Joint User Clustering and Passive Beamforming for Downlink NOMA System with Reconfigurable Intelligent Surface

Reconfigurable intelligent surface (RIS) is an emerging technology to achieve energy-efficient wireless communication. This technology has the potential of turning the wireless environment, which is highly probabilistic in nature, into a programmable and partially deterministic space. This paper focuses on the joint user clustering, passive beamforming and power allocation for the downlink RIS-assisted nonorthogonal-multiple-access (NOMA) system, with the target of maximizing energy efficiency. This is an optimization problem which is solved by optimizing three sub-problems iteratively. In particular, user clustering sub-problem is solved with a matching algorithm, power allocation sub-problem is solved with the difference of two convex functions (DC) programming, and passive beamforming sub-problem is solved with univariate search technique. Simulation results demonstrate that the downlink RIS-assisted NOMA system can improve the energy efficiency by 7.8%-23.1%, compared with NOMA system without RIS and traditional orthogonal-multiple-access (OMA) system without RIS.