Threshold optimization of finite sample based cognitive radio network

Abstract

In this paper, we consider a cooperative cognitive network containing two cognitive radios (CRs) and one primary user. Cooperation between the CRs is achieved through a fusion center (FC). The CRs forward energy estimates to the FC by using finite number of received data samples. The FC utilizes an optimized global threshold based on local thresholds of the CRs for determining presence or absence of the primary signal. We derive average probability of error of detecting the primary user in this set-up. An optimized global threshold in the FC is obtained by minimizing the average probability of error in detection of a spectrum hole. The average probability of error is assumed to be the sum of the probability of false alarm and missed detection in the FC.