IEEE Signal Processing Letters · 2004 · 33 citations · 4 references
ComputingComputer EngineeringComputer ScienceComputational AcousticsDigital CommunicationMachine LearningDigital Signal ProcessingKalman FilteringAcoustic Echo SuppressionAdaptive FilteringEcho CancellationNoise Cancellation
The letter presents a novel method for network echo cancellation, based on a combination of normalized least mean square (NLMS) and proportionate NLMS (PNLMS) adaptive filtering algorithms. First, based on a rough analysis of PNLMS adaptation, it is indicated why, after PNLMS initial fast convergence, it slows down. Then, the method used to overcome this deficiency is presented. Last, by showing some of the simulations, its improvement over the PNLMS algorithm is shown.
4