A Robust RIS-Enabled Spectrum Sensing Algorithm for Cognitive Radio Networks

Cognitive radio networks (CRNs) represent a promising technology for efficiently utilizing spectrum resources in next-generation wireless communication systems. Cooperative spectrum sensing (CSS) plays a pivotal role in CRNs, involving multiple secondary users (SUs) to detect the primary users (PUs). However, in conventional CSS systems, the PU signals received by SUs can be too weak for reliable detection due to the uncontrollable wireless environment. Furthermore, energy detection relies on accurate knowledge of noise power and is therefore susceptible to noise uncertainty, which can significantly degrade performance. To address these issues, this paper proposes a robust spectrum sensing method with the assistance of the reconfigurable intelligent surface (RIS), where the RIS is employed to enhance the signal power gain at the SUs to improve sensing performance. Specifically, we introduce an energy ratio detection (ERD) algorithm that leverages energy ratios from any two distinct SUs. We derive closed-form expressions for the decision threshold and detection probability of the proposed ERD algorithm. Notably, our approach is robust to noise uncertainty as it does not require prior knowledge of noise power. Furthermore, we optimize the RIS phase shifts in closed-form by maximizing the total signal-to-noise ratio across all SUs, resulting in significant detection performance improvements. Our optimization framework considers both instantaneous channel state information (CSI) and statistical CSI. Simulation results validate the effectiveness of the proposed algorithm, confirming the theoretical calculations and demonstrating that configuring the RIS under both instantaneous and statistical CSI substantially improves detection performance.