DOA Estimation Of Coherent Signals Based On Matrix Recontruction

Direction of arrival(DOA)estimation is an important branch of array signal processing.The traditional high-resolution algorithm can accurately estimate the incident angle of the signal under independent signal conditions.However,under coherent conditions,the signal and noise subspace cannot be accurately estimated due to the decrease in the rank of the signal covariance matrix,and the traditional algorithm fails.At the same time,the coherent signal will invalidate the conventional source number estimation algorithm.In terms of DOA estimation of coherent signals,the dimensionality-decomposition coherence algorithm based on spatial smoothing and matrix reconstruction,although able to effectively process coherent signals,is at the expense of array freedom,and they all depend on accurate estimates of the number of signals.In this paper,the problem of traditional decoherence algorithm is studied.The matrix reconstruction method based on oblique projection and joint diagonalization is used to study the DOA estimation of coherent signals and mixed signals.The specific content includes the following aspects:Firstly,the basic knowledge of DOA estimation is thoroughly studied.Besides,the principle of classical DOA estimation algorithm for coherent signals is theoretically deduced and proved,including spatial smoothing algorithm,vector singular value algorithm and Toeplitz matrix reconstruction algorithm.Secondly,aiming at the problem of large loss of array freedom under mixed signal background,a phased estimation algorithm based on oblique projection technique for modified covariance matrix is proposed.This method uses ESPRIT algorithm to estimate DOA value of independent signals,avoiding peak searching or polynomial root finding.At the same time,the oblique projection of the modified covariance matrix can effectively remove the noise interference and obtain the accurate covariance matrix of coherent signals.Finally,DOA estimation of coherent signals is carried out by using smoothing method.This method uses the idea of separate estimation of independent signals and coherent signals to improve the total number of estimable signals in the array,and has advantages in estimation accuracy and effective performance.Finally,for the traditional decoherence algorithm,which relies on the prior knowledge of the number of signals,a cost function construction algorithm based on joint diagonalization matrix is proposed in the context of fully coherent signals and mixed signals.Firstly,a set of covariance matrices with joint diagonalization structure and effective restoration of rank is constructed.Then the spatial spectrum of the signal is obtained by constructing the cost function,and DOA estimation of coherent and mixed signals is realized under the condition of unknown number of sources in the end.