<p>Environmental field conditions are highly variable in three-dimensions and unsuitable to be probed by a single sensor or weather station. In PLANTENNA, a team of electronics, precision and microsystems engineers and plant and environmental scientists collaborate to develop and implement 3D-sensor networks that measure plant and environmental parameters at high resolution and low cost. A first problem we aim to tackle in the field is 3D-monitoring of fruit farms for detection and mitigation of fruit frost damage. The objectives are two-fold:</p><p>- To quantify the time-dependent effects of frost mitigation measures on the 3D temperature profile, and to determine the resulting plant-physiological response to get a better understanding of the underlying mechanisms leading to frost damage.</p><p>- To develop low-cost, low power, wireless, distributed sensor networks, with automated mathematical data handling to give real-time visualization of subzero temperature regions as decision support system for the farmer.</p><p>Field implementation: A fruit farm will be equipped with optical fibre cables for Distributed Temperature Sensing, along horizontal and vertical profiles in the field. This will reveal how cooling penetrates the canopy as a function of time, and how this is influenced by changing atmospheric conditions and mitigation efforts. Detailed temperature monitoring is related to spatio-temporal physiological monitoring at the level of individual trees.</p><p>In a next step, the cables will be replaced by a 3D-network of&#160; temperature sensors. The aim is to develop an accurate (&#177;0.5&#176;C accuracy with a resolution << 0.1&#176;C), low cost sensor with ultra-low power consumption (~ 100 nW). The sensor is based on a PCB-based node that consists of a PV module to collect solar energy, a power management integrated circuit (PMIC), a supercapacitor to store energy, a temperature sensor, a microcontroller (&#181;C), a timing control unit (TCU) to enable/disable the system, and a radio frequency IC (RFIC) + antenna to transmit data to the network. To reduce energy consumption, it should operate in low-power &#8220;sleep mode&#8221; as much as possible, while still being able to capture sudden temperature changes as by ventilator activation: the sensor must decide when to &#8220;wake up&#8221; and how frequently to measure. The often &#8220;power-hungry&#8221; MCU and RF radio should operate in an event-driven mode and only &#8220;awakened&#8221; when the sensor detects a temperature change above a certain threshold.<br>We chose LoRa for its low power consumption and long-distance capability, which is a perfect match with our application.</p>