Federated Learning over LEO Satellite

Wang, Yiji; Zou, Cheng; Wen, Dingzhu; Shi, Yuanming

doi:10.1109/gcwkshps56602.2022.10008719

Cited by 11 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the aid of FL, LEOs can process the data locally and only exchange the encrypted model among LEOs through a candidate edge server, which reduces the transmission data size and the time induced from long distance between LEOs and the GS. Recent works have been discussing the benefits and challenges to establish FL on satellite systems [14], [15]. In [16], the authors improve the asynchronous FL with the scheduling algorithm in LEO constellation.…”

Section: Introductionmentioning

confidence: 99%

Edge Selection and Clustering for Federated Learning in Optical Inter-LEO Satellite Constellation

Chen¹,

Shen²,

Feng³

et al. 2023

Preprint

View full text Add to dashboard Cite

Low-Earth orbit (LEO) satellites have been prosperously deployed for various Earth observation missions due to its capability of collecting a large amount of image or sensor data. However, traditionally, the data training process is performed in the terrestrial cloud server, which leads to a high transmission overhead. With the recent development of LEO, it is more imperative to provide ultra-dense LEO constellation with enhanced onboard computation capability. Benefited from it, we have proposed a collaborative federated learning for low Earth orbit (FELLO). We allocate the entire process on LEOs with low payload intersatellite transmissions, whilst the low-delay terrestrial gateway server (GS) only takes care for initial signal controlling. The GS initially selects an LEO server, whereas its LEO clients are all determined by clustering mechanism and communication capability through the optical inter-satellite links (ISLs). The re-clustering of changing LEO server will be executed once with low communication quality of FELLO. In the simulations, we have numerically analyzed the proposed FELLO under practical Walker-based LEO constellation configurations along with MNIST training dataset for classification mission. The proposed FELLO outperforms the conventional centralized and distributed architectures with higher classification accuracy as well as comparably lower latency of joint communication and computing.

show abstract

Section: Introductionmentioning

confidence: 99%