Research On Problems Regarding DOA Estimation Under Non-ideal Array Conditions

Polarization-Direction of Arrival(DOA)estimation technology plays an important role in electronic reconnaissance and other fields.Most existing array parameter estimation methods have excellent direction finding performance generally in accurately known array manifolds,namely Obtained under the premise of ideal direction finding conditions wi thout array errors.Under actual conditions,there will be errors in amplitude,phase,position,and mutual coupling caused by non-ideal conditions such as inconsistent receiving channels and mutual interference between array elements.If ideal array manif olds are used for array polarization-DOA estimation,array parameter estimation performance will result.A sharp decline.This paper focuses on the research of non-ideal array polarization-DOA estimation and focuses on the reconstruction of array manifolds based on the manifold separation method under non-ideal conditions,and on the basis of coherent and non-coherent source conditions.Polarization-DOA estimation method and its fast algorithm.The main work of this article is as follows:(1)The signal and error model of the array and the classic basic DOA estimation algorithm required in the subsequent chapters are studied,and the array error models such as amplitude error,phase error,mutual coupling error,and position error are established.Through simulat ion analysis,various errors are applied to the array The effect of antenna direction finding performance.(2)For non-ideal polarization sensitive arrays,study the reconstruction and optimization techniques of manifolds based on manifold separation of non-ideal arrays.The effects of the number of effective aperture function truncation modes based on manifold separation technology and calibration noise on DOA estimation performance are analyzed;in order to improve the performance of manifold reconstruction under low calibration signal-to-noise ratio,the sub-channel decomposition method based on auxiliary channel is used to sample The matrix is optimized;the validity of the non-ideal array based on the manifold separation method for DOA estimation is verifi ed by the spatial spectrum estimation method.(3)In view of the high computational complexity of the 2D-multiple signal classification method(2D-MUSIC)algorithm based on manifold separation technology under the condition of non-coherent sources of conventional scalar arrays,this paper studies the two-dimensional fast Fourier transform MUSIC The algorithm reduces the computational complexity;for the polarization-sensitive array,the feasibility of applying the 2D-Polarimetric Element Space(PES)MUSIC algorithm to the polarization-sensitive array is studied.The simulation shows that the 2D-PES MUSIC algorithm The validity of the polarization-DOA parameter estimation.(4)In view of the high complexity of DOA estimation calculation based on the conventional subspace fitting algorithm aft er reconstructing array manifolds based on manifold separation technology under the con dition of coherent sources,an alternative oblique projection algorithm is studied.Only estimate the direction of the arrival of a single signal to achieve decoherence to reduce the computational complexity;for the problem of high computational complexit y due to polarization-space four-dimensional search in the generalized subspace fitting algorithm under polarization-sensitive arrays,the oblique projection algorithm is studied.On the basis of the DOA estimation result obtained by the generalized subspace fitting algorithm,the polarization vector of the signal is directly projected using oblique projection technology to obtain polarization information,which reduces the computational complexity.