Iterative robust beamformer with an estimation of uncertainty level

Abstract

In this paper, a new design of the robust adaptive beamformer (RAB) is developed to overcome the increased computation cost of the traditional RABs. This approach introduces a suboptimal of the distance between the actual array steering vector (ASV) and presumed ASV in tandem with the iterative doubly constrained robust capon beamformer using fixed uncertainty level (Fu-IDCRCB). Then, the uncertainty level of the first iteration is updated by the optimal distance and then the estimated ASV of the first step is in the vicinity of the actual ASV. As a result, the coherent iterations are needed to search the actual ASV with smaller uncertainty set. Hence, this method convergences faster than the other beamfomers (BFs) and the simulation results demonstrate the effectiveness of the proposed BF.