Robust Adaptive Beamforming Exploiting Coprime Virtual Iterative Adaptive Approach

Abstract

Coprime array can augment the array aperture and enhance the interferences suppression capability for array processing. In this paper, we propose a coprime virtual iterative adaptive approach (CVIAA) method to achieve beamforming performance improvement. Existing coprime array adaptive beamformers (CAABs) usually select the maximum virtual uniform linear array to estimate parameters with the non-uniform virtual sensors deleted, which will suffer from performance loss because some information is ignored. The proposed CVIAA method can cope with the non-uniformity of virtual coarray where entire data is preserved. Even though the virtual array signal is a single snapshot, the proposed CVIAA method can precisely estimate the powers of source signals and noise variance in an iterative way. Moreover, the proposed CVIAA estimator outperforms the existing virtual array Capon estimator when the source signals are closely located. Unlike the traditional iterative adaptive approach (IAA), the proposed CVIAA method reduces the computational complexity by dividing the data into several low-dimension subsets and estimates the noise variance iteratively. Finally, we calculate the weight vector of the proposed CAAB from its definition. Computer simulations verify that the proposed CVIAA estimator is superior to existing estimators and the proposed CAAB can handle environmental uncertainties.