A Sound Source Identification Algorithm Based on Bayesian Compressive Sensing and Equivalent Source Method.

Area of Science:

• Acoustics
• Signal Processing
• Computational Physics

Background:

• Near-field acoustic holography (NAH) using the equivalent source method (ESM) is vital for sound source identification.
• Conventional ESM struggles with high frequencies, while wideband acoustic holography (WBH) shows limitations with coherent sources at low frequencies.

Purpose of the Study:

• To develop an improved sound source identification algorithm that overcomes the limitations of conventional ESM and WBH.
• To enhance reconstruction accuracy across a broader frequency spectrum, particularly for coherent sound sources.

Main Methods:

• Proposed a novel algorithm integrating Bayesian compressive sensing (BCS) with the equivalent source method (ESM).
• Employed a hierarchical Laplace sparse prior probability distribution and adaptive regularization parameter adjustment.
• Incorporated a function beamforming concept with order 'v' to enhance dynamic range and positional accuracy.

Main Results:

• The BCS-ESM method demonstrated superior accuracy and wider frequency applicability compared to conventional ESM and WBH in simulations.
• Achieved significantly better reconstruction performance for coherent sound sources, with accurate localization possible by increasing order 'v'.
• Experimental validation confirmed the method's stability and reliability.

Conclusions:

• The proposed BCS-ESM algorithm offers a more robust and accurate solution for sound source identification across various frequencies.
• The technique effectively addresses the challenges posed by coherent sources, providing precise localization capabilities.
• The method's performance and reliability are validated through both simulations and experimental studies.