-
ANTÔNIO HENRIQUE DOS SANTOS RIBEIRO
Universidade Federal do Maranhão, São Luís, MA, Brasil, Departamento de Engenharia de Eletricidade, Laboratório de Controle Inteligente e Sistemas Embarcados
-
JOÃO VIANA DA FONSECA NETO
Universidade Federal do Maranhão, São Luís, MA, Brasil, Departamento de Engenharia de Eletricidade, Laboratório de Controle Inteligente e Sistemas Embarcados
-
FRANCISCO DAS CHAGAS DE SOUZA
Universidade Federal do Maranhão, São Luís, MA, Brasil, Departamento de Engenharia de Eletricidade, Laboratório de Sistemas Adaptativos e Processamento de Sinais
Keywords:
Affine combination, adaptive filters, beamforming, smart antennas, neural networks, adaptive algorithm RLS- LMS
Abstract
In this article, a Feed Forward Artificial Neural Network is applied to the RLS-LMS adaptive algorithm, with the objective of tuning the weight parameters, calculating the ideal or optimal weights used in the signal input of the linear filters that adapt the pattern of uniform linear array antenna radiation, directing multiple narrow beams to desired users and nullifying interference or unwanted users. The application of this neural network provides the ability to increase the efficiency and optimize the use of smart antennas. The results obtained are presented. This article is based on the work of (Ribeiro et al, 2019) taking the results obtained in the use of the combination of two adaptive filters LMS (Least mean- square) and RLS (recursive least-square) combined to obtain an adaptive algorithm RLS- LMS that surpassed classical algorithms in terms of convergence speed and stability.
