An IRWLS procedure for robust beamforming with sidelobe control

Recently, the application of support vector machines (SVM) has been proposed for increasing the robustness of array beamforming against signal mismatch situations and providing additional sidelobe control. In this approach, the conventional Capon cost function is modified by including a regularization term that penalizes differences between the actual and the ideal array responses. The resulting cost function is convex with a unique global minimum that can be found using quadratic programming (QP) techniques. This paper expands this approach in order to reduce the computational cost of the SVM solution by means of an iterative re-weighted least squares (IRWLS) procedure. The robustness of the proposed beam former is examined in several mismatch scenarios, showing a good performance even when the signal DOA mismatch error is underestimated: a situation where other robust approaches typically fail.