Long range wide area network for agricultural wireless underground sensor networks

Abstract

Wireless underground sensor networks (WUSNs) enable large-scale agricultural monitoring for improving farming efficiency and reducing pollution. A WUSN system based on the long range wide area network (LoRaWAN) standard is proposed. A novel LoRaWAN-based simulator is developed to model wireless signal attenuation and path loss in an underground environment by incorporating the Peplinski and modified Friis models. The simulator incorporates the full network stack of the LoRa physical and MAC layers. Simulation results show LoRaWAN-based WUSNs (with a node burial depth of 50 cm) can maintain network connectivity with a range of over several kilometres. The simulation results also show the regional duty cycle restriction significantly reduces network scalability due to acknowledgements from end-devices. For agricultural applications where such frequent acknowledgements are not required, the results show a LoRaWAN WUSN is scalable. A field experiment to evaluate the accuracy of the theoretical path loss model was conducted and results were found to agree with the simulations.