Towards Power Autonomous Wireless Sensors

Abstract

To enable widespread deployment of sensor systems in our world, they must be wireless and power autonomous. There is no universal solution to achieve such a device. Instead a system level approach must be taken that considers the sensor, its readout, wireless communications and powering methods available in the context of the application requirements and sensor system lifetime. Achieving power autonomy requires a two front approach; addressing both the need and the supply. The optimisation of power consumption in this work focuses on the radio communications, highlighted as one of the most power critical subsystems. A power-optimised system enables the exploitation of application appropriate energy sources; whether they be battery power in the case of short lifetime devices, energy harvesting where reliable sources exist, wireless power transfer where it can be applied, or passive radio backscattering techniques. This thesis presents four independent systems, each a power autonomous solution in the context of its application requirements. The first system integrates a voltammetric electronic tongue sensor for gastrointestinal disease diagnosis, with power provided by a high-density lithium manganese dioxide primary cell, ample for the 72 hours operation time of the capsule. The capsule contains an electrically small loop antenna and communicates wirelessly at 433 MHz to a remote receiver. The second system applies a MEMS acoustic emission sensor for online condition monitoring of valve leakage in the petrochemical industry, utilising thermal energy harvesting in the sensor node, and industrial current loop harvesting in the gateway to achieve power autonomy for continuous operation. The device communicates at 2.45 GHz and includes an ultra-low power wake-up radio based on passive down converting receiver to enable the sensor to be in an ultra-low power mode and asynchronously interrogated by the gateway. The third system exploits resonant inductive coupling for the actuation of a wirelessly powered soft robotic caterpillar which itself is potentially a fully passive wireless sensor platform. The system utilises a frequency sweeping power transmission mechanism, around 8.5 MHz, to wirelessly transfer energy to multiple shape memory alloy actuators, enabling the generation of a wave of force through the caterpillar body providing locomotion. The fourth system employs a fully passive wireless sensing platform based on radio backscattering at UHF frequencies (around 868 MHz), and a harmonic resonant sensor based on the third order intermodulation principle. In its entirety, this work represents progress on multiple fronts on the key challenge of powering the next generation of IoT devices.