Usability and Performance Analysis of Embedded Development Environment for On-device Learning
It addresses tool selection for developers implementing on-device learning in IoT, but is incremental as it compares existing tools without introducing new methods.
This research empirically evaluates embedded development tools for on-device TinyML on resource-constrained IoT devices, finding that Arduino Framework offers ease of use but higher energy consumption, while RIOT OS provides efficient energy consumption despite higher memory usage.
This research empirically examines embedded development tools viable for on-device TinyML implementation. The research evaluates various development tools with various abstraction levels on resource-constrained IoT devices, from basic hardware manipulation to deployment of minimalistic ML training. The analysis encompasses memory usage, energy consumption, and performance metrics during model training and inference and usability of the different solutions. Arduino Framework offers ease of implementation but with increased energy consumption compared to the native option, while RIOT OS exhibits efficient energy consumption despite higher memory utilization with equivalent ease of use. The absence of certain critical functionalities like DVFS directly integrated into the OS highlights limitations for fine hardware control.