can be reformulated as a convex optimization problem. Finally, we choose iterative optimization algorithm to solve the reformulated problem. Numerical results show that the proposed power allocation design is effective in achieving better energy efficiency comparing with the conventional methods.