Study On Direction Of Arrival Estimation For Coherent Signals Based On Matrix Reconstruction

With the fierce competition of electronic countermeasure technology in modern military field,the signal density and complexity increase significantly in the actual electromagnetic combat environment.When radar target coexists with coherent decoy jamming sources and other independent radiation sources,the performance of classical subspace DOA estimation algorithms severely degrades or even fails.Although non-subspace algorithms and sparse reconstruction algorithms have good theoretical performance for coherent sources,they still have difficulty to apply to practical engineering applications with complex construction and large computational complexity.The matrix reconstruction algorithms of typical subspace decoherence algorithms have attracted wide attention because of moderate computation,simple construction and can be directly combined with the subspace algorithms to achieve superresolution DOA estimation.However,these algorithms suffer from poor robustness under different phase difference,low estimation performance with close angle separation and strong dependence on array structure in practical engineering applications.To solve the problems,the construction mechanism of matrix reconstruction algorithms is comprehensively studied around the theory and application of the subspace decoherence algorithms relying on the National Natural Science Foundation of China.Based on the principle of signal eigenvector and data matrix reconstruction algorithms,the main research work is made to improve the estimation performance,probability of success resolution,robustness,array structure adaptability as well as reduction of computation,and used in engineering practice.The main research work and innovations are as follows:In order to solve the problem that the estimation performance of the classical matrix reconstruction algorithm is deteriorated due to the influence of phase difference,an eigenvalue decomposition algorithm based on multiple column signal eigenvector reconstruction with better robustness is proposed.Theoretical analysis and simulation results show that the algorithm can solve the phase difference failure problem of the classical matrix reconstruction algorithm,and has smaller array aperture loss(higher array freedom),higher information utilization and better estimation performance than the traditional eigenvector reconstruction algorithm.Aiming at the practical application requirements of target angle range prearrangement,a beam space weighted matrix reconstruction algorithm is proposed,which solves the problem of high computational complexity of classical algorithms.Theoretical analysis and simulation results show that the algorithm can effectively avoid the phase difference failure problem by reasonably selecting the weighting vector factor within the preset angle range,and the computational complexity is lower on the premise of ensuring the direction-finding performance of the algorithm.In order to meet the requirements of target angle range prearrangement,computational complexity and direction-finding accuracy,two square matrix reconstruction algorithms based on covariance matrix and correlation matrix information are proposed.Theoretical analysis and simulation results show that the algorithms have good engineering applicable values with better comprehensive performance,such as estimation accuracy,computational efficiency,and solving phase difference failure.To reduce the strong dependence of the existing matrix reconstruction algorithms on the array structure,and improve the utilization of the array aperture of the guidance system in the combat environment,the uniform circular array is selected as the typical array which is extended from the uniform linear array to other arrays.The mode space transformation eigenvector reconstruction algorithm,beam space weighted matrix reconstruction algorithm and square matrix reconstruction algorithms based on covariance matrix information are proposed respectively.Through theoretical derivation and simulation analysis,the effectiveness of each algorithm is verified,which provides a theoretical reference for its better array structure adaptability in practical engineering application and combat field.