Robust Beamforming Design for NOMA System Assisted by STAR-RIS

STAR-RIS represents an innovative technology that can simultaneously transmit and reflect signals, thereby extending the coverage from half space to full space. In practical situations, hardware failures at the base station's transmission and reception ends inevitably lead to distorted noise and performance degradation. To address this issue, this study proposes a revolutionary non-orthogonal multiple access (NOMA) system aided by STAR-RIS and explores the robust beamforming design of communication systems in the context of transceiver hardware damage. This work focuses on the joint optimization of active beamforming, transmission and reflection beamforming, power allocation factors, and decoding order for the purpose of optimizing the achievable total rate. A location-based matching algorithm has been introduced to facilitate subchannel allocation, allowing reflection and transmission users to be grouped on designated subchannels while determining the decoding order through optimizing the total channel gain. Furthermore, active beamforming optimization is done using the successive convex approximation approach, while transmission and reflection beamforming are optimized using the convex upper bound approach. Finally, optimize the power distribution coefficients. Numerical simulations indicate that the STAR-RIS-NOMA system exhibits greater resilience to hardware failures compared to the traditional non-robust RIS-NOMA system, which neglects the impact of hardware damage.