AI-Driven Optimization and Performance Analysis of LoRaWAN Network Deployment in Campus Non-Line-of-Sight (NLOS) Environment

Abstract

LoRaWAN is widely used in low-power wide-area network (LPWAN) applications. However, network planning in a campus environment is challenging because of non-line-of-sight (NLOS) conditions due to buildings and obstacles. Conventional propagation models are not always capable of modeling the irregular behavior of wireless signals in such environments. In this research, the LoRaWAN signal performance in NLOS scenarios is predicted using a machine learning based approach. Field measurement data were collected by drive testing in a university campus. K-Nearest Neighbors (KNN) and Random Forest (RF) algorithms used to predict received signal strength indicator (RSSI) and signal to noise ratio (SNR). The results demonstrate the ability of the machine learning models to successfully learn real world signal variations in NLOS conditions. This approach offers a more practical way to enhance LoRaWAN network planning and deployment in campus environments.