Wireless Link Quality Prediction in IoT Networks

Miguel Landry Foko Sindjoung (1) , Pascale Minet (1)

1 EVA - Wireless Networking for Evolving & Adaptive Applications

The knowledge of link quality in IoT networks will allow a more accurate selection of wireless links to build the routes used by data gathering. Therefore, the number of retransmissions on these links is decreased, leading to a shorter end-to-end latency, a better end-to-end reliability and a larger network lifetime. In this paper, we propose to predict link quality by means of machine learning techniques applied on two metrics: RSSI and PDR. The accuracy obtained by Logistic Regression, Linear Support Vector Machine, Support Vector Machine and Random Forest classifier is obtained on the traces of a real IoT network deployed at Grenoble.

Mots clés: Machine learning; Link quality estimation; Wireless network; TSC

Domaines: Réseaux et télécommunications [cs.NI]

Miguel Landry Foko Sindjoung, Pascale Minet. Wireless Link Quality Prediction in IoT Networks. PEMWN 2019 - 8th IFIP/IEEE International Conference on Performance Evaluation and Modeling inWired andWireless Networks, Nov 2019, Paris, France. ⟨hal-02432805⟩