Cluster-based cooperative spectrum sensing for cognitive radio under bandwidth constraints

Cooperative spectrum sensing is investigated for the purpose of detecting the primary user accurately in cognitive radio systems. As the report channel between cognitive users and fusion center experiences fading or shadowing, a cluster based cooperative sensing method is proposed to improve the sensing performance. However, when the number of the cognitive users in each cluster increases, the bandwidth for reporting the sensing results to the cluster head will be very enormous. In this paper, we assume that the cognitive users with enough information in each cluster will send their sensing results to the cluster head so as to decrease the number of sensing bits when performing spectrum sensing. The spectrum efficiency is also studied. Simulation results show that, by using this fusion method in spectrum sensing, the average number of sensing bits decreases greatly at the cost of a little loss in the performance of spectrum sensing.