LoRa flood messaging applied to remote soil-moisture monitoring

Soil-Moisture sensing has proven to be an important factor in decreasing the need for irrigation while maintaining crop yield and quality. Farmers’ finding ways to minimize water use is necessary given the world’s obvious climate change and, in some areas, the resulting onslaught of drought. As the world’s population grows and the demand for higher-quality food increases, precision irrigation becomes even more essential. As sensors are deployed, these have to be monitored in some way. Our project employs radio broadcasts over unlicensed frequencies to implement wireless remote sensing. Internet and cellular are alternative solutions. However, those require more than the acquisition and implementation of certain equipment and software. Those two approaches also involve third-party services, and their continuing fees. An additional assumption is that those services are available, cost effective, and reliable. Radio broadcast using the LoRa (Long Range) protocol is essentially a low-power means for implementing transceivers for low-rate data exchange. These are effective over longer ranges than technologies such as Bluetooth. LoRa is often deployed using LoRaWAN (LoRa for formal Wide Area Networks) but that involves additional equipment. This paper discusses a means of using only relatively low-cost hardware (Arduino-Uno microcontrollers with attached Dragino transceivers). A wireless network is formed upon which messages flow from sensors attached to microcontrollers, through relays (to account for terrain variability and other obstructions), to a receiver connected to some external system. The LoRa hardware unit remains consistent across all components, with only the installed software varying. Relays and receivers utilize their own unchanging software. However, the sensor units' baseline software is adapted based on the types and numbers of sensors attached. Each sensor, relay, and receiver has its own unique assigned network address. The authors have constructed and demonstrated a proof-of-concept of the ideas expressed in this paper. Discussion is offered on future work required as the project moves toward field application.