Research On Selective Cooperative Spectrum Sensing Methods

Cognitive Radio(CR)is the main technical method to effectively alleviate the scarcity of spectrum resources,and spectrum sensing is the first problem to be solved by the entire cognitive radio system.Spectrum sensing can accurately and accurately sense the spectrum occupancy in real time,find and determine idle frequency bands and use them reasonably,but single-user spectrum sensing will be affected by channel fading and other factors,resulting in poor sensing performance.Multi-user cooperative spectrum sensing significantly improves this problem.In order to improve the performance of cooperative spectrum sensing detection,this paper starts from two perspectives.On the one hand,the random nature of the wireless channel will cause uncertainty in the local detection result of the sensing user(SU),which makes the detection performance of the existing cooperative spectrum-sensing fusion algorithms poor,and none of the traditional cooperative sensing fusion algorithms Consider performance differences between perceived users.For this reason,a cooperative spectrum sensing algorithm based on D-S evidence theory is proposed.D-S evidence theory algorithm is known for its ability to handle uncertain information.The algorithm first extracts local evidence for each perceived user separately,and then calculates the reliability of each perceived user by considering the current reliability and historical reputation of the perceived user.Similarly,the basic probability distribution function of each perceived user is re-derived according to the credibility factor.Finally,in the Fusion Center(FC),D-S evidence theory fusion rules are used to fuse the evidence of each perceived user and make a final decision on the existence of the primary user(PU).Simulation results show that the proposed algorithm has a higher detection probability regardless of whether the signal to noise ratio(SNR)of the perceived user is the same.On the other hand,the energy detection algorithm is an important and most widely used local detection algorithm in the cooperative spectrum sensing model,but it is susceptible to noise and has poor sensing performance in the presence of noise uncertainty and low signal-to-noise ratio.Unable to maintain robustness.To solve this problem,a selective cooperative spectrum sensing algorithm based on dual-threshold energy detection is proposed.Unlike the traditional dual-threshold energy detection algorithm,the proposed algorithm sets three decision thresholds under the noise uncertainty model.Use the "OR" fusion algorithm to perceive the perceptual results of users who fall outside the two thresholds,and select the third threshold set by the perceptual users that fall between the two thresholds to filter the perceptual users with superior perceptual performance.Come out to participate in cooperative sensing,and use the "k-out-of-N" fusion algorithm to fuse the selected sensing user's sensing results.In the end,FC uses the "OR" fusion algorithm to finally fuse the perception results of the two parts of the user and make a decision.Simulation results show that the proposed algorithm can effectively solve the problem of noise uncertainty,and can maintain good robustness under high false alarm probability.Under QPSK modulation,when the perceived user signal-to-noise ratio is 0.1,the false alarm probability is 0.1,and the threshold adjustment factor is 0.5,the detection probability of the main user signal can reach 87%,and it has better detection in the presence of noise uncertainty Performance,reducing interference from perceived users to key users.