Application Research Of Compressed Sensing In OFDM Systems Channel Estimation

Performance of the wireless communication system is mainly restricted by the characteristics of the radio channel.Since the complexity and variability of wireless channel spread cause the receiving signals distort easily.In order to obtain accurate receive signal,it's necessary to estimate the wireless channel.Traditional channel estimation method requires a large amount of pilots,which leads to a low spectrum efficiency.Compressed sensing theory is a novel sampling theory based on the proposed of Nyquist sampling theorem,and can recovery signals accurately with a high probability in the case of to ensure no loss of information.With inherent sparsity characteristics of wireless multipath channel,the compressed sensing reconstruction algorithm can be applied to orthogonal frequency division multiplexing system,accuracy of OFDM system channel estimation will directly affect the performance of the channel,The main work for this paper is as follows:Firstly,selective fading channel estimation based on compressed sensing algorithm are studied.The classical reconstruction algorithm Orthogonal Matching Pursuit is used in performance estimation,simulation comparing it with the traditional channel estimation least squares method.Secondly,time-frequency double selective fading channel estimation based on StOMP algorithm are studied.Considering the reconstruction using long time of OMP algorithm,this thesis analyzes the underlying causes for a long time of the reconstruction algorithm matching pursuit class,StOMP algorithm of OMP improved algorithm is used in OFDM time-frequency double selective channel estimation.This algorithm selection algorithm for each of a plurality atoms rather than individual atoms,which can reduce the number of matches and improves the speed of operation.The simulation results show that the algorithm is used in this thesis for better performance complex reconstruction time and reconstruction precisionFinally,time-frequency double selective fading channel estimation based on StOMP improved algorithm are studied.Considering the StOMP algorithm for reconstruction precision is not ideal but also know sparsity channel as a priori condition,break regularization Orthogonal Matching Pursuit algorithm of StOMP improved algorithm is used in OFDM time-frequency double selective channel estimation.The algorithm sets threshold value based on the current residual in order to select atomic of correlation coefficient greater than the threshold value,thereby reduce the algorithm reconstruction time and improves the reconstruction accuracy.The simulation results show that the presented algorithm can achieve higher estimation accuracy and shorter reconstruction time with no prior knowledge of channel sparsity.