STOCHASTIC MODELLING OF THE TRANSFORM DOMAIN EPSILON-LMS ALGORITHM FOR A TIME-VARYING ENVIRONMENT (MonAmOR1)

Author(s) : Elen Lobato (Federal University of Santa Catarina, Brazil) Orlando Tobias (Federal University of Santa Catarina, Brazil) Rui Seara (Federal University of Santa Catarina, Brazil)

Abstract : This paper presents a stochastic model of the transform domain eLMS (eTDLMS) algorithm for a nonstationary environment. The proposed model is derived for Gaussian inputs, high-order adaptive filter and slow adaptation condition. Through simulations, we can verify a very good agreement between the results obtained by the Monte Carlo method and the predictions from the proposed analytical model.

Menu