Robust DOA Estimation Based On Sparse Respresentation

Array signal processing is an important branch of signal processing field. It uses the array which composed by sensors distributed in different positions in space to receive and process the spatial signal. The two main research direction in array signal processing are the direction of arrival estimation and the adaptive beamforming. By processing the signals received by array, direction-of-arrival estimation get the position information of the spatial signal, so as to determine the spatial position of the signal, while adaptive beamforming is to enhance the desired signal and suppress the interference signal and noise.In direction-of-arrival estimation algorithms, from the initial beam forming to the MUSIC super-resolution algorithm which break through the Rayleigh-Limit, and then to the algorithm based on sparse constraint, the resolution gradually upgrade. Although the existing algorithms have high resolution, but there are also some problems, these algorithms are based on the ideal condition, assume array channel amplitude and phase are consistent. But in the actual environment, the aging of components at the array, the signal environment and other factors will lead to the array channels magnitude phase inconsistency. This kind of inconsistency will affect the algorithm performance of positioning.It relates to the matrix inversion when solving the weight coefficients of the adaptive beamforming. The dimension of the inverse matrix is proportional to the number of channel. Actual antenna array often contain tens of thousands array ele ments. If each array element as a channel for processing, the solving calculation will be great. In order to solve the problem, we can divided the array into sub array, each sub array is used as a channel for processing, which can greatly reduce the number of processing channel, thereby reducing the amount of calculation.The main content of this paper consists of two parts. One is the direction of arrival estimation, based on the study of algorithm of DOA estimation, analysis the amplitude and phase errors and offgrid problem, to establish a new mathematical model and givesthe solving method.The simulation results shows that the proposed algorithm can effectively resolve the problems mentioned above. The second one is the adaptive beamforming, introduced the adaptive beamforming technology and some subarray partition method,and then applied the subarray partition method and subarray adaptive beamforming techniques to given diamond array, the simulation results shows that the subarray partitioning technique is effective.It reduce the amount of calculation and also ensure the adaptive performance.