Research On DOA Estimation With Noncircular Signal In Nonuniform Noise Environment

Direction of arrival estimation(Direction-of-Arrival,DOA)is an essential direction of array signal processing.It is essential for the progress in the fields of military radar,ocean exploration,smart cars,smart antennas,robots and other fields.Because of its wide application,it has attracted many scholars,and it has become an essential research direction.However,many scholars have studied algorithms for circular signals,and most of the algorithms are suitable for uniform noise.There is little research on noncircular signal with nonuniform noise.Since the noncircular signal model can expand the array aperture by adding virtual elements,the algorithms with noncircular signal model can provide superior performance.Since the receiving channel will be interfered by the hardware and the environment,the sensor will be affected by the uncorrelated non-uniform Gaussian white noise.In this case,the performance of most classical DOA estimation algorithms will be significantly reduced or even ineffective.So DOA estimation algorithms with high resolution,high precision and less computation under nonuniform noise will be popular among scholars.This paper makes an in-depth research on the DOA estimation of noncircular signals in nonuniform noise.First of all,the basic theory of DOA estimation is introduced in this paper,which provides theoretical support for the later study.Then the mathematical models of array signals are introduced in detail.Then,verify the basic knowledge through the existing algorithms,such as MUSIC algorithm and ESPRIT algorithm.Then simulation experiments are carried out according to different experimental conditions,and observe the performance of the algorithms.Then,the DOA estimation of noncircular signal is introduced in this paper.Firstly,the definition and properties of noncircular signal are introduced.The extended array and related mathematical model based on the characteristics of noncircular signal are given.The NCMUSIC algorithm and the optimized dimension reduction NC-MUSIC algorithm are introduced and deduced in detail.Then simulation experiments are carried out according to different experimental conditions,and observe the performance of the algorithms.Finally,two DOA estimation algorithms for noncircular signals in nonuniform noise environment are proposed in this paper.In the first algorithm,an interference suppression algorithm is proposed to reconstruct the array output covariance matrix.It reconstructs the covariance matrix by directly removing the diagonal elements of the output covariance matrix.This method removes the noise interference directly,and there is no need to estimate the original covariance matrix.The second algorithm gives a method to estimate the noise covariance matrix and uses the generalized eigenvalue decomposition method to obtain the noise subspace.There is no iterative process and convex optimization process in the two algorithms.Then,the simulation results have proved that the methods in this paper can effectively remove the interference of non-uniform noise and has high angle resolution and estimation accuracy.In addition,the Kramero bound under the noncircular signal model is also given in this paper.