Rethinking Power Minimization in a Downlink Hybrid NOMA Network

This paper investigates the problem of minimizing total transmission power in a downlink multi-user hybrid non-orthogonal multiple access (H-NOMA) network. The system consists of an access point (AP) equipped with multiple antennas serving multiple single-antenna users, divided into clusters. A closed-form beamforming solution is derived to achieve near optimal performance for the case where each cluster contains two users. The proposed solution significantly reduces reducing computational complexity compared to existing semidefinite programming (SDP) approaches. Based on the derived beamforming solution, an efficient grouping strategy is proposed to determine the optimal clustering plan for users. Simulation results show that the proposed approach outperforms conventional methods, such as orthogonal multiple access (OMA), zero-forcing (ZF), and other clustering benchmarks. It also has competitive performance compared to dirty paper coding (DPC), in terms of total transmission power reduction.