Enhanced RIS-Assisted DOA Estimation Through Coding Matrix Design

This paper proposes an enhanced direction of arrival (DOA) estimation method by optimizing the coding matrix of reconfigurable intelligent surfaces (RIS). Exploiting the sparsity in the spatial domain, the DOA estimation problem can be reformulated within the compressive sensing (CS) framework. According to CS theory, measurement matrix using random coding often exhibit high mutual coherence, leading to degraded DOA estimation performance. To address this issue, we introduce a coding matrix design strategy based on Genetic Algorithm (GA) to optimize the measurement matrix. Simulation results validate that the proposed method is significantly superior to the traditional random coding method, achieving greater efficiency, finer estimation resolution, and superior noise robustness.