Abstract：A novel algorithm for robust adaptive beamforming is presented in this paper to solve the problem of the mismatch errors between the actual and presumed steering vectors. By using the orthogonal component of the errors, the method can obtain the desired steering vectors successfully. Two orthogonal subspaces are constructed firstly, and then the mismatch errors are orthogonally decomposed, so the correcting process of steering vectors is transformed into an iterative quadratic convex optimization issue by using a slack variable. The proposed algorithm does not need the conditions of the norm or probability constraint of the mismatch errors. Therefore, it avoids estimating upper bound values and probability distribution modeling. Simulation results show the effectiveness of the proposed algorithm.
2010, Vol. 32 
