Exploitation of temporal persistence for accuracy improvement in primary user detection

Cognitive radio (CR) is a promising technology to alleviate current spectrum scarcity via enabling the secondary user (SU) to opportunistically access the unutilised spectrum band that has been licensed to the primary user (PU) for exclusive use. To identify the access opportunities, SU should reliably detect whether PU is present or not. Previous works on the problem of PU detection have suggested the approach of multi-SU cooperation in order to enhance detection reliability. In this study, the authors consider that PU maintains the same state (absence or presence) for at least M (M ≥ 2) slots and develop a new approach of exploiting M-slot temporal persistence for improving the accuracy of PU detection in terms of reducing the detection Bayesian cost. To achieve this goal, Bayesian energy detection (BED) scheme is firstly presented. Then based on BED scheme, the authors propose an improved scheme to exploit the M-slot temporal persistence. It is shown that, if M = 2, exploitation of M-slot temporal persistence is certainly beneficial and can reduce as much as 1/3 of Bayesian cost. If M > 2, more Bayesian cost reduction can also be obtained.