Uma solução de nó sensor sem fio para monitoração de anfíbios anuros e seu habitat

• Ano de defesa: 2011
• Autor(a) principal: Costa, Carlos Fransley Scatambulo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual de Maringá; Brasil; Programa de Pós-Graduação em Ciência da Computação; UEM; Maringá, PR; Departamento de Informática
• Palavras-chave em Português: Rede de sensores sem fio; Mana-m; Anuro; Multimídia; Acústica; Identificação; Arquitetura; Comunicação; Computação autonômica; Nó sensor .Vocalização; Monitoração ambiental; Anfíbuos anuros; Computação ubíqua; Wireless sensor networks; Brasil.; Amphibian; Vocalization; Sensor node; Multimedia; Acoustic; Identification; Environmental monitoring; Brazil.; Ciências Exatas e da Terra; Ciência da Computação
• Link de acesso: http://repositorio.uem.br:8080/jspui/handle/1/2587

The challenges encountered by researchers in the acquisition of environmental data and monitoring and identification of species in living environments is, as a rule, a long and expensive process. This stimulates the development of technologies such as Wireless Sensor Network (WSN), and in particular of Wireless Multimedia Sensor Networks (WMSN). These networks are important for this work which aims to develop a solution for wireless sensor node and a WMSN applied to the environmental monitoring and animal acoustic recognition, particularly monitoring of amphibians. These animals are highly sensitive to changes in their natural habitats resembling the natural sensors for evidence of environmental degradation. As proof of concept, a prototype consisting starting at commercial devices is proposed. The network will cover from the collection of data (environmental and species), data processing and visualization to the availability of this information in an Internet site. Aspects such as the optimum utilization of resources will be considered in the design of the network, due to their severe resource constraints and energy on the part of the network elements, because pretend itself to leave this network operating without human supervision for long periods, of months to years. As an example, among the hardware devices, was necessary a subdivision of tasks: environmental sensing and processing of acoustic data. Results showed a rate of up to 100% and 50% accuracy in identifying the amphibian for samples of vocalization without noise and others with noise interference of rain, respectively.