Analysis of Cognitive Radio Networks with Imperfect Sensing and Backup Channels

We study the performance of cognitive radio networks with imperfect spectrum sensing and backup channels (BCs). The cognitive radio network is an opportunistic spectrum access (OSA) network which consists of a primary system serving primary users (PUs) and a secondary system serving secondary users (SUs). There are N primary channels (PCs) licensed to PUs and M BCs dedicated to SUs. The PUs have strict priority over the SUs in utilizing PCs and the SUs first try to opportunistically access the PCs not occupied by the PUs temporarily. In case an SU is blocked from PCs, it then checks the BCs to find a free one. The system is modeled as a three-dimensional continuous-time Markov chain. We precisely specify the state-dependent transition rates due to imperfect sensing by simple recursive functions so that we are able to investigate the performance of the OSA system with imperfect sensing and BCs. Our results suggest that when the SUs opportunistically access the PCs alone with PUs operating in moderate or heavy loading conditions, the performance of SUs may be poor or unacceptable. By using some BCs to accommodate the SU calls blocked from PCs can significantly improve the performance.