Analysis of Distributed Consensus-Based Spectrum Sensing Algorithm in Cognitive Radio Networks

In distributed cognitive radio network (CRN), it is always not available to collect the sensing results from all cognitive users due to the lack of a common fusion center, which makes it critical to keep consistency and reliability of the spectrum sensing results. For consensus-based sensing algorithm, a security and connectivity analysis is implemented. Simulation shows that if there exists one attacker who can stay in cognitive users´ trust neighbor set after the set updates to a stable stage, it will mislead others to converge sensing data to the value provided by the attacker. Therefore, all users in the local network will make wrong decision. In addition, the simulation also shows that the whole network may be divided into small local networks which will make different sensing results. It is impossible to reach a consensus. For the features of spectrum sensing in cognitive radio networks, considering the possible problem in sensing data updating, some suggestions for spectrum sensing in distributed cognitive networks are proposed in the end of the paper.