Battery-free computer gaming offers a vision of sustainable interaction in which games run on hardware that does not require a battery, yet this approach introduces uncertainty due to frequent power failures. Rather than viewing these failures as limitations, this work examines how integrating energy harvesting with application design can encourage users to reimagine and work with such failures, thus shaping behaviour and supporting device use. We present TURNER, a state-of-the-art modular battery-free games console powered by a hand crank and solar cells, created as a research probe to study how energy harvesting mediates the relationship between power and interaction. In a mixed-methods study (N = 60), we explored the influence of energy harvesting on gameplay. Findings show significant variations in harvesting strategies, with interviews surfacing strategies for creating applications that respond to and build on the patterns of system power failure, the ergonomics of energy harvesting, and the value of embedding energy generation into play. Our work offers insights for interactive, sustainable battery-free computers.

Public testbeds are essential for replicable experiments and meaningful comparisons on shared physical infrastructure. While many testbeds exist for battery-powered Internet of Things (IoT) systems, there is a lack of public testbeds for observing and profiling the distributed operation of energy-harvesting IoT systems, including battery-free devices. We fill this gap and present Shepherd Nova, the first public testbed designed to support experiments under repeatable energy-harvesting conditions. Shepherd Nova uses field-recorded harvesting data to supply power to devices, consistently replicating real-world spatio-temporal energy availability across multiple experiments. Its virtual power source supports diverse ambient energy sources, harvesting circuitry, and energy storage devices. Moreover, Shepherd Nova provides services like general-purpose input/output (GPIO) tracing, power profiling, and serial output logging, all of which can run synchronously and with high resolution. Sub-microsecond synchronization enables precise correlation between these observations and emulated energy-harvesting conditions, offering unprecedented insights into distributed energy-harvesting IoT systems. In this paper, we describe Shepherd Nova's design, characterize its performance, and demonstrate its capabilities through controlled experiments and an example test case. To access the testbed, documentation as well as open-source harvesting data, hardware designs, and code, visit https://testbed.nes-lab.org/.

Battery-free sensing devices have gained growing popularity as they can operate relying solely on harvested energy and environmentally friendly capacitors. However, despite the increasing number of battery-free solutions, their design remains a difficult task. In fact, the limited energy storage capacity and the resulting coupling between energy supply and demand introduce new design trade-offs that cannot be explored using conventional tools that consider a constant power supply. To enable fast design space exploration and facilitate the development of battery-free systems, we introduce Simba, an open-source simulation framework that allows to investigate in detail the complex interplay between various device components. We demonstrate the benefits of Simba in two case studies, evaluated experimentally, targeting real-world, state-of-the-art battery-free devices. First, we illustrate how Simba can explore the dependencies between different component configurations and assess their impact on the overall system performance. Among others, we show that changing the storage capacity or slightly modifying the load behavior can improve data throughput by a factor of up to 5.1x and 9.7x, respectively. Second, we present how Simba allows to automatically select key parameters that optimize the operations of a battery-free system (e.g., its checkpointing mechanism), and showcase how Simba enables performance evaluations based on real-world energy harvesting traces.CCS CONCEPTS• Computer systems organization → Embedded systems.

We present an architecture for intermittently-powered wireless communication systems that does not require any changes to the official protocol specification. Our core idea is to save the intermediate state of the wireless protocol to non-volatile memory within each connection interval. The protocol state is then deterministically restored at a predefined (harvested energy-dependent) time, which follows the connection interval. As a case study for our architecture, we introduce FreeBie: a battery-free intermittently-powered Bluetooth Low Energy (BLE) mote. To the best of our knowledge FreeBie is the first battery-free active wireless system that sustains bi-directional communication on intermittent harvested energy. The strength of our architecture is articulated by FreeBie consuming at least 9.5 times less power during device inactivity periods than a state-of-the-art BLE device.

Battery-free and intermittently powered devices offer long lifetimes and enable deployment in new applications and environments. Unfortunately, developing sophisticated inference-capable applications is still challenging due to the lack of platform support for more advanced (32-bit) microprocessors and specialized accelerators---which can execute data-intensive machine learning tasks, but add complexity across the stack when dealing with intermittent power. We present Protean to bridge the platform gap for inference-capable battery-free sensors. Designed for runtime scalability, meeting the dynamic range of energy harvesters with matching heterogeneous processing elements like neural network accelerators. We develop a modular "plug-and-play" hardware platform, SuperSensor, with a reconfigurable energy storage circuit that powers a 32-bit ARM-based microcontroller with a convolutional neural network accelerator. An adaptive task-based runtime system, Chameleon, provides intermittency-proof execution of machine learning tasks across heterogeneous processing elements. The runtime automatically scales and dispatches these tasks based on incoming energy, current state, and programmer annotations. A code generator, Metamorph, automates conversion of ML models to intermittent safe execution across heterogeneous compute elements. We evaluate Protean with audio and image workloads and demonstrate up to 666x improvement in inference energy efficiency by enabling usage of modern computational elements within intermittent computing. Further, Protean provides up to 166% higher throughput compared to non-adaptive baselines.

Debugging and testing battery-free intermittently-powered systems is notoriously difficult. This is not only due to the additional complexity of maintaining state through power failures but also due to the lack of proper tools to test and debug these systems. As a solution, we present DIPS: a fully-featured hardware debugger for battery-free intermittently-powered systems capable of automatically verifying memory and peripheral state between power failures. Our solution seamlessly integrates an emulator allowing for emulation of any power scenario to the device under test. This allows our debugger to pause emulation and program execution when debugging or when state restoration issues are detected. Our new system is built around GNU Debugger (GDB): a widely-used debugging tool. Therefore, DIPS allows for a debugging process identical to state-of-the-art debuggers for continuously-powered devices. User studies found that our debugger is easy and intuitive to use. It allows embedded system developers to find bugs quicker in code written for battery-free devices. With our debugger we found unseen errors in state-of-the-art software frameworks for intermittently-powered systems.

Energy-harvesting devices have enabled Internet of Things applications that were impossible before. One core challenge of batteryless sensors that operate intermittently is reliable timekeeping. State-of-the-art low-power real-time clocks suffer from long start-up times (order of seconds) and have low timekeeping granularity (tens of milliseconds at best), often not matching timing requirements of devices that experience numerous power outages per second. Our key insight is that time can be inferred by measuring alternative physical phenomena, like the discharge of a simple RC circuit, and that timekeeping energy cost and accuracy can be modulated depending on the run-time requirements. We achieve these goals with a multi-tier timekeeping architecture, named Cascaded Hierarchical Remanence Timekeeper (CHRT), featuring an array of different RC circuits to be used for dynamic timekeeping requirements. The CHRT and its accompanying software interface are embedded into a fresh batteryless wireless sensing platform, called Botoks, capable of tracking time across power failures. Low start-up time (max 5 ms), high resolution (up to 1 ms) and run-time reconfigurability are the key features of our timekeeping platform. We developed two time-sensitive batteryless applications to demonstrate the approach: a bicycle analytics tool, where the CHRT is used to track time between revolutions of a bicycle wheel, and wireless communication, where the CHRT enables radio synchronization between two intermittently-powered sensors.