Two-Dimensional Off-Grid Generalized Inverse Beamforming Based on Regularization Matrix

To address the inherent limitations of grid basis mismatch associated with conventional fixed-grid beamforming techniques, this study introduces a novel two-dimensional off-grid generalized inverse beamforming approach for acoustic source identification based on a regularization matrix (OG-RMGIB). The proposed methodology establishes a novel off-grid framework based on generalized inverse beamforming (GIB) principles, effectively leveraging the inherent advantage of superior spatial resolution characteristic of generalized inverse beamforming techniques, and introducing a three-dimensional regularization matrix. The source strength expression is derived while performing a first-order Taylor expansion of the initial point transfer vector. At this point, the source strength expression contains all the information about the off-grid deviation. The source coordinates as well as the source strength quantization are then solved by alternating iterations. Numerical simulation results substantiate that the proposed method significantly enhances both the accuracy of source localization and the low-frequency identification capability in source identification applications.