The future of Internet of Things (IoT) envisions billions of sensors integrated with the physical environment. At the same time, recharging and replacing batteries on this infrastructure could result not only in high maintenance costs, but also large amounts of toxic waste due to the need to dispose of old batteries. Recently, battery-free sensor platforms have been developed that use supercapacitors as energy storage, promising maintenance-free and perpetual sensor operation. While prior work focused on supercapacitor characterization, modelling and supercapacitor-aware scheduling, the impact of mobility on capacitor charging and overall sensor application performance has been largely ignored. We show that supercapacitor size is critical for mobile system performance and that selecting an optimal value is not trivial: small capacitors charge quickly and enable the node to operate in low energy environments, but cannot support intensive tasks such as communication or reprogramming; increasing the capacitor size, on the other hand, enables the support for energy-intensive tasks, but may prevent the node from booting at all if the node navigates in a low energy area. The paper investigates this problem and proposes a hybrid storage solution that uses an adaptive learning algorithm to predict the amount of available ambient energy and dynamically switch between two capacitors depending on the environment. The evaluation based on extensive simulations and prototype measurements showed up to 40% and 80% improvement compared to a fixed-capacitor approach in terms of the amount of harvested energy and sensor coverage.
This is an open access article distributed under the Creative Commons Attribution License
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited