Robust Adaptive Beamforming Algorithm Research In Array Signal Processing

Array signal processing has the advantage of flexible beam control, high signal gain and strong interference suppression, Adaptive beamforming is an important research direction in array signal processing, by adjusting the weights adaptively to make the direction of the main lobe aim at the desired signal and zero align points to the direction of interference,which can improve the output SNR. In practice, Because of the steering vector error due to discrete scanning interval and the array elements, and, covariance matrix estimation error caused by a limited number of snapshots, Therefor, The research to robust beamforming algorithm which is error robustness has important significance.In this thesis, First, We build the mathematical model of array signal processing, Introduced the theory of the adaptive beamforming algorithm,Then analyses the three optical beamforming criterion (MMSE, MSINR, LCMV) and several classical adaptive algorithms (LMS, RLS, GSC). In order to overcome the shortcomings of classical adaptive algorithm sensitive to model error, the paper introduces robust adaptive algorithm which still ensure good output performance under model mismatch, The classical robust adaptive algorithm has LSMI, ESB, RCB, Then analyze the performance advantages and disadvantages.When there are desired signal component in the sampling snapshots data, Conventional algorithms performance degradation, This paper proposes two robust adaptive algorithm based on matrix reconstructing, the first algorithm uses the idea of joint algorithm, Though the method of Music to reconstruct the interference noise covariance matrix which removals desired signal component in the sampling matrix, and then by solving the optimization problem to compensate the desired signal steering vector, MATLAB Simulation results show that the algorithm has a strong robustness in a low snapshot condition, Meanwhile The algorithm enhanced interference suppression ability. Another algorithm is the improvements of orthogonal projection algorithm (OP) which uses the reconstruction of the interference-noise association covariance matrix to the orthogonal projection (OP)algorithm, MATLAB simulation results show that the improved algorithm can solve the problem of original OP algorithm’s signal cancellation when the sample data contains the desired signal component, and reduce the impact of noise disturbance on algorithm performance, enhanced the ability of interference suppression.