Research On Angle Estimation Algorithm Of Distributed Arrays Under Impulse Noise

In recent years,the development of direction of arrival(DOA)estimation with sparse array has become increasingly mature.As a new type of special sparse array;distributed arrays have the advantages of strong flexibility and high degree of freedom,which has aroused extensive attention from scholars at home and abroad.Based on the advantage that non-circular signals can increase the amount of receiving matrix data and obtain more available information by using their non-circular characteristics,this thesis combines the advantages of non-circular characteristics and distributed arrays,and designs distributed arrays DOA estimation algorithms based on non-circular signals,circular and non-circular mixed signals.At present,the DOA estimation of distributed arrays is carried out under the background of Gaussian noise.However,most of the noises in real life are impulsive noises with strong impulsive characteristics,which will interfere with the DOA estimation of distributed arrays and degrade its performance.In order to solve this problem,the following contents are studied in this thesis:1.The DOA estimation of distributed arrays under impulsive noise is studied.Taking the Alpha stable distribution as the impulse noise model,the array received signal is preprocessed by two nonlinear transformation methods,fractional low-order transformation(FLOT)and compression transformation(CT),to effectively suppress the impulse noise.Based on the covariance matrix of the preprocessed signal,two robust dual-size unitary ESPRIT(U-DS-ESPRIT)algorithms are proposed.In order to evaluate the performance of parameter estimation,the Cramer-Rao Bound(CRB)of DOA estimation for distributed array under impulse noise is analyzed.Simulation results show that the two robust U-DS-ESPRIT algorithms proposed in this thesis can effectively suppress impulse noise under low signal-to-noise ratio(SNR)and high pulse intensity.2.The DOA estimation of distributed arrays for noncircular signals under impulsive noise is studied.Firstly,the distributed arrays DOA estimation of non-circular signals in Gaussian background is studied,and a dual-size unitary non-circular ESPRIT(U-DS-NC-ESPRIT)algorithm is proposed.This thesis also analyzes the CRB of DOA estimation of non-circular signals under distributed arrays.Simulation results show that the proposed algorithm can expand the physical aperture of the array and further improve the performance of DOA estimation.Considering the performance degradation of non-circular signals under impulse noise,the non-circular characteristics are extended to Alpha stable distribution signals by using FLOT and CT.The fractional low-order covariance matrix(FLOC)and compressed transform covariance matrix(CTC)of the signals received by the extended array are constructed.Combining FLOC and CTC with U-DS-NC-ESPRIT algorithm,two robust DOA estimation algorithms are proposed.Simulation results show that the two algorithms can improve DOA estimation accuracy while suppressing impulse noise.3.The DOA estimation of distributed arrays for circular and non-circular mixed signals under impulsive noise is studied.Firstly,the DOA estimation of distributed arrays for circular and non-circular mixed signals in Gaussian background is studied,and proposes a dual-size unitary mixed-ESPRIT(U-DS-Mix-ESPRIT)algorithm.Aiming at the problem of impulse noise in the actual environment,the received signals of circular and non-circular mixed extended array can be preprocessed by using FLOT and CT,and the FLOC and CTC of the received signals are constructed which have the same structure with the second-order covariance matrix under Gaussian distribution.Two robust DOA estimation algorithms are proposed by combining with the U-DS-Mix-ESPRIT algorithm.Simulation results show that the proposed algorithm has more robust DOA estimation performance under impulse noise.