Design and Optimization of Specular Reflecting RIS-Aided VLCP System With Low SNR

This paper investigates the optimization of the integrated visible light communication and positioning (VLCP) system performance with the aid of specular reflecting reconfigurable intelligent surface (RIS) when the received signal is subject to low signal-to-noise ration (SNR). Specifically, we focus on the low SNR asymptotic capacity as the criterion of communication indicator whereas the Cramer-Rao low bound (CRLB) as the positioning performance indicator. The VLCP system performance in low SNR is discussed since the scenario may occur if the receiver encounters substantial noise interference (low illumination conditions such as corridors) or the transmitter imposes restrictive limitations on the intensity (such as biology darkroom lab). We derive the CRLB and asymptotic capacity of the RIS-aided multi-LED VLCP system in low SNR and dig out the relationship between the communication or positioning performance and the parameters of RIS. Two optimization problems are formulated with the maximization of the capacity as the optimization object while satisfying the constraints on the CRLB and RIS parameters. To address the non-convex optimization problems, we leverage convex relaxation, the monotonicity of the function, semidefinite relaxation (SDR), and greedy algorithms. Based on these techniques, we propose the integrated two-stage greedy algorithm and greedy strategy-based convex relaxation algorithm to determine the optimal configuration of RIS. From the simulation results, we show that the application of RIS can enhance the VLCP system performance in low SNR obviously, and meanwhile, the proposed algorithms are effective methods for the proposed system performance optimization problems.