Smart Line-of-Sight Identification in LoRaWAN Networks

A significant issue involving Internet of Things (IoT) systems is the fluctuation of a link between line-of-sight (LOS) and non-line-of-sight (NLOS) states due to time-varying environments. Discrimination between LOS and NLOS conditions is important for a variety of purposes in wireless systems, including channel modeling, transmitter design, and scheduling. This article implements machine learning techniques to classify LOS and NLOS, based on data collected on the campus of the Federal University of Juiz de Fora (UFJF) and results indicates better performance for Random Forest and K-Nearest Neighbors in this scenario.