For soil moisture monitoring, Wageningen University trial in the Netherlands in 2023 shows that the lorawan sensor network deployed achieves the single gateway coverage of 12.5 square kilometers in the frequency band of 868MHz and offers 73% savings on base station construction cost compared to the NB-IoT solution. In SF12 (spread spectrum factor) mode, A single probe consumes just an average daily power of 48μA, enabling the CR2032 battery to power the same for 8 years and doubling the maintenance cycle. AgriTech’s 1,850 node network in Brazil’s soybean fields has reduced irrigation water consumption by 41% using adaptive Data Rate (ADR) technology. However, it must be noted that when the soil moisture is more than 40%, the attenuation rate of the signal will increase from 0.6dB/m to 2.1dB/m, and the transmission power should be dynamically adjusted from 14dBm to 20dBm.
The field operation of the precision irrigation system has validated the scalability of lorawan. The 3,200 nodes deployed in the Israeli smart farm of Netafim manage an 8-square-kilometer area with a single gateway, and save 230 cubic meters of water per hectare. The chance of collision among data packets is no more than 0.7% when the node density is 0.1 nodes per hectare. But in conditions of humidity higher than 85% during the rainy season, the bit error rate will increase from 0.05% to 1.2%. According to the FAO report, this solution increases corn yield by 19%, however, it is required to combine it with an edge computing gateway in order to decrease the latency of data processing from 12 seconds to 0.8 seconds for the purpose of real-time control.
In the example of livestock tracking, the lorawan ear tags employed by the Australian Livestock Bureau to track 38,000 Angus cows have a 10-kilometer range with a 125kHz bandwidth and cost $2.3 per year for operating and maintenance for a single device. In environmental conditions of -20℃ to 45℃, the equipment’s battery life is 5.8 years, and the standard deviation of the positioning error is controlled at ±82 meters. However, based on ABI Research, if cattle density is more than 200 per square kilometer, data packet loss ratio will rise from 3% to 21%, and a multi-gateway timeslot receiving mechanism needs to be configured.
The economic assessment of the pest and disease early warning system shows that the lorawan trap network adopted in Punjab, India, has achieved 98% detection of cotton bollworm at a device density of 0.5 devices per hectare. Utilizing the FEC forward error correction technology, it sustained a data integrity rate of 92% even in interference due to 30dB wind noise, and reduced the pesticide intake by 37%. But as per the ISRO tests, the false alarm rate for the equipment in sandstorm conditions (visibility <50 meters) will increase from 0.8% to 6.7%, and satellite remote sensing information have to be integrated to optimize the model.
The economic benefits of greenhouse automation control are significant. Westland tomato farm in the Netherlands connects 12,000 environmental sensors via lorawan to a single gateway having the capability to support 600 nodes and reduces the deployment cost per square meter to 0.47 euros. With the TSCH (Time Slot Frequency Hopping) protocol, data arrival remains at 99.3% in 98% humid environment, reducing energy consumption by 280,000 kilowatt-hours annually. But the Delphy consulting report warns that if more than 65% of the frames are greenhouse metal, then the median signal strength will drop from -87 DBM to -102 DBM, and relay nodes should be inserted every 50 meters.