Parihar Nikhilsingh Pradipsingh supervised by Dr. Praful Mankar received his Master of Science by Research in Electronics and Communication Engineering (ECE). Here’s a summary of his research work on Spectrum Sensing for RIS-aided Communication Systems: Analysis and Design

Spectrum sensing facilitates an opportunistic secondary user to access underutilized spec-trum while protecting primary users from interference. In practice, the reliability of spectrum sensing is constrained by low signal-to-noise ratios (SNR), multipath fading, correlated noise, and channel state information (CSI) uncertainty. This thesis develops statistical signal process-ing frameworks to mitigate these challenges for the reconfigurable intelligent surface (RIS)-assisted systems. Under a complex Gaussian observation model, the received signal statistics are captured by a sample covariance matrix following a central complex Wishart distribution. Distinguishing the presence of a primary user from a noise-only scenario is based on the max-imum eigenvalues of this matrix. However, deriving the exact distribution of the maximum eigenvalue becomes highly challenging in the presence of joint spatial fading and noise corre-lations.

To address this, a maximum eigenvalue detection (MED) framework is presented for RIS-assisted spectrum sensing. Using random matrix theory, the exact finite-dimensional cumula-tive distribution function of the largest eigenvalue is derived for both the null and alternative hypotheses. This facilitates the direct derivation of closed-form false alarm and detection prob-abilities, bypassing need of the standard asymptotic analyses. Furthermore, RIS phase shifts are optimized to maximize the expected test statistic. The results, which includes simulation validation, indicate that the RIS-integrated MED significantly improves detection reliability as a function of the number of reflecting elements with the optimized phase shift configuration.

Further, for the case of unknown channel distribution and transmission power, a generalized likelihood ratio test (GLRT)-based framework is proposed for RIS-assisted systems in corre-lated noise environments. This joint estimation-detection methodology utilizes a group-wise RIS activation scheme and maximum likelihood estimation (MLE) to estimate the unknown channel and transmit power. Subsequently, the RIS phase shifts are optimized using these parameter estimates to maximize the total sensing gain. The derived channel and power esti-mates are then utilized to formulate a GLRT detector, enabling robust hypothesis testing under unknown parameter. The framework achieves reliable sensing under CSI uncertainty and out-performs conventional energy-based detection.

July 2026