Hibernus++: A Self-Calibrating and Adaptive System for Transiently-Powered Embedded Devices

Abstract

Abstract-Energy harvesters are being used to power autonomous systems, but their output power is variable and intermittent. To sustain computation, these systems integrate batteries or supercapacitors to smooth out rapid changes in harvester output. Energy storage devices require time for charging and increase the size, mass and cost of systems. The field of transient computing moves away from this approach, by powering the system directly from the harvester output. To prevent an application from having to restart computation after a power outage, approaches such as Hibernus allow these systems to hibernate when supply failure is imminent. When the supply reaches the operating threshold, the last saved state is restored and the operation is continued from the point it was interrupted. This work proposes Hibernus++ to intelligently adapt the hibernate and restore thresholds in response to source dynamics and system load properties. Specifically, capabilities are built into the system to autonomously characterize the hardware platform and its performance during hibernation in order to set the hibernation threshold at a point which minimizes wasted energy and maximizes computation time. Similarly, the system auto-calibrates the restore threshold depending on the balance of energy supply and consumption in order to maximize computation time. Hibernus++ is validated both theoretically and experimentally on microcontroller hardware using both synthesized and real energy harvesters. Results show that Hibernus++ provides an average 16% reduction in energy consumption and an improvement of 17% in application execution time over stateof-the-art approaches.

Energy-driven computing: Rethinking the design of energy harvesting systems

“…As a result, the only volatile state left is in the registers, which are copied to NVM when an imminent power outage is detected using a voltage interrupt monitoring ++ (the system's decoupling capacitance allows this decay to be detected and the volatile state saved before the decreasing ++ causes the system to lose power). This is similar to that of our own work [2][9], discussed in Section III. One disadvantage of this approach is that NVM typically consumes greater power than SRAM, hence a quiescent overhead is always incurred in terms of energy.…”

“…As an extension to hibernus, hibernus++ [2] performs adaptive, run-time calibration and management of the platform and energy harvesting source, to avoid the need to provide the design-time calibration mentioned above. Through this approach, hibernus++ allows the system to operate effectively with an amount of energy storage that was unknown at designtime.…”

“…At first glance, energy harvesting appears to be a very attractive offering, and has gained increasing traction over the past decade; however it is not without its own challenges. ', and (bottom) the available power from an indoor photovoltaic cell over a period of two days [2].…”

Using Sleep States to Maximize the Active Time of Transient Computing Systems

“…• Hibernus++ [2] overcomes the limitations of previous transient computing systems by saving a snapshot only before an expected power failure. Using voltage comparator interrupts, the supply voltage is monitored against hibernate and restore voltage thresholds.…”

“…A state-of-the-art approach for transient computing is Hibernus++ [2], which self-calibrates its voltage thresholds to match the platform it is deployed on. However, even with this self-calibration capability, it tends err on the side of caution and hibernate more than necessary; this is a particular problem with low-current power sources.…”