Real world multi environment analysis of LoRaWAN performance

Modern agriculture increasingly depends on real-time data to guide irrigation, fertilizer application, pest control, and crop health decisions. However, many rural farms lack reliable internet and power infrastructure, limiting continuous monitoring, making it difficult for farmers to collect the required data to make data-driven decisions to increase farm produce. LoRa and LoRaWAN provide long-range, low-power connectivity for distributed sensors. This paper evaluates LoRaWAN performance across multiple real-world environments (urban, lake, forest, and mixed urban-lake settings) using commercial off-the-shelf sensors and a custom SX1262-based node. We measure RSSI, SNR, and packet delivery success over multi-mile distances using a Dragino gateway and Raspberry Pi processing hub, and analyze how antenna configuration, transmit power, and terrain affect range and reliability. Results provide actionable guidance for designing robust LoRaWAN deployments for agricultural and environmental monitoring.
Keywords: LPWAN, LoRaWAN, Smart Agriculture, Wireless Sensor Networks, Field measurements.

Davis is a PhD student at the University of Idaho. He holds a Master of Science in Computer Science from the University of Idaho. With over 5 years of experience in applied AI in Agriculture and Automation, his work focuses heavily on building custom AI models to optimize farm produce. Davis was previously a Senior Machine Learning Engineer at a multi-national multimillion-dollar corporation, leading large-scale AI initiatives. His current work bridges industrial expertise with academic innovation to solve complex agricultural challenges.