Spectrum Sensing for Cognitive Radio Using Laplacian Kernel Correlator in the Presence of Laplacian Noise

This paper proposes the Laplacian kernel correlator (LKC) to address the limitations of current spectrum sensing techniques in the presence of Laplacian noise interference. The proposed method enhances robustness against impulsive noise by employing a Laplacian kernel function to mitigate outliers deviating from the sample median of received signals. We derive closed-form expressions for the mean and variance of LKC, and further establish explicit expressions for the false alarm probability and detection probability. Performance analyses are further conducted in terms of receiver operating characteristic (ROC) and area under the ROC curve. Monte Carlo simulations not only verified the correctness of the established theoretical results, but also demonstrated the superiority of LKC over other commonly used methods in terms of detection probability under Laplacian noise.