English
Ciência_Iscte
Publicações
Descrição Detalhada da Publicação
Prediction of link quality for IoT cloud communications supported by machine learning
2021 IEEE World AI IoT Congress (AIIoT)
Ano (publicação definitiva)
2021
Língua
Inglês
País
Estados Unidos da América
Mais Informação
Web of Science®
Scopus
Google Scholar
Esta publicação não está indexada no Overton
Abstract/Resumo
This paper introduces a study done to evaluate the use of machine learning regression techniques to predict the link quality of communications done by IoT nodes. The proposed methodology is able to predict the link quality of the most typical cloud communication protocols, such as cellular, Wi-Fi, SigFox and LoRaWAN, based on the node location. To discover the best model to achieve this, a set of machine learning techniques were implemented, including Linear Regression, Decision Tree, Random Forest and Neural Networks, being the results compared. Results showed that Decisions Trees achieve the best efficiency, with a margin of error of 7.172 dBm, after cross-validation. This paper includes a detailed description of the methodology, its implementation and the experimental results.
Agradecimentos/Acknowledgements
--
Palavras-chave
Machine learning,Wireless communications,Internet of Things,Regressions
Registos de financiamentos
| Referência de financiamento | Entidade Financiadora |
|---|---|
| ISCTE-IUL-ISTA-BM-2018 | Iscte - Instituto Universitário de Lisboa |
