Eigenvalue Based Spectrum Sensing Scheme In Cognitive Radio

With the progress of society and the rapid development of 5G communication technology,the demand for spectrum resources is also increasing day by day,and the utilization rate of some frequency band resources is low due to unreasonable allocation.Cognitive radio technology can effectively solve the problems of spectrum resource shortage and low utilization.The basic idea of cognitive radio technology is that secondary users continuously detect the target spectrum in multi-dimensional spaces such as time domain and frequency domain.Secondary users dynamically access the spectrum in order to improve the utilization of spectrum resources when the target spectrum is idle.Based on the traditional eigenvalue-based spectrum sensing algorithm,this thesis proposes fusion eigenvalue sensing and selective spectrum sensing algorithms.The main contents of this thesis are as follows:(1)The traditional eigenvalue-based spectrum sensing algorithm only uses part of the eigenvalue information of the covariance matrix,so the sensing performance is not good in low SNR.Aiming at this problem,this thesis proposes a fusion eigenvalue perception algorithm.The detection algorithm uses the maximum eigenvalue,minimum eigenvalue,average eigenvalue and fusion parameters of the covariance matrix to construct new statistics.Due to the information used in this algorithm is more comprehensive,the simulation results show that the algorithm has higher detection performance and the false alarm probability meets the system requirements.?(2)A selective spectrum sensing method is proposed for low SNR users in the sensing environment.A credibility is set for each cognitive user,which represents the accuracy of the cognitive user in the previous perception.Then this algorithm uses the credibility to transform the traditional "K-N" criterion to complete the energy detection.When the energy detection results are not accurate,based on the criteria of proximity to the primary users and high reliability,the appropriate secondary users are selected to participate in the eigenvalue detection.The simulation results show that there are low SNR users in the perceptual environment,this algorithm can effectively avoid the influence of the users on system detection performance and improve system performance.