Multi-Source Direction-of-Arrival Estimation Using Improved Estimation Consistency Method

TL;DR

This paper proposes improved estimation consistency methods for multi-source DOA estimation in reverberant and noisy environments using spherical microphone arrays, enhancing robustness and accuracy over existing approaches.

Contribution

The paper introduces modified parametric weights to enhance the estimation consistency approach for better multi-source DOA estimation in challenging acoustic conditions.

Findings

Superior performance over existing EC methods

Effective in reverberant and noisy environments

Handles closely spaced sources better

Abstract

We address the problem of estimating direction-of-arrivals (DOAs) for multiple acoustic sources in a reverberant environment using a spherical microphone array. It is well-known that multi-source DOA estimation is challenging in the presence of room reverberation, environmental noise and overlapping sources. In this work, we introduce multiple schemes to improve the robustness of estimation consistency (EC) approach in reverberant and noisy conditions through redefined and modified parametric weights. Simulation results show that our proposed methods achieve superior performance compared to the existing EC approach, especially when the sources are spatially close in a reverberant environment.