Research On Time Delay Estimation Algorithm Based On Compressed Sensing

Time delay estimation is a classical signal processing technology,which is widely used in the field of wireless location.The correlational research based on small sample is the hotspot and difficulty in the field.With the development of compressed sensing theory and technology,the research on signal restoration and reconstruction based on sparse information has made great progress,which provides a new method to solve the problem of time delay estimation under the condition of small sample.This paper focuses on the research of time delay estimation based on compressed sensing.The main work is as follows:1.In order to solve the problem of the modeling research deficiency of the compressed sensing time delay estimation,a delay estimation model based on compressed sensing is proposed according to the study of sparse representation of angle domain.The sparse representation of gain coefficient is achieved by extending the steering vector matrix of time delay estimation.Furthermore,the model is abstracted as the problem of minimizing the zero norm,which provides a theoretical basis for further research.2.In order to solve the problem of the low precision of the existing compressed sensing time delay estimation algorithm,a novel sparse reconstruction algorithm based on backtracking filter is proposed to estimate the time delay.In this algorithm,the idea of backtracking filter is used to improve the method of atoms selection,which reduces the error probability and improves the estimation accuracy.Simulation and analysis show that the mean square error of the proposed algorithm is reduced by 11.7% compared with the ROMP algorithm,and the accuracy of the time delay estimation algorithm based on compressed sensing is improved effectively.3.In order to solve the problem of the high storage expense of the existing compressed sensing time delay estimation algorithm,a novel compressed sensing time delay estimation algorithm based on progressive edge-growth is proposed.The LDPC matrix measurement matrix with quasi cyclic structure is constructed by using the idea of progressive edge-growth based on the derivation of the design criterion,which effectively reduce the storage cost of measurement matrix.Simulation and analysis show that,the storage space of proposed algorithm is decreased by an order of magnitude under the same estimation performance conditions compared with the Gauss random measurement matrix.4.In order to solve the problem of the low precision of the existing compressed sensing time delay estimation algorithm under the small snapshots environment,a novel compressed sensing subspace algorithm is proposed.The method derives the condition of Eigen decomposition,the orthogonal signal and noise subspace is constructed by sparse reconstruction.And then,RootMusic algorithm is used to realize spectral peak searching,which effectively solve the problem when the snapshot number is less than the multipath number.Simulation and analysis show that the proposed algorithm can effectively improve the estimation accuracy under the small snapshots environment,enhance the universality of the algorithm.