Bayesian Sound Field Estimation Using Moving Microphones

Abstract

A sound field estimation method using moving microphones is proposed. The sound field is modelled with an infinite sequence of spherical harmonic coefficients for each frequency in the discrete Fourier transform domain, and the harmonic coefficients are estimated using Bayesian inference. The proposed method allows for the use of microphones with arbitrary directivities and moving along any trajectory. It can also be adapted for simultaneous measurement with multiple loudspeakers. The approach avoids truncation of the harmonic coefficients, as well as the specification of an expansion center. The proposed method is evaluated through simulations and compared against other state-of-the-art methods. The proposed method is shown to produce estimates with low error, and can be applied in a wider range of situations compared to other methods.