Study On Spherical Beamforming Sound Source Identification Technique And Its Application

With the development of the vehicle technology, the vehicle interior noise getsmore and more attention. Spherical microphone array has recently become an importantresearch subject because its special spherical symmetry is very convenient foridentifying simultaneously noise sources from all directions in three-dimensional space,and separating the incident noise and reflective noise, and it can complete themeasurement and identification fast.This paper studied on the spherical beamforming algorithm for identifying acousticsources by spherical microphone array. It identified the acoustic sources by a set of‘decomposition’ and ‘combination’ of spherical harmonics for the sound pressuremeasured by the microphones on the sphere, and based on spherical Fouriertransformation and scattering theory. Firstly, Spherical Harmonics Beamforming (SHB)for plane wave was studied, and the spatial resolution, errors and white noise gain wereanalyzed. And on this basis, Spherical Harmonics Angularly Resolved Pressure(SHARP) for a single monopole point source was derived. Then, performancesimulations of noise source identification were carried out by MATLAB for the rigidspherical microphone array with36and50channels. The results showed that thetruncated length of the spherical harmonics, frequencies and positions of the noisesources had significant effects on the output of arrays, and the best truncated lengthvalue was related to the frequencies, radius of the sphere, positions of microphones, etc.Furthermore, software for analyzing the dynamic performance of spherical array andimaging the acoustic sources was designed. Lastly, the test example imaging of thegiven sources using the software proves that the algorithm and the software is accurate.The noise source identification for a car was carried out by36channels sphericalmicrophone array, and the positions of major noise sources in a given condition wereidentified.