Sobre a Modelagem Estocástica do Algoritmo NLMS em Ambientes Não Estacionários para Filtros Adaptativos e Plantas do Sistema com Ordens Diferentes

This paper presents a stochastic model describing the behavior of the normaliz ed least-mean-square (NLMS) algorithm in a context for estimating time-varying systems. In particular, the proposed model is valid for white input data with Gaussian distribution and take into account scenarios in which the order of the adaptive filter is different from that of the system to be identified. Specifically, model expressions are derived describing the mean weight behavior, the mean-square error (MSE), and the optimum step size that minimizes the steady-state MSE. Simulation results confirm the accuracy of the proposed stochastic model