Research On DOA Estimation Method Based On Coprime Sparse Array

In the field of modern signal processing,Direction of Arrival(DOA)estimation technology has been a main research direction of array signal processing.The traditional DOA estimation technology is generally based on the research of uniform linear arrays.This often requires higher performance demand such as the signal-to-noise ratio and the number of snapshots due to the limited number of array elements.Moreover,the uniform linear array must satisfy the Nyquist sampling law,which limits its application in practice.As a special sparse array,the coprime array proposed in recent years can obtain a larger array aperture and degree of freedom while using the same number of physical array elements.Therefore,this thesis studies the DOA estimation based on the coprime array.The main work studied in this thesis is as follows:1.The basic theory of coprime sampling and the principle of coprime array signal processing are introduced.The in-depth analysis of the mutual prime array signal model reveals the principle of its virtual domain signal processing to improve the degree of freedom performance.A DOA estimation algorithm for coprime arrays without spatial smoothing is studied.The DOA estimation is performed by constructing a new matrix instead of the spatial smoothing covariance matrix,and the effectiveness of the spatial smoothing algorithm is verified by experimental simulation.2.Aiming at the problem of optimizing the location of array elements of mutual prime arrays,two methods of optimal array arrangement based on second-order virtual domains are analyzed;then based on the concept of fourth-order difference set arrays,a new fourth-order virtual domain sparse mutual prime array is analyzed The extended structure can be found to effectively increase the degree of freedom of the virtual array.Aiming at the problem of error in the DOA estimation of the off-grid target by traditional methods,a DOA estimation algorithm of off-grid target is proposed based on the array element structure of the fourth-order virtual domain.The alternating descent method is used to alternate between the two variables.The optimal sparse solution is obtained by descent,so that the DOA estimation of the fourth-order virtual array deviating from the grid target is realized.3.Aiming at the problem that the traditional far-field co-prime array DOA estimation method degrades the parameter estimation performance under near-field conditions,a near-field parameter estimation algorithm based on near-field shifted co-prime arrays is proposed using fourth-order cumulant and sparse reconstruction algorithm.First,the fourth-order cumulant is used to convert the two-dimensional parameter estimation problem to the one-dimensional parameter estimation problem,and the sparse reconstruction algorithm is used for DOA estimation to avoid the aperture loss caused by spatial smoothing.The accuracy and freedom of effective parameter estimation are reduced.The number of beats required.