Research On Performance Of Cooperative Spectrum Sensing In Cognitive Radio Networks

During the last decade, as the wireless services develop rapidly, the spectrum resource is more and more scarce; meanwhile the utilization rate of some allocated fixed spectral bandwidth is low. In order to alleviate the contradiction, cognitive radio come up and is believed to be the best solution for the problem of low utilization ratio, which can realize the spectrum sharing and doesn’t disturb the primary user. This paper mainly refers to the schemes of cooperative spectrum sensing and the fusion algorithms.At first, this paper introduces research condition of cognitive radio at home and abroad, then illustrates some key technologies of single cognitive user on spectrum sensing detailed, finally we focus on the technologies of cooperative spectrum sensing and mainly study two improved schemes.1. Considering that the cognitive users of low SNR will pull the whole detection performance down in cognitive radio network, we bring the SNR-compared method in the clustering scheme. The process of this new scheme is that the fusion center separates the cognitive users into some clusters; after that the users send their energy statistics and SNR to their own cluster head; the next each cluster head orders the users in its cluster from big to small according to their SNR, and then fuses the energy statistics of the half users which have the higher SNR; finally every cluster head sends their results to the fusion center to make a decision. Simulation results prove that the detection performance of this proposed scheme is significant.2. Based on the time frame of sequential cooperative spectrum sensing, this paper discusses a sequential cooperative spectrum sensing method. To improve every cognitive user’s detection performance, the cluster head orders the cognitive users from big to small according to their SNR, which make the users of low SNR will get more sensing time. Simulation results prove that this proposed method can improve sensing performance of the system. Besides, under the AWGN environment, we present a method of compared sequential probability ratio test which can reduce the number of users to make a decision. To get higher reaction speed, the fusion center compares each user’s log likelihood ratio before fusing. Through1000test statistics, the results indicate that the proposed scheme can reduce the users in cooperation, and improve the reaction speed of spectrum change status.