EUVIMEDEUVIMED entdecken
Crossref

FedOrbit: Energy Efficient Federated Learning for Orbital Edge Computing Using Block Minifloat Arithmetic

Mohammad Reza Jabbarpour, Bahman Javadi, Philip H.W. Leong, Rodrigo N. Calheiros, David Boland

IEEE Transactions on Services Computing · 2024-11 · 17 (6) · 3657-3671

DOI: 10.1109/tsc.2024.3478768 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Services Computing, 2024-11. Verfasst von Mohammad Reza Jabbarpour und 4 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 53 Literaturangaben. Gefördert von: NSW Defence Innovation Network, NSW State Government through DIN Pilot Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Low Earth Orbit (LEO) satellite constellations have diverse applications, including earth observation, communication services, navigation, and positioning. These constellations have evolved into a valuable data source; however, their use in a ground station (GS) for analysis via machine learning algorithms presents challenges due to constraints on power consumption, communication bandwidth, and onboard computing capabilities. While the combination of Federated Learning (FL) and Orbital Edge Computing has been employed to address these challenges, its heavy reliance on the GS for model aggregation and edge resource limitations remains a research challenge. This article presents FedOrbit, a novel energy-efficient and decentralised FL method to optimise communication with the GS and reduce power consumption. FedOrbit utilises reinforcement learning for cluster formation, satellite visiting patterns for master satellite selection, and block minifloat arithmetic for power reduction. Extensive performance evaluation under Walker Delta-based LEO constellation configurations and different datasets reveals that FedOrbit can maintain high accuracy while significantly reduce communication demand, power consumption and training time in comparison to state-of-the-art FL approaches. The proposed technique can also reduce the training time by 5× compared with the centralised FL approaches. In addition, the utilisation of block minifloat representation as low-precision arithmetic enhanced the energy consumption by 3.5× compared with the single-precision (FP32) format.

Förderung

  • NSW Defence Innovation Network
  • NSW State Government through DIN Pilot Project

Zitieren

APA

Jabbarpour, M. R., Javadi, B., Leong, P. H., Calheiros, R. N., & Boland, D. (2024). FedOrbit: Energy Efficient Federated Learning for Orbital Edge Computing Using Block Minifloat Arithmetic. IEEE Transactions on Services Computing, 17(6), 3657-3671. https://doi.org/10.1109/tsc.2024.3478768

Vancouver

Jabbarpour MR, Javadi B, Leong PH, Calheiros RN, Boland D. FedOrbit: Energy Efficient Federated Learning for Orbital Edge Computing Using Block Minifloat Arithmetic. IEEE Transactions on Services Computing. 2024;17(6):3657-3671. doi:10.1109/tsc.2024.3478768

BibTeX
@article{Jabbarpour2024,
  author = {Jabbarpour, Mohammad Reza and Javadi, Bahman and Leong, Philip H.W. and Calheiros, Rodrigo N. and Boland, David},
  title = {FedOrbit: Energy Efficient Federated Learning for Orbital Edge Computing Using Block Minifloat Arithmetic},
  journal = {IEEE Transactions on Services Computing},
  year = {2024},
  volume = {17},
  number = {6},
  pages = {3657--3671},
  doi = {10.1109/tsc.2024.3478768}
}
Kostenlos starten →