Research On Spectrum Detection Algorithm Based On Wavelet Denoising And Compressed Sensing

With the booming of wireless communication technology and the current increasing popularity of various wireless intelligent terminals,wireless spectrum resources,the radio frequency band on which wireless communication is reliant,are getting scarcer due to its limited non-renewability.Fixed partitioned spectrum allocation is widely used all over the world.Users with different levels and priorities occupy a fixed number of independent subbands called authorized spectrum bands which are exclusively for primary users who are their corresponding users.Other unauthorized users are strictly prohibited from using them.However,this spectrum allocation method is irreconcilable with the current skyrocketing demand for frequency bands.Relevant research also found that a large number of frequency bands are idle at different time and locations,without being used efficiently.As a key technology to solve the above problems,cognitive radio,which has been widely concerned by academia since proposed,is considered as a key driver of future wireless communication systems.Users with "cognitive" function access cognitive radio system through spectrum sensing,interference avoidance,resource allocation and other technologies.While improving spectrum utilization,they should endeavor to avoid introducing interference in order to share authorized frequency band with primary users.The center of this thesis is the spectrum sensing algorithm in cognitive radio technology.Firstly,the background and current development of cognitive radio technology and spectrum sensing algorithm are introduced,and the advantages and disadvantages of different algorithms at present are compared and analyzed.In order to improve the performance of spectrum sensing under low SNR and make the spectrum sensing algorithm less complicated,the soft-hard threshold compromise method is adopted to combine the differential energy detection model with the wavelet threshold denoising algorithm.In view of the inadequacies of the current algorithms,the selection and detection process of the threshold in the dualthreshold spectrum sensing algorithm according to energy and wavelet transform are optimized.The simulation results suggest that the refined dual-threshold spectrum sensing algorithm improve the performance of spectrum detection under the influence of uncertain ambient noise.Secondly,the application of the dual-threshold spectrum sensing algorithm is expanded to the detection of broadband spectrum.The utilization of wireless spectrum resources by users has now changed from the narrowband signal to the broadband one.The sampling of broadband signals is costly due to its constraining by Nyquist sampling theorem.In order to complete the sampling and detection of broadband signals at a lower cost and faster speed,with the advantage that compressed sensing technology can reconstruct the original signal with fewer numbers than Nyquist sampling number and using wavelet transform technology for spectrum edge detection,a dual-threshold broadband spectrum cooperative sensing algorithm is proposed.The specific step is to sample and reconstruct the broadband signals to be detected through compressed sensing technology,then use wavelet transform technology to conduct spectrum edge detection,and divide the broadband spectrum into several subbands whose real-time status is judged by dual-threshold cooperative detection one by one,until the detection range cover the whole interested broadband bands.Through this,the cost of broadband spectrum detection system is lowered,and the effectiveness of the proposed algorithm is demonstrated by simulation analysis.Finally,the content of this thesis is summarized,and the shortcomings of spectrum sensing algorithm which need to be improved in the practical application and its future research direction are analyzed.