Spherical Sector Harmonics Based Acoustic Source Localization Using Circular Array

Abstract

Spherical harmonics (SH) domain processing for source localization provides ease of array processing, reduced dimensionality and higher resolution. However, it requires microphones mounted over spherical structure. Attempts have been made in literature to exploit the SH processing benefits for circular microphone arrays (CMA) by utilizing possibly multiple CMAs and an additional supporting sphere. In this work, CMA is conceived as a special case of spherical sector microphone array. Hence, spherical sector harmonics (S<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>H) become the default choice of basis function to decompose the pressure over CMA. This results in accurate sound field decomposition in the three-dimensional space of the array. Multiple signal classification (MUSIC) and minimum variance distortionless response (MVDR) spectra are reformulated for multiple source localization over CMA in S<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>H domain. The performance is analyzed using various experiments on source localization.