Research On Channel Estimation Of Underwater Acoustic OFDM Communication System Based On Compressed Sensing

Underwater acoustic channel estimation is a very important part of underwater acoustic communication,which is very important to improve the signal demodulation performance of the receiver.Underwater acoustic channel is a complex and sparse channel.The application of Orthogonal frequency division multiplexing(OFDM)system and compressed sensing(CS)theory to underwater acoustic channel estimation can restore underwater acoustic signal at high speed and effectively.Most of the existing underwater acoustic OFDM channel estimation algorithms based on CS are performed symbol by symbol in a static underwater acoustic en vironment,which ignores the dynamic characteristics of the underwater acoustic channel,making the algorithm more complex and poor in real-time.Based on the original underwater acoustic channel estimation algorithm,this paper uses the time-varying characteristics of the underwater acoustic channel to improve the existing algorithm.The main research contents are as follows:Firstly,the research background and significance of the subject are summarized,the characteristics of underwater acoustic channel are briefly described,and the principles of CS theory,block CS theory and time varying block sparse signal CS theory are introduced in detail,which lays a theoretical foundation for the follow-up research of underwater acoustic channel estimation.Secondly,to solve the problem that the traditional underwater acoustic channel estimation algorithm does not consider the time-domain correlation of signals in the time-varying underwater acoustic channel,a dynamic block generalized orthogonal matching pursuit(D-Bg OMP)channel estimation algorithm for underwater acoustic OFDM system based on CS is proposed.The first-order auto-regressive(AR)model is used to model the underwater acoustic channel dynamically,which makes the algorithm make full use of the time-domain correlation of the signal,and then uses the characteristics of the underwater acoustic signal with block structure.The model is combined with the structure characteristics of underwater acoustic signal block to study the dynamic CS algorithm under block structure.Experimental results show that the improved algorithm improves the performance of channel estimationand and reduces the complexity of the algorithm.Thirdly,for the situation that the sparsity of underwater acoustic channel is unknown in many environments,the D-Bg OMP algorithm is improved to dynamic block sparse adaptive generalized orthogonal matching pursuit(D-BSAg OMP)algorithm.This algorithm first predicts the sparsity of the channel through continuous iteration,and then updates the index block set and residual to get the channel estimation value,and determines whether the channel estimation value needs to be deleted by setting the threshold value,so as to continuously improve the performance near optimal channel estimation.The experimental results show that the improved algorithm has better reconstruction performance and better performance.Finally,in order to make full use of the correlation between and internal signals,a distributed compressed sensing block sparse adaptive matching pursuit(DCS-BSAMP)algorithm is proposed based on distributed CS theory.This algorithm converts the recovery of a sparse signal from a single block to a sparse recovery of multiple blocks.Under the JSM-1 model,the channel estimation is divided into two parts: common taps and independent taps,and error taps are deleted through continuous iteration to improve channel estimation.Experimental results show that the improved algorithm has better channel estimation performance and lower complexity.