Circuit Techniques for Wirelessly Powered Sensors and Actuators

In conventional sensors and actuators, batteries are used to provide a stable source of energy. Unfortunately, batteries deplete over time, reduce the lifetime of the sensors, make them bulky, and increase the overall cost of the system. Eliminating batteries makes medical implants smaller, reduces the risk of failure, and in some applications lowers the bar for regulatory approvals. In industrial applications, eliminating batteries makes it possible to make the sensors small enough such that they can be deployed in small geometries such as fractures in an oil/gas reservoir. These sensors can be used to perform localization, identify the geometry of a fracture, and perform reservoir characterization in previously impossible locations. Eliminating batteries reduces the cost of the sensors, making it economical to deploy them in large quantities. The information provided by these sensors is highly valuable for energy producers helping them to improve safety, increase the efficiency of energy production, and ultimately help to secure the future of energy. In addition to medical and industrial applications, the Trillion Internet-of-Things (IoT) vision is becoming a reality. Trillion IoT sensors need to operate without a battery eliminating the need for their replacement. Eliminating the batteries makes the design of a sensor very challenging. Batteryless sensors need to harvest energy from other sources, such as electromagnetic waves. To harvest electromagnetic waves, the sensors need to rectify the received signals and produce a stable supply for a sufficient duration to complete the task of sensing, digitization, processing, and wireless transmission of the sensed data. The electromagnetic power radiated to a batteryless sensor generates a large interference making it difficult to sense a weak electrical signal with a low amplitude. Batteryless sensors have a very limited power budget and, in many cases, cannot afford to produce a stable clock locally. In these cases, a clean clock signal needs to be extracted from the same electromagnetic waves used to power the sensor. In addition, batteryless sensors cannot stay on for a long duration and need to operate on a duty-cycled mode. The duty-cycled operation makes it challenging to perform a complex computational task and requires a non-volatile memory to store the state of the computation before the sensor uses its power. Alternatively, the sensor needs to send the raw data wirelessly to an external unit before losing power so that the processing can be done on the external unit. In this presentation, we will review some of these important challenges and introduce several circuit and system level techniques to address them.