RIS-Aided MU-NOMA Systems with Imperfect CSI and Generalized Hardware Impairments

The combination of reconfigurable intelligent surface (RIS) and non-orthogonal multiple access (NOMA) is a promising technique to enhance spectral efficiency and energy efficiency. However, the imperfect channel state information (CSI) of the wireless channel and hardware impairments (HWI) of the radio frequency chain will cause significant performance loss on RIS-aided NOMA communication. In this paper, we focus on the effect of the imperfect CSI and generalized HWI on RIS-aided NOMA communication and propose an efficient robust precoding and phase shift design scheme to enhance the energy efficiency of systems. Specifically, we minimize the transmit power with quality of service constraints of users, where the imperfect CSI and generalized HWI are modeled as Gaussian error models, and the successful successive interference cancellation condition is characterized by the variance criterion. The minimization problem is greatly challenging due to the non-closed form outage constraints and non-explicit phase shift design. To solve it, we develop an efficient probabilistic technique to reformulate the outage constraints and restrict an explicit objective function to obtain a more efficient phase shift design. Numerical results demonstrate that our scheme has superior convergence performance, energy efficiency, and feasibility rate compared to existing schemes.