Data Fusion in Internet of Things

• Autor(a) principal: Mendes, Tiago Nobre de Albuquerque
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.6/10222
• topic: Data Fusion; Deep Learning; Health-Care; Human Activity Recognition; Internet of Things; Neural Networks; Smart Home; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

This dissertation reviews Internet of Things concepts and implementations, state-of-the-art technology with practical examples, as well as data fusion methods applied in different problems. The purpose of this study is to review different data fusion methods and develop a system to provide recognition of human activity that can be applied in day care homes and in hospitals to monitor patients. The system’s objective is to study human activity recognition based on the data recovered by sensors like accelerometers and gyroscopes. In order to transform this data to useful information and practical results to monitoring patients with accuracy and high performance, two different neural networks were implemented. To conclude, the results from the two different neural networks are compared to each other and compared with systems from other authors. It is hoped this study will inform other authors and developers about the performance of neural networks when managing human activity recognition systems.