Research On DOA Estimation Algorithm Based On Subspace Analysis

Estimation of direction of arrival(DOA)for spatial signals is an important branch in the area of array signal processing.Two dimensional(2D)DOA estimation based on subspace analysis can obtain more accurate estimation for spatial signals.Subspace-based DOA estimation algorithms make full use of the intrinsic characteristics or second order statistics of the array data matrix.These methods have the advantages of high resolution,low complexity and good real-time performance,so it is a hot topic in the field of DOA estimation.With the continuous deepening of 2D DOA estimation for application study in engineering practice,people have higher requirements for estimation accuracy,resolution and computational complexity.This thesis proposed three improved algorithms on 2D DOA estimation based on the existing subspace-based DOA estimation algorithms.The main research work is as follows:(1)To improve the algorithm performance of 2D DOA estimation under the condition of planar array,an improved algorithm using propagator method(PM)is proposed.First of all,this method extracts four overlapped sub-arrays from the received signal in planar array under some rules.Then it constructs a new data matrix with correlation matrixes of four sub-arrays.Next the rotational invariance matrix via utilizing PM is gotten.Automatically-paired 2D DOA estimation is obtained at last.The computational complexity of the proposed algorithm is less than that of the traditional 2D ESPRIT algorithm.Simulation results show that the estimation accuracy of the improved algorithm is higher than ESPRIT when signal to noise ratio(SNR)is low or sample is less because this algorithm reuses the data of correlation matrixes.(2)Extending the array aperture is an important method to improve the performance of DOA estimation.This thesis proposes an efficient 2D DOA estimation algorithm under the condition of two-parallel uniform linear array.The improved method exploits the conjugate symmetry property of the array manifold matrix to increase the effective array aperture.The method obtains the automatically-paired 2D DOA estimation via combining PM.Simulation results reveal that the proposed algorithm has higher estimation accuracy with low complexity compared with some existing DOA estimation algorithms.(3)For DOA estimation of coherent signals,this thesis proposes an improved 2D DOA estimation algorithm based on a L-shaped array.The method extracts the first column of the correlation matrix and constructs a new matrix by utilizing the conjugate symmetry property.The new matrix's column rank has no relationship with sources' relevance.In addition,the method constructs a polynomial with the new orthogonal matrix to reduce the computational complexity.Simulation results show that the effective array aperture is extended with higher performance compared with some existing de-correlation algorithms.