Multi-STAR-RIS-Assisted Multi-User Communication With Symbiotic RIS-Feedback Channels and BS-User Direct Links

Modern urban communication networks face challenges such as signal propagation blockages, which can be effectively mitigated using simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS). This paper explores the integration of multiple STAR-RISs into a symbiotic radio framework to enhance overall network performance, supporting both primary users and STAR-RISs that transmit their own data over the shared spectrum. The system enables RIS-feedback links within the same spectrum while accounting for direct links between the base station (BS) and user equipments (UEs). Two optimization problems are formulated based on different symbiotic resource-sharing policies and are equivalently transformed into weighted minimum mean square error (WMMSE) problems, which are then solved using the block coordinate descent method. Furthermore, an iterative penalty dual decomposition is adopted to determine the optimal BS beamformer and the STAR-RIS transmitting and reflecting coefficients. The effectiveness of the proposed STAR-RIS optimization framework is demonstrated through in-depth evaluation results. The findings indicate that deploying multiple STAR-RISs significantly enhances overall communication performance. While direct BS-to-UE links enhance user performance, they negatively impact RIS feedback performance. Moreover, performance trends vary according to the selected symbiotic resource-sharing policy.