Abstract—As sensor equipped wearable systems enter the mainstream, system longevity and power-efficiency issues hamper large scale and long-term deployment, despite substantial foreseeable benefits. As power and energy efficient design, sampling, processing and communication techniques emerge to counter these issues, researchers are beginning to look on wearable energy harvesting systems as an effective counterpart solution. In this paper, we propose a novel harvesting technology to inconspicuously transduce mechanical energy from human foot-strikes and power low-power wearable systems in a self-sustaining manner. Dielectric Elastomers (DEs) are high-energy density electrostatic transducers that can transduce significant levels of energy from a user while appearing near-transparent to her, if configured and controlled properly. Towards this end, we propose DE-based harvester configuration that capitalizes on properties of human gait to enhance transduction efficiency, and further leverage these properties in an adaptive control algorithm to optimize the net energy produced by the system. We evaluate system performance from detailed analytical and empirical models of DE transduction behavior, and apply our control algorithm to the modeled DEs under experimentally collected foot pressure datasets from multiple subjects. Our evaluations show that the proposed system can achieve up to 120mJ per foot-strike, enough to power a variety of low-power wearable devices and systems. I.

Abstract—Recent advances in the scope of wearable devices and networks make body area sensor networks (BASNs) an extremely attractive tool to the fields of mobile and tele-health, owing to the range of medical applications they can serve and the diagnostic richness of patient data they can offer. However, for BASNs to achieve true ubiquity, they must be scalable in their support of automated patient data collection, making usability and reliability key considerations. Its designers must wrestle with the tradeoff be-tween usability, hindered by device intrusiveness into the behav-iors it measures, and lifetime, enhanced by large power supplies and expensive, sturdy components. Furthermore, the validity and reliability of the collected data are paramount. In this paper, we consider these issues in the context of localizedmulti-sensory wear-able networks and present a method to generate low-power sam-pling schedules that are resilient to sensor faults while achieving high diagnostic fidelity. We jointly formulate this as a power-con-strained sampling problem wherein the number of sensors sam-pled per epoch are limited, and, a fault tolerant scheduling problem wherein the sampling scheme offers enough redundancy to endure up to a predefined number of sensor faults while maintaining diag-nostic accuracy. This formulation is based on, 1) the localized scope of BASNs that engenders strong spatio-temporal interactions in the samples, and, 2) the periodic nature of human behaviorsmeasured. We present our algorithm in the context of gait diagnostics derived from a foot plantar pressure measurement platform and illustrate its performance based on real datasets collected by it. Index Terms—Body area sensor networks, energy-efficient sam-pling, fault-tolerant sampling, power-constrained sampling. I.