An energy autonomous electronic transmitting system for green plant sensing applications

This dissertation deals with the design of a completely autonomous electronic transmitting system meant to be applied in plant sensing applications. It is grafted on a previously performed study carried on tobacco plants. It concerned the study of the variation of plant stem's electrical impedance with respect to the drying process and analysis carried to understand which is the best frequency range for signal propagation. It has been found that electrical impedance grows over time if no watering events occur, and it decreases right after the plant gets water. This mechanism offers a considerable possibility: understanding plant health conditions inspecting its impedance by directly exploiting electrical signals propagation inside it. This kind of analysis has never been done: watering status information has always been acquired through the reading of external sensors (for example, soil humidity, sun irradiation, and temperature) or human knowledge. This work aims to provide a system able to overcome this limit and offer the possibility to get direct information from plants. The target device must show essential characteristics: low power consumption, reliability, long durability, biocompatibility, and (hopefully) compactness. In the thesis, every component used to create the transmitting system will be presented together with the reasons leading to its choice, advantages, and drawbacks. Since the transmitter has to be implemented directly on plants, it must rely on nature's energy sources. The chosen one has to be renewable, reliable, and easy to convert. To this purpose, firstly, an overview of all the exploitable sources, related harvesting device, and the reasons that led to discard or choose them, is presented. Renewable resources are not entirely reliable (for example, there is no sunlight to convert at night, and the wind does not always blow). Thus a storage device must be implemented to guarantee a higher level of durability. Such a device must have specific features and satisfy certain performance standards. Therefore a brief overview of storage devices will be done to choose the most suited, and reasons leading to the final choice will be described in detail. Then stratagems used to prevent useless power consumption and improve power management will be reported together with devices allowing them, advantages, and disadvantages that their implementation implies. At the end of every transmitter component description, the whole device and its working principle will be shown, highlighting the reasons that led to discard or choose it. After that, the complete transmitter is presented assembling every component forming the whole device, and its working principle will be entirely detailed. Finally, tests performed on a real tobacco plant will be presented. They highlighted that, even in the total absence of energy source, power saving gimmicks are quite useful, leading to the power consumption of about $18 \ \mu W$ per each measurement. It implied that the estimated transmitter autonomy is quite satisfactory: approximately one day and a half. Moreover, it showed sufficient autonomy even when sunlight was weak (cloudy days) or highly scattered (foggy days). The system lasted for more than two days, corroborating the goodness of the choices made. At the end of this work, possible improvements that can be implemented to improve performances will be presented.