Research On Channel Estimation Based On Compressed Sensing

With the development of science and technology,mobile communication technology is improving little by little every day.From the first generation of frequency division multiple access technology,to the second generation of time division multiple access technology,to the third generation of code division multiple access technology,and then to the fourth generation of orthogonal frequency division multiplexing technology,the transmission rate and transmission quality of wireless communication has improved over time.Nowadays,the development of cognitive radio and ultra-wideband technology has led to the increasing demand for sampling technology.If the sampling frequency of the original Nyquist is followed,then the sampling rate of the signal bandwidth of at least two times is required,which is a waste of spectrum resources and a severe requirement for analog digital converter.The emergence of the theory of compressive sensing has brought hope to all of this.And the proof of channel sparsity provides the theoretical feasibility for the application of compressive sensing theory to channel estimation.This paper mainly studies the channel estimation of channel based on orthogonal frequency division multiplexing and multi-input multi-output.By introducing the compression sensing technology,the number of pilot frequencies needed for channel estimation is greatly reduced,and the utilization rate of spectrum resources is increased.the fading characteristics and propagation model of wireless channel is introduced,and the technical principle of OFDM and MIMO-OFDM and the feasibility of using compressed sensing in channel estimation is analyzed in this paper.In this paper,several common algorithms of compressed sensing are studied for sparse channel recovery,and an improved compression sampling matching tracking algorithm is proposed in this paper.the channel estimation based on compressed sensing needs fewer pilot frequencies to achieve a good reconstruction effect than the conventional channel estimation is verfied in the simulation.Compared with the sparse adaptive algorithm,the improved compression sampling matching tracking algorithm proposed in this paper has an improved error ratio of 1dB.Further the estimation of compressed sensing channel based on bayesian theory is studied.After the greedy algorithm is used to reconstruct the channel estimation based on Bayesian compressed sensing,the improved Bayesian algorithm and the improved sparse adaptive Bayesian algorithm are proposed.Compared with the orthogonal matching tracking algorithm,the error ratio of bayesian compressed sensing algorithm is improved by about 1dB.Compared with bayesian compressed sensing algorithm,the improved bayesian compressed sensing algorithm also has about 1dB improvement.Compared with the sparse adaptive algorithm,the improved sparse adaptive bayesian compression sensing algorithm has a gain of about 3dB.