ARMONIA: an Automated Remote MOnitoring system of NItrogen in Agriculture

An innovative monitoring system based on an IoT sensor network with LoRaWAN technology has been developed to monitor ammonia (NH3) and carbon dioxide (CO2) concentrations in buffalo barns. The system uses electrochemical DOL 53 [1] sensors for ammonia detection and DOL 119 [2] sensors for CO2 measurement, selected for their precision, robustness, and ability to operate in challenging agricultural environments. The sensors, powered at 24V, were also chosen for their ceiling-mounting capability, a strategic position where greenhouse gases tend to accumulate, allowing for more accurate concentration monitoring. These sensors have an energy consumption of approximately 2-4W, ensuring continuous and reliable measurements. The IoT nodes were designed using USI® STM32™ Nucleo expansion boards for LoRa™ (I-NUCLEOLRWAN1), integrating an STM32L052T8Y6 microcontroller based on Cortex®-M0+ and a Semtech SX1272 radio transceiver. This configuration enables efficient long-range data transmission with reduced energy consumption. The STM32L0 microcontroller consumes about 3 mA in active mode, while the SX1272 module uses around 40 mA during transmission. Both components consume approximately 1 μA in standby mode. This setup allows the node to operate with an average consumption of 0.444 W, providing an autonomy of about 33 days with a 360 Wh lithium-ion battery. The sensor network leverages LoRaWAN technology for data transmission, ensuring stable and secure communication over distances of up to 15 km in rural environments. The nodes are enclosed in IP65-rated housings, resistant to dust, humidity, and corrosive agents, which are typical conditions in barns. The data collected is sent to a central gateway and then transmitted to a custom-developed IoT platform for data processing, management, and real-time visualization. This platform enables continuous and precise environmental monitoring, quickly identifying any anomalies or threshold exceedances. The ultimate goal of the system is the creation of a “Digital Twin” of the farm, a virtual model that integrates realtime data to provide a dynamic and detailed representation of environmental conditions. This Digital Twin will, in the future, allow for the simulation of operational scenarios, resource management optimization, and the prediction of potential environmental issues, supporting more efficient and sustainable farm management. This system represents a significant improvement in environmental monitoring for buffalo barns, offering a technological solution for efficient data collection and analysis. The insights gained can be used to optimize operational and environmental management, supporting informed decision-making and promoting more sustainable agricultural practices [4]. Future research will focus on enhancing system scalability, integrating additional environmental parameters, and improving predictive analytics within the Digital Twin framework.