EUVIMEDEUVIMED entdecken
Crossref

Supervised QoS-Aware Energy Saving in O-RAN

Iran M. Braga, Adriely L. Brandelli, Francisco H. Costa Neto, João V. B. Labres, Rodrigo K. Y. Aoki

IEEE Communications Letters · 2026 · 30 · 2322-2326

DOI: 10.1109/lcomm.2026.3705559 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Communications Letters, 2026. Verfasst von Iran M. Braga und 4 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben. Gefördert von: EMBRAPII/CPQD EXCellence Centre in Open Networks.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This letter proposes a supervised solution deployed in a rApp that optimizes cell switch-on/off decisions to maximize energy efficiency in Open Radio Access Network (O-RAN) networks while guaranteeing Quality of Service (QoS) regarding outage probability. To address the lack of labeled data, we develop a local optimal solution leveraging Gaussian approximation and the Dinkelbach method. Validated in an environment aligned with O-RAN Alliance specifications, the proposed supervised model performs closely to the local optimal solution, achieving a gain of 98% during low-demand periods while strictly maintaining QoS requirements compared to the benchmark.

Förderung

  • EMBRAPII/CPQD EXCellence Centre in Open Networks

Zitieren

APA

Braga, I. M., Brandelli, A. L., Neto, F. H. C., Labres, J. V. B., & Aoki, R. K. Y. (2026). Supervised QoS-Aware Energy Saving in O-RAN. IEEE Communications Letters, 30, 2322-2326. https://doi.org/10.1109/lcomm.2026.3705559

Vancouver

Braga IM, Brandelli AL, Neto FHC, Labres JVB, Aoki RKY. Supervised QoS-Aware Energy Saving in O-RAN. IEEE Communications Letters. 2026;30:2322-2326. doi:10.1109/lcomm.2026.3705559

BibTeX
@article{Braga2026,
  author = {Braga, Iran M. and Brandelli, Adriely L. and Neto, Francisco H. Costa and Labres, João V. B. and Aoki, Rodrigo K. Y.},
  title = {Supervised QoS-Aware Energy Saving in O-RAN},
  journal = {IEEE Communications Letters},
  year = {2026},
  volume = {30},
  pages = {2322--2326},
  doi = {10.1109/lcomm.2026.3705559}
}
Kostenlos starten →