Hemispherical Harmonics based Acoustic Source Localization in Mixed-field
Deepika Kumari, Vibhu Kailkhura, Lalan Kumar · 2019
In this paper, a unified data model is presented for mixed-field (MF) source localization using hemispherical microphone array (HMA). Mode strength based data model for MF source localization is formulated. The proposed MF mode strength accounts for both near-field (NF) and far-field (FF) sources. As the data is only present over hemisphere, rather than spherical harmonics (SH), hemispherical harmonics (HSH) basis functions have been utilized. The HSH basis functions provide a more accurate representation of the data over the hemisphere. Optimum source localization algorithms such as multiple signal classification (MUSIC) and minimum variance distortionless response (MVDR) are formulated in HSH domain to estimate the range and bearing for the MF sources. Root mean square error (RMSE) and probability of resolution measures at various signal to noise ratio (SNR) are provided to evaluate the proposed algorithms.