Módulo de predição de dados visando a economia de energia em dispositivos para internet das coisas
IoT (Internet of Things) presents a myriad of solutions to problems in many areas of human knowledge. Despite its wide applicability, developing IoT applications is not an easy task due to the heterogeneity of devices and non-functional requirements such as power saving. One of the most energy-inten...
Na minha lista:
Autor principal: | |
---|---|
Outros Autores: | |
Formato: | bachelorThesis |
Idioma: | pt_BR |
Publicado em: |
Universidade Federal do Rio Grande do Norte
|
Assuntos: | |
Endereço do item: | https://repositorio.ufrn.br/handle/123456789/37921 |
Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
Resumo: | IoT (Internet of Things) presents a myriad of solutions to problems in many areas of human knowledge. Despite its wide applicability, developing IoT applications is not an easy task due to the heterogeneity of devices and non-functional requirements such as power saving. One of the most energy-intensive activities on the Internet of Things is using the wireless network to send or receive information. This is a critical activity because IoT devices often have limited power sources such as batteries. In order to solve the such problem, the objective of this work is to develop a Data Prediction Service using Machine Learning, whose function is determined or the exact moment for a Middleware to manage the IoT device work cycle through data models. Where the predicted data can be used instead of the actual device, thus never failing to provide data to the application. An evaluation was performed by establishing to quantify the efficiency of the prediction service and to analyze data prediction models with the intention of maximizing device autonomy. The execution of the application with the Prediction Service was compared to another without use. Thus, a 63% increase in usage time has been reported for power with a 9v (volts) battery. |
---|