A Novel T–Test Based Spectrum Sensing Technique for SIMO and MIMO Cognitive Radio Networks

Cognitive radio (CR) is perceived as an intelligent solution to the incompetence of spectrum utilization. The efficacy of any CR system depends on the accuracy of the spectrum sensing (SS) technology used. Therefore, the SS is considered as a substantial task of the CR systems. In this paper, the SS is handled by a different perspective as a goodness of fit hypothesis testing problem. Therefore, a T-Test based spectrum sensing (TTBSS) algorithm is proposed based on the statistical properties of the student's t-distribution. It does not require any prior knowledge of the primary user (PU) signal. It determines the PU absence/presence by calculating the dissimilarity between the two samples means of the received signal when the PU is present and when the PU is absent. Closed-form expressions for the probability of detection, probability of false alarm, and their associated thresholds are derived considering both single-input multi-output (SIMO) and multi-input multi-output (MIMO) systems configurations. The simulations revealed that the proposed algorithm is superior to the other state-of-the-art PU detection algorithms particularly at low signal-to-noise ratios (SNR) and few samples conditions. Moreover, the TTBSS addressed many problems that restrict the performance of the SS process such as noise uncertainty and interference. In addition, the numerical results are highly matched to the theoretical analyses.