RIS-GLRT for Robust Spectrum Sensing in MIMO Cognitive Radio Networks
Abstract
Efficient utilization of the radio spectrum is essential for supporting the rapid growth of wireless communication services. Cognitive Radio (CR) offers a promising solution by enabling secondary users (SUs) to opportunistically access spectrum when primary users (PUs) are inactive. Reliable spectrum sensing remains a critical challenge, particularly in low signal-to-noise ratio (SNR) and fading environments, where conventional detectors such as energy detection and eigenvalue-based methods often fail. This paper proposes a Reconfigurable Intelligent Surface (RIS)-empowered Generalized Likelihood Ratio Test (GLRT) framework for robust spectrum sensing in Multiple-Input Multiple-Output (MIMO) CR systems. The RIS introduces controllable reflections that enhance the effective channel gain, thereby improving separability between signal-present and noise-only hypotheses. Analytical formulations for the GLRT test statistic, threshold estimation under a fixed false-alarm constraint, and Monte Carlo calibration are developed. Simulation results demonstrate that the proposed RIS-GLRT achieves up to 35–40% improvement in detection probability at moderate-to-low SNR compared to traditional detectors, while reducing missed detection probability and exhibiting favourable scaling with the number of antennas and RIS elements. These findings confirm the potential of RIS-assisted GLRT to significantly enhance spectrum sensing reliability and establish a strong foundation for next-generation spectrum-sharing systems.