Enhancing Agricultural Efficiency through an IoT-Based Soil Moisture Monitoring Network Utilizing LoRaWAN and Edge Computing

This study explores an IoT soil moisture monitoring network designed to improve agricultural efficiency and sustainability. The system integrates LoRaWAN-enabled soil moisture and temperature sensors, strategically deployed across agricultural fields, with a Raspberry Pi 4 gateway that processes and transmits data to the cloud. The combination of low-power, long-range communication and dual connectivity options—Wi-Fi and LTE 4G—ensures reliable operation even in remote areas, making the system ideal for large-scale agricultural monitoring.

The core of the network is a robust edge processing framework that enhances data accuracy, security, and efficiency. The framework begins with noise filtering, using techniques such as median filtering to remove anomalies from raw data. Once filtered, the data is aggregated over specific time periods to reduce transmission bandwidth and provide actionable summaries of soil conditions. Adaptive data rate adjustments further optimize resource use by increasing data collection frequency during significant environmental changes and reducing it during periods of stability.

Data security is ensured through encryption at the edge, protecting sensitive environmental information from unauthorized access. Local processing also supports predictive analytics, using models like decision trees or linear regression to forecast future soil moisture and temperature conditions based on historical trends. These forecasts enable proactive decision-making, such as adjusting irrigation schedules to maintain optimal soil moisture levels, improving resource efficiency and crop health.

Anomaly detection is another critical component of the system, identifying unusual patterns in sensor readings that could indicate malfunctions or unexpected environmental changes. This ensures data integrity by flagging or excluding erroneous data. In addition, real-time event-driven alerts notify users of critical thresholds, such as dangerously low soil moisture or rapid temperature changes, allowing for immediate interventions. Alerts are delivered through SMS, email, or cloud dashboards for maximum accessibility and responsiveness.

The system's scalability supports the seamless addition of sensors, accommodating expanding agricultural operations without significant modifications. Local data logging provides redundancy, preserving raw and processed data even during network outages. This ensures uninterrupted monitoring and allows for post-event analysis, enhancing reliability and resilience.

The network’s design offers substantial benefits for agriculture. Adaptive resource management conserves bandwidth, power, and computational resources, reducing operational costs while extending system lifespan. By combining edge processing with cloud analytics, the system provides timely and actionable insights, empowering farmers to make data-driven decisions. Enhanced security through encryption protects sensitive data, while predictive analytics and anomaly detection ensure proactive and accurate responses to changing field conditions.

Overall, the IoT soil moisture monitoring network is a robust and efficient solution for modern agriculture. It enhances real-time monitoring, decision-making, and resource management, enabling farmers to optimize irrigation, improve crop health, and boost productivity. The system's scalability and adaptability make it a practical choice for addressing the growing demands of precision agriculture, contributing to sustainable farming practices and improved food security.

Acknowledgement:

This research has been funded by European Union’s Horizon Europe research and innovation programme under ScaleAgData project (Grant Agreement No. 101086355).