A Novel RIS-Aided Optimization Strategy for Semantic Communication System

Reconfigurable intelligent surfaces (RISs) are programmable metasurfaces capable of optimizing signal strength and reducing interference, serving as key components in maintaining the integrity of semantic information during transmission. This letter explores the establishment of additional semantic transmission and reflection pathways by deploying RISs in different cells. An optimization strategy, maximizing mutual information (MI) for quality of experience (QoE)-aware modeling of the R-SC system (QR-SC), is proposed to enhance both semantic and communication performance. Additionally, a QoE-aware model is utilized for users to gauge semantic transmission performance. Experimental results indicate that QR-SC can elevate the performance of semantic communication while ensuring reliable transmission, highlighting the substantial potential of RIS in digital and energy simultaneous transmission.