Dynamic RIS Partitioning in Downlink NOMA Systems based on Quality of Service Optimization

The integration of reconfigurable intelligent surfaces (RIS) with non-orthogonal multiple access (NOMA) is a promising solution for improving connectivity and spectral efficiency in next-generation wireless networks. However, joint optimization of all RIS elements could be computationally challenging. This letter proposes a quality of service (QoS)-driven RIS partitioning for downlink RIS-assisted NOMA systems. Herein, the RIS is dynamically divided into sub-surfaces assigned to different user subsets. User requirements are captured through utility functions, and the resulting constrained optimization problem is solved numerically to maximize system performance. Numerical results confirm the effectiveness of the proposed RIS partitioning strategy.