SPECTRUM AND ENERGY EFFICIENCY OPTIMIZATION FOR SWIPT-BASED HYBRID PRECODING-BASED MMWAVE MASSIVE MIMO-NOMA SYSTEMS

Non-orthogonal multiple access (NOMA) and millimeter-wave (mmWave) massive MIMO systems are pivotal for enhancing spectrum efficiency in 5G networks while integrating simultaneous wireless information and power transfer (SWIPT) offers a pathway to improve energy efficiency. This study addresses the joint optimization of transceiver design and power splitting in SWIPT-enabled hybrid precoding for mmWave massive MIMO-NOMA systems. Three main techniques were proposed: (1) a Dynamic Cluster-Head Selection Algorithm to assign beam-specific users, (2) a Compressed Sensing-based Committee Machine Design (CS-COMADE) to suppress inter-user interference and optimize antenna array gain, and (3) an iterative framework to resolve the non-convex problem of jointly optimizing power allocation and power-splitting factors. Compared to conventional SWIPT-based MIMO-OMA and existing hybrid precoding schemes, the proposed approach achieved superior spectrum and energy efficiency, nearing the performance of fully digital zero-forcing (ZF) precoding in energy efficiency. These advancements highlight the potential of SWIPT-NOMA integration to support sustainable, high-capacity 5G networks.