Year: 2025
Type: article
Abstract: ABSTRACT This paper proposes an improved fast independent vector analysis (FastIVA) algorithm. The proposed algorithm replaces the second‐order derivative with the original function and its first derivative, incorporating an adaptive scaling factor to effectively reduce the number of iterations and computation time. The experiments are conducted based on a hybrid dataset constructed from the PNL Non‐Speech Corpus and the ATR Japanese Speech Database. Experimental results demonstrate that the improved algorithm achieves separation performance comparable to the traditional FastIVA while significantly reducing both the iteration count and runtime.
Source: Electronics Letters
DOI: 10.1049/ell2.70246
Authors: Dahu Wang, Chang Liu
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,590
OpenAlex ID: W4409313063