Técnicas de geolocalização em redes LoRaWan como abordagem de tolerância a falhas em dispositivos de rastreamento baseado em GPS

• Autor(a) principal: Pastório, André Franscisco
• Data de Publicação: 2021
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UTFPR (da Universidade Tecnológica Federal do Paraná (RIUT))
• Texto Completo: http://repositorio.utfpr.edu.br/jspui/handle/1/25629
• topic: Sistemas de transmissão de dados; Internet das coisas; Sistema de Posicionamento Global; Data transmission systems; Internet of things; Global Positioning System; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO

The wide adaptation of the Internet of Things (IoT) concept involves increasing the reliability and availability of its applications in the event of failure. The heterogeneity of IoT components, combined with computationally limited devices integrated to the Internet, pose additional challenges to achieving fault tolerance. Smart cities comprise an IoT scenario where the city is connected. Long-range, low-power networks are used in such applications to provide connectivity. In this context, object tracking is an important functionality used, for example, to track fleet vehicles in real time over the Internet. The use of common location technologies such as GPS results in high energy consumption. Allied to that, or malfunction of a GPS device can compromise the application. This work investigates how fault tolerance in IoT can be efficiently developed through a systematic literature review. Furthermore, to mitigate a GPS failure in tracking devices, this work implements a hybrid location strategy in LoRaWAN networks. The strategy considers that in case of GPS failure, coordinates are estimated using location algorithms. The implementation is compared to a commercial tool, called LoRa Cloud, which only allows for a thousand daily requests. An environment with three gateways LoRaWAN was set up to perform an assessment. The results show errors below 151 and 126 meters in the worst case considering the implementation performed and the LoRa Cloud tool, respectively.