In the modern age, the proliferation of electronic devices and their subsequent waste presents an increasing environmental challenge. With the continuing growth of technology and the urgency of sustainability becoming ever more important, there is an ever-increasing need to reimagine how we power and utilize our devices. This has been exacerbated in the last years with the proliferation of electric cars, which need a large amount of batteries to match the range of existing internal combustion cars. The motivation behind this thesis stemmed from the desire to address these challenges head-on and create a new interactive research platform to help re- searchers improve user interactivity in a battery-free, energy-constrained system.
In this thesis, I cover the design, implementation, and evaluation of an innovative low-power gaming research platform, in the shape of a portable gaming console. The console’s uniqueness is derived from its key characteristics: an interactive nature, a lack of traditional batteries, and a robust system for handling intermittent power. The term “interactive” has been redefined in the context of a battery-free console, creating a dynamic user-console relationship that allows real-time responses to diverse user inputs and environmental factors. For example, one of the energy harvesting methods of the console, a crank, can also behave as an input to the system, allowing the mapping of in-game actions to crank movements.
The architecture of the gaming console, both in hardware and software aspects, is designed from the ground up to support key characteristics such as ease of use, future expandability, and app portability. Due to the intermittent nature of the console, the software implementation has been designed around a seamless checkpointing system to handle power interruptions gracefully, where, from the point of view of the user, the game is resumed after a power failure without any apparent loss of progress.
Finally, in order to facilitate user experiments and future work, the console also features an efficient logging system to capture user interaction and system performance data during experiments, alongside an easy-to-use interface that facilitates porting of existing software to this platform.

Existing tools for debugging battery-free applications are limited to specific architectures or require code changes of the Device Under Test (DUT) to function. These tools also cannot measure the efficiency of the application designed for battery-free systems. Currently, there is a lack of independent broad comparisons of intermittent systems. Our work, therefore, evaluates state-of-the-art frameworks and their artifacts and finds shortcomings in reproducibility and their performance. To overcome these shortcomings, we introduce DIPS+, a multi-platform debugger and measurement platform for intermittent systems with ARM and MSP430 support. DIPS+ introduces new methods to analyse applications for battery-free systems. One method finds the minimum energy budget required for forward progress, crucial for determining the minimal capacitor size for intermittent systems. Furthermore, DIPS+ offers functions to perform automatic profiling tests, like code start-up time, which gives valuable insights into the system's efficiency. DIPS+ achieves significant improvements in debugging performance, with 11 times faster connection time and reduced code execution by 157 times on the DUT before full reconnecting after intermittency occurs. The evaluation of selected frameworks reveals substantial overheads caused by the additional overhead of saving and restoring of system's state. This causes certain benchmarks to take up to 110 times longer to complete than their uninstrumented counterparts. These findings raise concerns about the viability of task-based approaches as an effective solution for managing intermittency in battery-free Internet of Things devices.

The growing field of research into batteryless or intermittent systems has enabled Internet of Things applications that were previously impossible. For example, the FreeBie system recently introduced Bluetooth Low-Energy (BLE)
to intermittent devices, making medium to long range bi-directional communication a reality for the first time. However, this achievement also highlighted
that some inefficiencies considered acceptable for conventional systems are unacceptable when working in the intermittent domain. Our key insights are that
intermittent peripherals should dictate the connection parameters and not the
central, that connection setup overhead should be reduced as much as possible,
and that connection parameter updates should be applied faster. To achieve
these goals, we 1) introduce a method of sharing connection parameters before
a connection is made, 2) introduce methods of caching connection setup packets
together with a reconnect procedure called Fast Reconnect that reduces connection setup to a single packet, and 3) apply 1 and 2 in a dynamic algorithm
called FRAPPUCcInO that controls connection rate based on energy harvesting capabilities. These three solutions allow intermittent BLE to be used in
environments with less ambiently available energy than before by improving
efficiency and responsiveness.