Abstract

This article presents a novel adaptive algorithm that combines the Normalized Subband Adaptive Filter (NSAF-NLMS) with a Data-Reuse (DR) strategy to address the challenges of Acoustic Echo Cancellation (AEC) and System Identification. These tasks are critical in modern communication systems, where effective echo suppression and accurate modeling of acoustic environments are essential for improving overall communication quality. Unlike conventional approaches that apply the data-reuse technique in full-band adaptive signal processing, the proposed method introduces data reuse within the subband processing framework. Specifically, the reuse of data is performed independently within each subband, enabling more efficient adaptation. Extensive simulations conducted under various signal and environmental conditions validate the performance of the proposed algorithm. The results demonstrate that the algorithm achieves notable improvements in convergence speed, misalignment, and tracking accuracy. In particular, the proposed method consistently outperforms the conventional Normalized least mean square (NLMS), DR-NLMS and NSAF-NLMS algorithms in dynamic and noisy scenarios, confirming its robustness and effectiveness.

Key words: Adaptive filtering, Acoustic echo cancellation, Subband Adaptive Filter, NLMS, NSAF, Data Reuse.