Location Prediction Empowered RIS Phase Optimization for ISAC System With High-Mobility

Reconfigurable Intelligent Surface (RIS) is a key technology for 6G communications. However, in high-mobility scenarios, traditional approaches based on statistical location information respond slowly, while those relying on instantaneous CSI are impractical due to excessive pilot overhead. To this end, we propose a location prediction empowerd RIS phase optimization (LP-RISPO) approach. Its core idea is to shift from passive “static optimization” to proactive “dynamic predictive optimization”. Specifically, it predicts the future motion state of user equipment (UE) based on real-time position, velocity, and uncertainty (covariance), and uses this prediction as prior knowledge to perform a one-time optimization of the RIS phase configuration. As a result, LP-RISPO significantly reduces pilot transmission and RIS phase reconfiguration frequency in high-mobility scenarios while improving the system’s average achievable rate, compared to benchmarks.