Otimização do tempo de vida de uma rede de sensores sem fio baseado em autômatos celulares de aprendizagem

• Ano de defesa: 2020
• Autor(a) principal: Medeiros, Rafael Pereira de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Engenharia Elétrica; Programa de Pós-Graduação em Engenharia Elétrica; UFPB
• Palavras-chave em Português: Rede de nós sensores sem fio; Autômatos celulares de aprendizagem; Otimização do tempo de vida de redes; Wireless sensor network; Cellular learning automata; Otimization of network lifetime; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/18309

With the advent of automation, whether in industry or residences, more and more information is being generated, either in actuators or by acquisition. Thereby, increasingly sensors are being used, featuring increasingly dense networks. In the case of sensor nodes, is necessary to have usage management of them, to have better energy management. The Discrete Event Systems (DES) are presented as a solution to a better elaboration of applied logics and harnessing of sensors.Andmore,throughtheuseofCellularLearningAutomata(CLA),thereisthepossibility of elaborate intelligent groups that attends the systems requirement of coverage, in an optimized way. Therefore, in this work is presented a model for characterization of a wireless network sensor, based on discrete event systems theory and considering cellular learning automata, to the optimization of energy consumption in a wireless sensor network, through the formation of groups aimed at increasing the network’s lifetime. The model was developed, analyzed and validated using the computational tool Stateflow, in MATLAB®/Simulink®. Algorithms corresponding to the models and assemblies were developed to validate the methods. From the results obtained in simulation, was verified a decrease in average consumption of network up to 67,54 % and an increase of the network’s lifetime up to 192,86 %, in scenarios under analysis. The experimental results were performed using a wireless sensor network with five sensors, with an increase in the lifetime of 31.71 %.