EUVIMEDEUVIMED entdecken
Crossref

COSMO: COmpressed Sensing for Models and Logging Optimization in MCU Performance Screening

Nicolò Bellarmino, Riccardo Cantoro, Sophie M. Fosson, Martin Huch, Tobias Kilian, Ulf Schlichtmann, Giovanni Squillero

IEEE Transactions on Computers · 2025-02 · 74 (2) · 652-664

DOI: 10.1109/tc.2024.3500378 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Computers, 2025-02. Verfasst von Nicolò Bellarmino und 6 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 44 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In safety-critical applications, microcontrollers must meet stringent quality and performance standards, including the maximum operating frequency $F_{\max}$ F max . Machine learning models have proven effective in estimating $F_{\max}$ F max by utilizing data from on-chip ring oscillators. Previous research has shown that increasing the number of ring oscillators on board can enable the deployment of simple linear regression models to predict $F_{\max}$ F max . However, the scarcity of labeled data that characterize this context poses a challenge in managing high-dimensional feature spaces; moreover, a very high number of ring oscillators is not desirable due to technological reasons. By modeling $F_{\max}$ F max as a linear combination of the ring oscillators’ values, this paper employs Compressed Sensing theory to build the model and perform feature selection, enhancing model efficiency and interpretability. We explore regularized linear methods with convex/non-convex penalties in microcontroller performance screening, focusing on selecting informative ring oscillators. This permits reducing models’ footprint while retaining high prediction accuracy. Our experiments on two real-world microcontroller products compare Compressed Sensing with two alternative feature selection approaches: filter and wrapped methods. In our experiments, regularized linear models effectively identify relevant ring oscillators, achieving compression rates of up to 32:1, with no substantial loss in prediction metrics.

Zitieren

APA

Bellarmino, N., Cantoro, R., Fosson, S. M., Huch, M., Kilian, T., Schlichtmann, U., & Squillero, G. (2025). COSMO: COmpressed Sensing for Models and Logging Optimization in MCU Performance Screening. IEEE Transactions on Computers, 74(2), 652-664. https://doi.org/10.1109/tc.2024.3500378

Vancouver

Bellarmino N, Cantoro R, Fosson SM, Huch M, Kilian T, Schlichtmann U, et al. COSMO: COmpressed Sensing for Models and Logging Optimization in MCU Performance Screening. IEEE Transactions on Computers. 2025;74(2):652-664. doi:10.1109/tc.2024.3500378

BibTeX
@article{Bellarmino2025,
  author = {Bellarmino, Nicolò and Cantoro, Riccardo and Fosson, Sophie M. and Huch, Martin and Kilian, Tobias and Schlichtmann, Ulf and Squillero, Giovanni},
  title = {COSMO: COmpressed Sensing for Models and Logging Optimization in MCU Performance Screening},
  journal = {IEEE Transactions on Computers},
  year = {2025},
  volume = {74},
  number = {2},
  pages = {652--664},
  doi = {10.1109/tc.2024.3500378}
}
Kostenlos starten →