Research On Compressed Sensing-based Channel Detection Algorithm In Intelligent Anti-interference

With the rapid development of wireless communication technology and the steady expansion of wireless communication services,the electromagnetic environment is becoming more and more complex and volatile.The ubiquitous artificial and natural interference has become the biggest obstacle to high-quality,high-speed and high-efficiency transmission of broadband communications.Research on intelligent anti-interference systems is particularly urgent.As the channel detection module at the core of the intelligent anti-interference system,its performance directly determines the accuracy of subsequent parameter decisions and the quality of the system's anti-interference performance.Traditional channel estimation methods generally use uniform pilots and linear interpolation,resulting in problems such as low frequency band utilization and low recovery accuracy.The estimation method based on compressed sensing can greatly reduce the data rate through sparse expression and accurate reconstruction,thereby reducing hardware pressure and reducing system overhead.Based on this,this paper studies the channel estimation algorithm based on compressed sensing and Bayesian compressed sensing in the interference environment to provide accurate channel information for the subsequent parameter decision of the intelligent anti-interference system,thereby further improving the anti-interference ability of the system.First,the intelligent anti-interference system architecture is introduced,and the advantages of channel detection algorithms based on compressed sensing are introduced.Then,the compressed sensing theory and its important concepts are analyzed,and the implementation steps,advantages and disadvantages of the convex optimization algorithm,greedy tracking algorithm and Bayesian-based compressed sensing algorithm are analyzed,which lays the foundation for the subsequent chapters of this article.Second,starting from the OFDM channel estimation model,its system framework,key features and channel estimation model are introduced;its focus on the research of OMP,CoSaMP and SAMP based on three compressed sensing estimation algorithms,performance and application scenarios,and traditional Compared with the LS and MMSE algorithms,not only can the number of pilots be saved,but also the estimation error can be reduced.For the problem of sparse channel estimation,the effect of different interference and pilot structures on channel estimation performance is theoretically analyzed.Experiments show that compressed sensing method has better performance on sparse channels,but it must be on the premise that the channel sparseness is known,and the algorithm has poor adaptability.Then,the compressed sensing reconstruction algorithm under Bayesian theory is studied to realize adaptive sparse channel estimation.Under the framework of JSM-2 model,the Bayesian compressed sensing and Laplace prior based complex Bayesian channel estimation algorithm are derived in detail and simulated with five reconstruction algorithms such as LS,LMMSE,OMP,SAMP,Co SAMP.By comparison,the results show that the OMP performance is the best when the sparsity is known;the two Bayesian-based algorithms have better estimation performance when the channel sparsity is unknown.Finally,in order to further improve the reconstruction accuracy under low signal-to-noise ratio environment,combined with the wavelet tree structure model,the DCT base is used to replace the wavelet base,and a channel estimation algorithm based on TSBCS is proposed,which further improves the convergence speed and the estimation under low SNR Precision.In addition,the simulation compares the reconstruction accuracy of the above algorithm in the case of multi-tone interference.The results show that TSBCS is most suitable for low signal-to-interference ratio and has strong anti-interference ability.