EEF criterion based user selection for cooperative spectrum sensing in cognitive radio network

Abstract

Spectrum sensing is a fundamental technique of cognitive radio (CR) system to detect the presence of primary user (PU) transmissions in the licensed spectrum. This paper investigates secondary user (SU) selection based cooperative spectrum sensing under exponentially embedded family (EEF) criterion. With an aim to estimate the optimal number of cooperative users who are better fitting for participating in cooperative sensing, we propose AIC, MDL, and EEF criteria to select the potential users among all cooperative users in the CR network. Based on the estimated user number, the global test statistic (GTS) is generated, and finally the fusion center (FC) makes the global decision on the presence/absence of the PU signal. Analysis and simulations verify that the proposed user selection based cooperative sensing schemes can significantly improve the spectrum sensing performance.