Efficient MIMO-NOMA Clustering Integrating Joint Beamforming and Power Allocation

Multiple-input, multiple-output non-orthogonal multiple access (MIMO-NOMA) approach has been considered as a promising multiple access technology for the fifth generation (5G) networks to improve the system capacity and the spectral efficiency. In this paper, we propose a linear optimization approach to jointly optimize beamforming vectors and power allocation coefficients for a MIMO-NOMA cluster, and then minimize the total power consumption through mobile user (MU) clustering. The proposed approach avoids the peer effect during MU clustering and can obtain a closed-form expression of the cluster beamforming matrix. Then, we employ our approach in two different MIMO-NOMA scenarios termed as MIMO-NOMA1 and MIMO-NOMA2. Based on the analytical results, we propose a ranking scheme to obtain the clustering result for the large-scale MIMO-NOMA networks. Simulation results show that MIMO-NOMA1 is better than MIMO-NOMA2, and the proposed approach is superior in finding power efficient MIMO-NOMA clusters than counterparts.