Search papers, labs, and topics across Lattice.
This paper introduces a geometrically constrained decentralized independent vector analysis (GC-Dec-IVA) method designed for distributed microphone arrays, addressing limitations in the previously proposed Dec-IVA method. By incorporating direction-of-arrival (DOA) information, GC-Dec-IVA effectively mitigates permutation mismatches and enhances source alignment across arrays, which is critical for accurate source separation. Experimental results demonstrate significant improvements in both separation performance and cross-array permutation consistency, particularly in noisy environments.
Incorporating direction-of-arrival information, GC-Dec-IVA significantly enhances source separation in distributed microphone arrays, overcoming critical limitations of previous methods.
This paper proposes a geometrically constrained decentralized independent vector analysis (GC-Dec-IVA) method for distributed microphone arrays. Recently proposed Dec-IVA method enables source separation by exchanging only power-related statistics to exploit cross-array information. However, this initial attempt often provides negligible improvement over applying IVA locally at each array, mainly due to the potential permutation inconsistency among arrays and the strong cross-array dependency implied by its source model. To address these limitations, we incorporate direction-of-arrival (DOA) information to derive GC-Dec-IVA, which mitigates permutation mismatch across arrays and enhances source alignment. Furthermore, a new source model is introduced to weaken cross-array dependency, improving robustness against permutation inconsistency in noisy environments. Experiments show the proposed method improves both the separation performance and cross-array permutation consistency.
