Asynchronous event-based self-powering, computation, and data logging

Abstract

Asynchronous self-powering refers to an energy scavenging approach where energy for sensing, computation, and nonvolatile storage is harvested directly from the signal being sensed. The approach eliminates the need for energy regulation modules, energy storage, analog-to-digital converters, microcontrollers, and random-access memory, all of which are commonly used in traditional energy scavenging sensors. In this chapter, we describe the fundamental principles of asynchronous self-powering by considering a case study of a sensor designed for structural health monitoring (SHM) applications. In this regard, we describe how the device physics governing the operation of nonvolatile analog memory could be combined with the physics of piezoelectric and electrostatic transducers such that the resulting circuits can operate at fundamental limits of self-powering. For the sake of completeness, we describe an architecture of a system-on-chip that uses ambient strain variations to asynchronously self-power and compute signal-level and signal-velocity statistics.