Joint Beamforming and Power Allocation for RIS-assisted OTFS-NOMA Systems

Orthogonal time frequency space (OTFS) has emerged as a promising physical layer technology for enhancing communication reliability in high-mobility scenarios. This paper focuses on the investigation of reconfigurable intelligent surface (RIS)-aided multiple-input multiple-output (MIMO)-OTFS non-orthogonal multiple access (NOMA) systems, where RIS is employed to assist in enhancing communication, and NOMA is utilized to improve the spectral efficiency of the system. To maximize the achievable system rate, we formulate a joint optimization problem involving active beamforming vectors, power allocation coefficients, and RIS phase shifts. To address this non-convex problem with highly coupled variables, we adopt an alternating optimization algorithm for decoupling. Specifically, for the subproblem of jointly optimizing active beamforming and power allocation, semidefinite relaxation (SDR) and successive convex approximation (SCA) techniques are employed. In particular, we prove that the optimal solution to the SDR problem is a rank one solution. Furthermore, for the RIS phase-shift optimization subproblem, the sequential rank-one constraint relaxation (SROCR) algorithm is adopted instead of Gaussian randomization, thereby avoiding potential non-convergence issues caused by the solution quality of Gaussian randomization. Finally, simulation results verify the advantages of the RIS-aided OTFS-NOMA system, the effectiveness of the proposed algorithm, and the impact of key system parameters on system performance.