Design and Specification of Batteryless Sensing Systems

Over the past few decades, batteries played a central role in the design of wireless sensing systems. Large storage devices provide a stable energy supply, ensuring long system lifetimes even when energy demands are highly variable. Current trends point towards the deployment of billions of interconnected sensing devices gathering information from their surrounding, also known as the Internet-of-Things (IoT). While energy flow is absolutely necessary for IoT devices to function, large energy storage capacity is not. Minimized energy provisioning will make the IoT more economically viable and environmentally friendly. It also restricts the use of high-power peripherals and introduces intermittence, raising new challenges in application development. In this dissertation, we address how rich data sensing systems can be designed for robust and efficient operation with a minimized transducer and storage element. Batteryless systems are particularly suited for energy-driven sensing applications, where energy and information are simultaneously available in the environment. To this end, we design a general-purpose power subsystem compatible with transducers whose output cannot directly sustain system operation. Then, we introduce a data aggregation scheme which can drastically reduce average transactional costs to power hungry peripherals. Finally, we present a specification model for intermittence-tolerant applications. With it, developers can write arbitrarily long (single-core) sensing applications and automatically calculate the optimal energy burst partitioning with optimized data state retention and restoration. Specifically, the following contributions are presented in this dissertation. • We design an Energy Management Unit (EMU), which acts as a broker between a low-power transducer