Research On Robust Adaptive Beamforming Technology Based On Uniform Linear Array

With the rapid development of modern electronic information technology,adaptive beamforming plays an indispensable role in signal processing of array sensor networks.However,in complex and changeable application scenarios,due to the mismatch between the real working environment and the ideal electromagnetic signal model,there will be mismatch errors in the signal steering vector and the estimated covariance matrix,which will seriously affect the output performance of the beamformer.Therefore,this thesis will study how to improve the robustness of the adaptive beamforming algorithm.The research work on robust adaptive beamforming algorithms in this thesis mainly includes:(1)Most robust adaptive beamforming algorithms are non-convex optimization problems which cananot be solved directly,because of the limitation of the nonconvex constraint conditions.It is necessary to introduce the relaxation condition approximation to solve the convex optimization problem.However,the approximation errors will inevitably affect the output performance of the beamformer.Aiming at the approximation errors introduced in the solving process of existing algorithms,this thesis proposes a robust adaptive beamforming algorithm based on biconvex optimization.The algorithm redefines the amplitude response constraint based on prior knowledge to accurately control the angle error range to ensure the desired signal gain and perform global interference suppression,and then introduces auxiliary variables and dual variables to directly model the robust beamforming problem as a biconvex optimization problem.The introduction of approximate errors is avoided,and the performance of the beamformer is improved.(2)At present,most algorithms directly modify the expected signal characteristic information based on the echo data,which depends on the data itself.However,when the echo data is seriously distorted,the correction ability of the beamformer will not reach the expected effect,which will reduce the output performance of the beamformer.In order to avoid processing the echo data directly,a robust adaptive beamforming algorithm with sparse constraints is proposed based on the characteristics of filters.In this algorithm,the optimal filter weight vector is analyzed,which is decomposed into matched filter and narrow-band filter.Then,the sparsity of the narrowband filter is used to establish the optimization problem.The algorithm has not only better robustness,but also the nature of rapid convergence under mismatching error.(3)Adaptive beamforming by adjusting the power amplifier and antenna array phase modulator of signal amplitude and phase weighted with directional antenna pattern formation.However,in the process of weighted signal power may be limited by loss led to the decrease of the ability of detection range,and frequent adjustment will cause great loss to hardware devices,system processing time will increase.Therefore,the thesis presents a new beamforming algorithm with only phase change.Given the arrival angle deviation of interferences in practice,this algorithm maximizes the ratio of the filtered desired signal and the signal within the interference error angle range under the constant modulus constraint to achieve null broadening.In the solving process,using the decomposability of constraint condition,the high-dimensional problem is transformed into several one-dimensional problems.Then the coordinate descent method is used to solve iteratively the robust beamforming problem to obtain the optimal filtering weight vector.The simulation results show that the algorithm has shorter calculation time under the condition of constant modulus constraint,and can form a wide and deep null at the interference.