Wireless Link Quality Prediction in IoT Networks

Abstract

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