Depth-Resolved Three-Dimensional Sound Source Reconstruction Using Acoustic Arrays: Imaging Performance, Localization Error, and Validation.

Beamforming signal processing based on acoustic microphone arrays is widely used for sound source localization and imaging. However, when investigating sound sources in complex environments, such as cockpit environments or on complex aircraft surfaces, planar imaging becomes less suitable. In addition, the array aperture and microphone density significantly influence imaging capability in different directions, not only in the frontal direction. Therefore, this paper extends the conventional beamforming algorithm into three dimensions, focusing on sound source localization and imaging quality along the depth direction of the array aperture. Theoretically, a cross-spectral matrix and an error function are introduced into the delay-and-sum algorithm to comparatively analyze the imaging quality. The influence of arranging microphones along the depth direction on the system's imaging performance is investigated, and an experimental validation scheme comparing a planar array with a depth-wise asymmetric array is designed. The beamforming imaging performance in the X-, Y-, and Z-directions is analyzed, including comparisons of main-lobe width, the influence of frequency, and steering-vector-induced error phenomena. Test results indicate that microphone placement along the depth affects imaging performance in the Z-direction, and that appropriate amplitude compensation in the algorithm can effectively control localization errors.