Analysis of direct and diffuse sound field components via sparse plane wave decomposition

Date

2025-7-4

Author

Genç, Ece Beren

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

3064
views

0
downloads

Cite This

Room acoustic quality significantly impacts overall auditory experience in enclosed spaces. The characteristics of sound propagation within rooms forms the foundation of acoustic perception and are crucial considerations in architectural acoustics, audio reproduction, and virtual reality applications. This thesis investigates advanced methods for the analysis, quantification, and decomposition of sound fields in enclosed spaces. Sound fields in enclosures comprise a combination of strongly directional components alongside a diffuse field that gradually builds up due to the multipath nature of sound propagation. We propose two complementary approaches based on spherical microphone array recordings. First, a spatio-temporal analysis method using sparse plane wave decomposition provides insights into both temporal and directional aspects of diffuse fields. It is validated through experiments with 4th-order spherical arrays. Building on this foundation, we introduce a novel direct and diffuse field decomposition framework that leverages the statistical distribution of directional data on the unit sphere. By modeling unit-magnitude vectors as mixtures of von Mises-Fisher (movMF) distributions, we employ posterior probabilities as masking mechanisms to effectively separate direct and diffuse components. A comparative analysis with state-of-the-art techniques demonstrates the method's efficacy, alongside an objective quality evaluation. Throughout the thesis, we address the challenges and inherent limitations of diffuse field analysis and decomposition, while discussing practical applications and advantages of our proposed methodologies. Finally, we outline potential improvements and future research directions.

Subject Keywords

Sound field separation, Mixing time, Spherical microphone array, Spherical harmonics, Room acoustics

URI

https://hdl.handle.net/11511/115173

Collections

Graduate School of Informatics, Thesis