Uma solução baseada em agentes para redes sensores sem fios

Detalhes bibliográficos
Ano de defesa: 2009
Autor(a) principal: Spadoni, Igor Marçal Botamede
Orientador(a): Araujo, Regina Borges de lattes
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de São Carlos
Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
Departamento: Não Informado pela instituição
País: BR
Palavras-chave em Português:
Área do conhecimento CNPq:
Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/430
Resumo: Typically the current monitoring systems for physical environments subject to situations of emergency, such as industrial plants, petroliferous platforms, nuclear plants, etc., do not have the accuracy required to observe emergency events due to lack of information and/or accuracy about what is occurring in the physical environment during the event. Therefore, there is no way to know exactly where, how, and why some situations of emergency have occurred. On the other hand wireless sensor networks (WSN) are being used more and more to monitor systems of critical safety. In these systems the level of data reliability is critical, mainly when situations of emergency (events) that involve risks to life and assets occur. Therefore, it is essential to monitor incidents in an accurate way all through the event. Clearly in these environments the sensors flow of information must feed a system of continuous monitoring during the event. Besides, following up all phases of the incident may be used for future investigation and prevention by revealing the source of the problem. It may also help the rescue team, regarding managing and making decisions. The WSN present limited resources, such as communication, processing, and memory, which generate great challenges for designers of solutions for networks, such as protocols of routing and applications. In this Dissertation a routing protocol, which uses the mobile agents paradigm to perform data collecting and forwarding was designed, implemented, and evaluated. With the mobile agents there is no need for sensors to send its reading to a centralized node (sink) every time an abnormality is detected in the environment being monitored. The mobile agent designed as part of this work is personalized according to the needs of the application (for example, lower latency, higher delivery rate, etc). The agent is responsible for visiting the nodes of the WSN that have identified problems, collecting the data read and performing the fusion of the data of that region. The highlight of this work relies on the flexibility of the agent for changes of behavior that grant the service quality during the collecting and delivery of data even in diverse conditions.