FedSpace: An Efficient Federated Learning Framework at Satellites and Ground Stations

So, Jinhyun; Hsieh, Kevin; Arzani, Behnaz; Noghabi, Shadi A.; Avestimehr, Salman; Chandra, Ranveer

doi:10.48550/arxiv.2202.01267

Cited by 6 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Without this capability, several extensions of the approach in [5] are possible that might lead to faster convergence speed. Indeed, the authors of [7] consider a heuristic GS update procedure and gradient buffering to speed up convergence.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Scheduling for Ground-Assisted Federated Learning in LEO Satellite Constellations

Razmi¹,

Matthiesen²,

Dekorsy³

et al. 2022

Preprint

View full text Add to dashboard Cite

Distributed training of machine learning models directly on satellites in low Earth orbit (LEO) is considered. Based on a federated learning (FL) algorithm specifically targeted at the unique challenges of the satellite scenario, we design a scheduler that exploits the predictability of visiting times between ground stations (GS) and satellites to reduce model staleness. Numerical experiments show that this can improve the convergence speed by a factor three.

show abstract

Section: Introductionmentioning

confidence: 99%

“…In this paper, we take a different approach and design a scheduler based on the predictability of GS-satellite visiting lengths to reduce the model staleness. This leads to significantly faster convergence speed than the baseline approach in [5] and can be combined with different aggregation rules, such as the one in [7].…”

Section: Introductionmentioning

confidence: 99%

Scheduling for Ground-Assisted Federated Learning in LEO Satellite Constellations

Razmi¹,

Matthiesen²,

Dekorsy³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, when satellites in one orbital plane are traveling in the opposite direction to satellites in another orbital plane, the Doppler shift will be maximized which is known as crossseam [4]. Also, due to the high mobility of LEO's satellites, some satellites can be in contact with the GS several times a day, while other may be blocked due to the Earth's rotation [5], [6]. Therefore, all of these challenges will increase the uncertainty of the communication between satellites to each other and the GS, in turn, will impact the FL convergence.…”

Section: Introductionmentioning

confidence: 99%

“…Several recent studies have examined the use of FL in LEO constellations [5]- [8]. Chen et al [7], for example, have applied a traditional FL algorithm to satellite constellations and compared their results to the centralized training at the GS.…”

Section: Introductionmentioning

confidence: 99%

“…A new asynchronous FL algorithm called FedSat is also proposed in [8], however, it is only applicable if the satellites visit the GS once a period. Additionally, So et al [5] proposed another new FL algorithm called FedSpace that trades off local model staleness and idleness connectivity in asynchronous FL and synchronous FL, respectively. However, FedSpace can only be used if the GS has a dataset that is similar to the satellite-generated dataset or if the satellites have to download low-resolution images to enable FedSpace to determine the satellite's connectivity with the GS.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Federated learning for LEO constellations via inter-HAP links

Elmahallawy¹,

Luo²

2022

Preprint

View full text Add to dashboard Cite

Low Earth Obit (LEO) satellite constellations have seen a sharp increase of deployment in recent years, due to their distinctive capabilities of providing broadband Internet access and enabling global data acquisition as well as large-scale AI applications. To apply machine learning (ML) in such applications, the traditional way of downloading satellite data such as imagery to a ground station (GS) and then training a model in a centralized manner, is not desirable because of the limited bandwidth, intermittent connectivity between satellites and the GS, and privacy concerns on transmitting raw data. Federated Learning (FL) as an emerging communication and computing paradigm provides a potentially supreme solution to this problem. However, we show that existing FL solutions do not fit well in such LEO constellation scenarios because of significant challenges such as excessive convergence delay and unreliable wireless channels. To this end, we propose to introduce high-altitude platforms (HAPs) as distributed parameter servers (PSs) and propose a synchronous FL algorithm, FedHAP, to accomplish model training in an efficient manner via inter-satellite collaboration. To accelerate convergence, we also propose a layered communication scheme between satellites and HAPs that FedHAP leverages. Our simulations demonstrate that FedHAP attains model convergence in much fewer communication rounds than benchmarks, cutting the training time substantially from several days down to a few hours with the same level of resulting accuracy.

show abstract

AsyncFLEO: Asynchronous Federated Learning for LEO Satellite Constellations with High-Altitude Platforms

Elmahallawy

Luo

2022

2022 IEEE International Conference on Big Data (Big Data)

View full text Add to dashboard Cite

The advances in satellite technology developments have recently seen a large number of small satellites being launched into space on Low Earth orbit (LEO) to collect massive data such as Earth observational imagery. The traditional way which downloads such data to a ground station (GS) to train a machine learning (ML) model is not desirable due to the bandwidth limitation and intermittent connectivity between LEO satellites and the GS. Satellite edge computing (SEC), on the other hand, allows each satellite to train an ML model onboard and uploads only the model to the GS which appears to be a promising concept. This paper proposes FedLEO, a novel federated learning (FL) framework that realizes the concept of SEC and overcomes the limitation (slow convergence) of existing FL-based solutions. FedLEO (1) augments the conventional FL's star topology with "horizontal" intra-plane communication pathways in which model propagation among satellites takes place; (2) optimally schedules communication between "sink" satellites and the GS by exploiting the predictability of satellite orbiting patterns. We evaluate FedLEO extensively and benchmark it with the state of the art. Our results show that FedLEO drastically expedites FL convergence, without sacrificingin fact it considerably increases-the model accuracy.

show abstract

FedSpace: An Efficient Federated Learning Framework at Satellites and Ground Stations

Cited by 6 publications

References 0 publications

Scheduling for Ground-Assisted Federated Learning in LEO Satellite Constellations

Scheduling for Ground-Assisted Federated Learning in LEO Satellite Constellations

Federated learning for LEO constellations via inter-HAP links

AsyncFLEO: Asynchronous Federated Learning for LEO Satellite Constellations with High-Altitude Platforms

Contact Info

Product

Resources

About