Sum Rate Maximization for STAR-RIS Aided NOMA System With Two-Way Communication

Simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) has received substantial attentions due to its capability of manipulating incident signals. This letter investigates two-way communication for STAR-RIS-assisted non-orthogonal multiple access (NOMA) system with frequency division duplex (FDD). Specifically, a sum rate maximization problem is decomposed into three subproblems for STAR-RIS-aided NOMA system with mode switch (MS) and energy splitting (ES), i.e., the uplink power allocation subproblem, the downlink power allocation subproblem, and the passive beamforming optimization subproblem. The uplink power allocation is solved by transforming it to a linear program (LP) problem. Furthermore, a successive convex approximation (SCA) method is invoked to solve the downlink power allocation subproblem, and a penalty based semi-definite programming (SDP) method is developed to achieve the optimal passive beamforming matrix. Simulation results show that the proposed STAR-RIS-aided NOMA method has the capability of enhancing the sum rate compared with various benchmarking schemes.