Stochastic modelling of the transform domain LMS algorithm for a time-varying environment

This paper presents a stochastic model of the transform domain εLMS (εTDLMS) 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.