Robust adaptive beamforming using partial least squares

The goal of this paper is to present an iterative robust adaptive beamformer using partial least squares (PLS) based on the generalized sidelobe canceller (GSC) parameterization of the minimum variance (MV) beamformer. The proposed method can be implemented simply as an iterative process and does not require the choice of a user parameter. The shrinkage properties and the relations among ridge regression (RR) Capon beamformer, principal component regression (PCR) MV beamformer and our proposed method are discussed. The performance of our approach is illustrated by numerical simulations and compared to other robust adaptive beamformers.