Comparative and experimental research on robust adaptive beamforming algorithm
Lin Guan-chen · Journal of Northwest University · 2015
Traditional adaptive beamforming algorithms can be extremely sensitive to slight errors in array characteristics. In order to improve the robustness of adaptive beamforming techniques against array imperfections,after discussing the improvement on robustness of beamformer with the traditional method of diagonal loading based on analysis of linear constrained minimum variance beamformer,the Support Vector Machine( SVM) algorithm which is based on structural risk minimization principle is applied to robust beamforming.Finally,it analyses and compares the robust performance of diagonal loading algorithm and SVM-based algorithm. Experimental results in pool show that the method of diagonal loading can reduce the diffusion of eigenvalues corresponding to noise in the covariance matrix and improve the robustness of beamformer against steer vector error and limited number of snapshots. And the new SVM-based beamformer performs better than the other beamformers for high Signal-Noise Ratio( SNR) values or with numerous interferences signals.