Joint dynamic user pairing and resource allocation for RIS-aided multi-cluster NOMA networks

The integration of reconfigurable intelligent surfaces (RIS) with non-orthogonal multiple access (NOMA) has emerged as a transformative approach to enhance the performance of next-generation wireless networks. However, managing energy efficiency (EE) and addressing the challenges posed by dynamic network environments remain critical issues. In this paper, we propose a novel joint dynamic user pairing and resource allocation framework tailored for RIS-aided multi-cluster NOMA networks. Specifically, we consider scenarios where users experience varying degrees of obstruction, categorizing them as having either direct or non-direct links to the base station (BS). To mitigate interference and enhance system performance, a dynamic user pairing algorithm is developed to cluster users with distinct channel conditions based on channel gain difference and correlation. Building on this, an EE maximization problem is formulated, involving the joint optimization of the beamforming matrix, power allocation factors, and RIS phase shifts. The inherent non-convexity and coupling of the problem are addressed using a solution framework that combines auxiliary variables, successive convex approximation (SCA), and alternating optimization (AO) to convert the problem into a tractable convex form. Simulation results demonstrate the superior performance of the proposed algorithms in improving energy efficiency and mitigating interference compared to existing benchmarks.