A new simplified fast transversal filter algorithm based on subband approach (SSFTF) for acoustic echo cancellation

Bessekri, A.; Djendi, Mohamed; Guessoum, Abderrezak

doi:10.1016/j.apacoust.2019.107178

Cited by 4 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Audio‐conferencing is a crucial application that demands adaptive filters with large numbers of coefficients and fast convergence rate in order to properly avoid perceptible acoustic echo. These requirements are stringent even for state‐of‐the‐art adaptive filtering (AF) algorithms [1]. It is widely known that colored input signals (which are common in practice) reduce the convergence rate of AFs, and subband adaptive filtering (SAF) algorithms have been proposed to mitigate such issue.…”

Section: Introductionmentioning

confidence: 99%

A non‐uniform NSAF‐SF adaptive algorithm

Xavier

Petraglia

Haddad

2021

Electronics Letters

View full text Add to dashboard Cite

Speech enhancement and acoustic noise reduction are two important tasks where adaptive filtering algorithms emerge as a competitive solution. Unfortunately, in such applications the convergence rate of the system identification is hampered when the excitation data is highly correlated. Subband adaptive algorithms have been developed to address such issue. A recently proposed low‐complexity subband adaptive structure with sparse subfilters is generalised, in order to permit a non‐uniform filter bank structure. This generalisation brings additional flexibility to the resulting critically decimated adaptive structure, which allows one to adapt it to the idiosyncrasies of the application of interest without losing the perfect‐reconstruction property. A closed‐form solution for the optimal values that the adaptive coefficient should assume in order to accurately emulate a given impulse response is derived. Simulations reveal that the resulting algorithm may outperform recently published subband adaptive filtering algorithms, thereby requiring even less computational effort.

show abstract

Section: Introductionmentioning

confidence: 99%