Cooperative Spectrum Sensing Algorithm Of Cognitive Radio System Based On Tensor Analysis

In recent years,5G technology,Internet of things,Internet of vehicles and other network information technologies are developing rapidly.Moreover,they have been put into use on a large scale in the fields of family,work and public area,etc.Naturally,people need more and more spectrum resources.But the spectrum resources are more and more scarce in the actual environment.At the same time,the researchers found that the current mainstream spectrum allocation mechanisms mainly include fixed allocation and access methods.These two mechanical allocation mechanisms lead to the low utilization rate of the effective spectrum.Therefore,the researchers focus on how to improve the spectrum utilization.However,the application of cognitive radio technology improves the utilization rate of spectrum to a certain extent.Spectrum sensing as one of its core technologies has attracted widespread attention in the scientific and technological community.Although the current spectrum sensing technology has different advantages in spectrum detection,it causes to the loss of original signal in the face of increasing spectrum demand and high-dimensional signal.Due to the development of modern network information technology,the signal data obtained are all high-dimensional data.The processing of high-dimensional data is usually in the form of vector or matrix,which reduces the complexity of the operation.But it causes the loss of the original signal,or be polluted by noise and other interference in the transmission process.As a result,the measured spectrum data is often incomplete,which leads to the signal cannot recover the original signal better in the reconstruction process.As a result,the detection performance of the algorithm is affected and the detection probability is too low.So that the fusion center cannot accurately determine whether the corresponding spectrum is working from the primary user,which affects the use of the effective spectrum by the secondary users and reduces the utilization rate of the effective spectrum.In order to solve the above problems,the third-order power signal tensor model is established based on the three-dimensional information of channel and main user's position,which is considered the influence of channel fading and noise interference on the spectrum information in the framework of cognitive radio system network.And the low rank tensor analysis algorithm is used to recover the missing and damaged information of the original signal.The detection probability and false alarm probability are calculated by judging the main user's working state of the corresponding spectrum.Moreover the sensing performance of the algorithm can be measured by these two standards.So that the fusion center can better allocate the spectrum to the secondary users and improve the spectrum utilization.Due to the high complexity and long running time of low rank tensor completion algorithm and in the presence of noise interference,the effect of reconstruction of the original signal is poor.For solve the problems of the above low rank tensor completion algorithm,an improved low rank tensor analysis algorithm is introduced to transform the processing of signal data from large-scale singular value decomposition to small-scale singular value decomposition.Finally,compared the two algorithms,the improved low rank tensor analysis algorithm can further improve the spectrum sensing detection performance in MATLAB simulation platforms.