Robust Differential Beamformer Design and Error Performance Analysis for Small-Aperture Vector Sensor Arrays
Abstract
Conventional differential beamforming algorithms inevitably sacrifice white-noise gain (WNG) to achieve high directivity, resulting in degraded robustness. To address this issue, this article proposes a robust differential beamforming method for small-aperture acoustic vector sensor (AVS) arrays based on Kronecker-product decomposition. The proposed method decomposes the steering vector into the Kronecker product of the scalar pressure-array (SPA) steering vector and the steering vector of a single AVS, enabling a maximum WNG ideal beamformer design. A unified higher order analytical solution for weight vectors is derived via least squares and Maclaurin series expansion. The main contributions are threefold: First, a robust differential beamforming technique that overcomes white-noise amplification while maintaining the directivity index; second, a unified higher order closed-form solution for arbitrary-order beamformer design; and third, an error analysis reveals that Maclaurin truncation is the primary source of beampattern and directivity factor errors, which increase with frequency, spacing, and sensor count. Simulation results show a 4.8-dB WNG improvement over conventional SPA-based methods, with superior WNG performance compared with other AVS array algorithms.