Super-resolution sound field imaging with sub-space pre-processing

Spherical microphone arrays are a powerful tool for sound field analysis. In previous work, we have shown that sparse recovery can be used to arbitrarily increase the resolution of the sound field recorded by a spherical microphone array. Because these super-resolution techniques rely on the assumption that the sound field results from a few dominant plane waves, they are not robust to the presence of noise or reverberation. In this paper we propose a simple method to separate the sound field into a directional component and a diffuse component prior to applying sparse recovery techniques. Simulations show that this pre-processing could dramatically improve the results of sparse recovery in noisy or reverberant environments.