Energy Efficient Hybrid Precoding for SWIPT in mmWave Massive MIMO NOMA Systems

Abstract

The integration of simultaneous wireless information and power transfer (SWIPT) into millimeter-wave (mmWave) massive MIMO–non-orthogonal multiple access (NOMA) systems is a promising paradigm for sustainable, high-capacity wireless networks. However, achieving high energy efficiency (EE) in these systems is challenging due to severe propagation losses and the constraints of hybrid precoding. This paper proposes a novel hybrid analog–digital transceiver framework that jointly optimizes power allocation and power-splitting factors. The key contributions include a dynamic cluster-head selection (DCHS) algorithm for forming interference-resilient user groups and a compressed sensing–based committee machine design (CS-COMADE) to enhance array gain and mitigate interference. To solve the non-convex EE maximization problem, we develop a low-complexity iterative resource allocation algorithm. Simulation results demonstrate that our method achieves substantial gains, improving EE by up to 20% over conventional SWIPT-based MIMO–OMA and hybrid NOMA schemes. Notably, it attains near-optimal performance compared to fully digital zero-forcing precoding while reducing power consumption by over 60%. These findings confirm the framework’s viability as an energy-efficient solution for green 5G and 6G communication systems, effectively balancing high data rates with sustainable operation.