Algoritmo ZA-PNLMS com Fatores de Ativação Individuais e Ganhos dos Coeficientes Limitados Superiormente

Alzeneide Dutra Pessoa; Francisco das Chagas de Souza

doi:10.20906/CBA2022/3452

Alzeneide Dutra Pessoa Laboratório de Sistemas Adaptativos e Processamento de Sinais – LSAPS, Programa de Pós-graduação em Engenharia Elétrica, Universidade Federal do Maranhão – UFMA Av. dos Portugueses, 1966, CEP 65080-805, Bacanga, São Luís, MA
Francisco das Chagas de Souza Laboratório de Sistemas Adaptativos e Processamento de Sinais – LSAPS, Programa de Pós-graduação em Engenharia Elétrica, Universidade Federal do Maranhão – UFMA Av. dos Portugueses, 1966, CEP 65080-805, Bacanga, São Luís, MA

DOI: https://doi.org/10.20906/CBA2022/3452

Keywords: Activation factors, adaptive algorithm, sparse plants, upper limited gains, zero-attracted

Abstract

In this paper, a zero-attracted proportionate normalized LMS adaptive algorithm (ZA-PNLMS) is proposed, which combines individual activation factors with upper limited gains. The proposal presented here upper limits the gains of the algorithm coefficients for identifying plants with a high degree of sparseness, leading to better distribution of the adaptation energy between the coefficients. Computer simulations, considering sparse systems with disturbance and tracking, attest to the effectiveness of the proposed algorithm.