Improved GESD test for cooperative sensing over impaired cognitive radio networks

Cooperative sensing is a process of getting spatial diversity gains to improve the performance of system over impaired cognitive radio networks (CRN). In this paper, an improved generalized extreme studentized deviate (IGESD) method is proposed for effective cooperative sensing, where the algorithm checks the normality of the soft decision data using Shapiro-Wilk (SW) test and estimates the expected number of malicious users in the network. These are the two essential inputs required for GESD test to eliminate significant malicious users accurately in the cooperative network. Hence, in this work, the cooperative spectrum sensing with differential evolution and improved GESD test is performed to enhance the sensing accuracy. The energy detection method is considered as local spectrum sensing due to less complexity and the property of incoherent detection. The results reveal that the cooperative spectrum sensing method with proposed IGESD method improves the sensing accuracy. It can detect low SNR signals of -17dB by employing eight nodes in the cooperation with sample size of 128 at desired sensing probabilities, specified in IEEE 802.22 WRAN standard.