SOLAR POWERED IN SITU SENSOR NETWORK FOR REMOTE SOIL HEALTH MONITORING

Abstract

This thesis presents the design, development, and testing of a fleet of long-range wireless, solar powered sensors for the study of soil conditions throughout a given geographical location for the measurement of important metrics that affect overall soil fertility. Farmers require active monitoring for informed maintenance on their fields to maintain optimal crop growth. Having practical, real-time data to analyze assists in determining tasks that need to be completed for an optimal crop yield. The common practice is sending soil samples off to a lab for chemical analysis. Therefore, no known system has been developed to remotely measure soil-gas, soil moisture, and other environmental variables in situ through the use of an IoT sensor network across a large farm. This thesis discusses the implementation of a real-time and historical soil health monitoring system that is used for the measurement of temperature, humidity, soil moisture, volatile organic compounds, carbon monoxide, carbon dioxide, methane, ammonia, nitrogen dioxide, hydrogen, ethanol, atmospheric pressure, and sunlight. The network functions on a gateway-bridge architecture, whereas the gateway connects to the database server via the internet using a cellular modem, and subsequent bridge nodes send telemetry to the gateway via LoRa radio protocol. This system directly communicates to the ThingsBoard open source software via the MQTT messaging protocol for the uploading of data to the SQL database, resulting in the ability to query and plot telemetry values.