Emergent Mind

Abstract

Limited by fixed step-size and sparsity penalty factor, the conventional sparsity-aware normalized subband adaptive filtering (NSAF) type algorithms suffer from trade-off requirements of high filtering accurateness and quicker convergence behavior for sparse system identification. To deal with this problem, this paper proposes variable step-size L0-norm constraint NSAF algorithms (VSS-L0-NSAFs). We first analyze mean-square-deviation (MSD) statistics behavior of the L0-NSAF innovatively in according to novel weight recursion form and arrive at corresponding expressions for the cases that background noise variance is available and unavailable, where correlation degree of system input is indicated by scaling parameter r. Based on derivations, we develop an effective variable step-size scheme through minimizing the upper bounds of the MSD under some reasonable assumptions and lemma. Furthermore, an effective reset strategy is incorporated into presented algorithms to tackle with non-stationary situations. Finally, numerical simulations corroborate that the proposed algorithms achieve better performance in terms of estimation accurateness and tracking capability in comparison with existing related algorithms in sparse system identification and adaptive echo cancellation circumstances.

We're not able to analyze this paper right now due to high demand.

Please check back later (sorry!).

Generate a summary of this paper on our Pro plan:

We ran into a problem analyzing this paper.

Newsletter

Get summaries of trending comp sci papers delivered straight to your inbox:

Unsubscribe anytime.