Sub-100 $\mu$W Multispectral Riemannian Classification for EEG-Based Brain-Machine Interfaces
Lists
Only you see this
Abstract
Motor imagery (MI) brain–machine interfaces (BMIs) enable us to control machines by merely thinking of performing a motor action. Practical use cases require a wearable solution where the classification of the brain signals is done locally near the sensor using machine learning models embedded on energy-efficient microcontroller units (MCUs), for assured privacy, user comfort, and long-term usage. In this work, we provide practical insights on the accuracy-cost trade-off for embedded BMI solutions. Our multispectral Riemannian classifier reaches 75.1% accuracy on a 4-class MI task. The accuracy is further improved by tuning different types of classifiers to each subject, achieving 76.4%. We further scale down the model by quantizing it to mixed-precision representations with a minimal accuracy loss of 1% and 1.4%, respectively, which is still up to 4.1% more accurate than the state-of-the-art embedded convolutional neural network. We implement the model on a low-power MCU within an energy budget of merely 198 <inline-formula><tex-math notation="LaTeX">$\mu$</tex-math></inline-formula>J and taking only 16.9 ms per classification. Classifying samples continuously, overlapping the 3.5 s samples by 50% to avoid missing user inputs allows for operation at just 85 <inline-formula><tex-math notation="LaTeX">$\mu$</tex-math></inline-formula>W. Compared to related works in embedded MI-BMIs, our solution sets the new state-of-the-art in terms of accuracy-energy trade-off for near-sensor classification.
BibTeX
@article{wang2021sub,
author = {Xiaying Wang and Lukas Cavigelli and Tibor Schneider and Luca Benini},
title = {{Sub-100 $\textbackslash\{\}mu$W Multispectral Riemannian Classification for EEG-Based Brain-Machine Interfaces}},
journal = {IEEE Trans. Biomed. Circuits Syst.},
year = {2021},
url = {https://openreview.net/forum?id=AY5dijSPOYy},
}