arxiv: 2604.16518 · v1 · submitted 2026-04-15 · 💻 cs.RO · astro-ph.IM· cs.AI

Recognition: unknown

On-Orbit Space AI: Federated, Multi-Agent, and Collaborative Algorithms for Satellite Constellations

Ziyang Wang

Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:57 UTC · model grok-4.3

classification 💻 cs.RO astro-ph.IMcs.AI

keywords satellite constellationson-orbit AIfederated learningmulti-agent systemscollaborative sensingdistributed inferencespace autonomySWaP-C constraints

0 comments

The pith

Satellite constellations need federated learning, multi-agent coordination, and distributed inference to achieve on-orbit autonomy.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This survey organizes work on space AI for groups of satellites into three main approaches that together address the move from isolated spacecraft to networked, adaptive systems. Federated learning lets satellites improve shared models while keeping data local and respecting bandwidth and privacy needs. Multi-agent methods handle joint tasks such as planning routes, allocating resources, and avoiding collisions under changing connections. Collaborative sensing and inference combine observations from several satellites and split computations to fit strict power and compute budgets. Readers care because single-satellite AI cannot meet the coordination, fault tolerance, and real-time demands that arise once constellations operate as one platform.

Core claim

The emerging field of on-orbit space AI is consolidated through three complementary paradigms: federated learning for cross-satellite training, personalization, and secure aggregation; multi-agent algorithms for cooperative planning, resource allocation, scheduling, formation control, and collision avoidance; and collaborative sensing and distributed inference for multi-satellite fusion, tracking, split/early-exit inference, and cross-layer co-design with constellation networking, together with a system-level view and taxonomy that unifies collaboration architectures, temporal mechanisms, and trust models.

What carries the argument

The taxonomy of collaboration architectures, temporal mechanisms, and trust models that organizes the three paradigms of federated learning, multi-agent algorithms, and collaborative sensing and distributed inference for satellite constellations.

If this is right

Federated learning enables satellites to train models jointly without exchanging raw data while supporting local personalization.
Multi-agent algorithms permit real-time cooperative decisions for formation flying, scheduling, and collision avoidance.
Collaborative sensing improves tracking accuracy by fusing measurements from multiple satellites and reduces load through split inference.
Cross-layer co-design links AI algorithms directly to networking layers to maintain performance under variable connectivity.
The taxonomy and ongoing curation of papers give researchers a shared structure for classifying and extending new work.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Constellations that adopt these methods could reduce reliance on ground stations for routine decision making in Earth observation.
Hybrid systems that combine federated updates with multi-agent policies might let satellites adjust coordination rules on the fly from peer data.
Large-scale deployments would require explicit handling of concept drift and non-IID data across thousands of nodes beyond what current surveys emphasize.
Hardware-in-the-loop tests under radiation could reveal whether the surveyed algorithms preserve safety guarantees when faults occur.

Load-bearing premise

The selected literature and the taxonomy of architectures, mechanisms, and trust models supply a stable and comprehensive way to unify the field that will stay useful as new methods appear.

What would settle it

A significant on-orbit coordination technique for satellite groups that cannot be placed into any of the three paradigms or the taxonomy categories would show the unification is incomplete.

read the original abstract

Satellite constellations are transforming space systems from isolated spacecraft into networked, software-defined platforms capable of on-orbit perception, decision making, and adaptation. Yet much of the existing AI studies remains centered on single-satellite inference, while constellation-scale autonomy introduces fundamentally new algorithmic requirements: learning and coordination under dynamic inter-satellite connectivity, strict SWaP-C limits, radiation-induced faults, non-IID data, concept drift, and safety-critical operational constraints. This survey consolidates the emerging field of on-orbit space AI through three complementary paradigms: (i) {federated learning} for cross-satellite training, personalization, and secure aggregation; (ii) {multi-agent algorithms} for cooperative planning, resource allocation, scheduling, formation control, and collision avoidance; and (iii) {collaborative sensing and distributed inference} for multi-satellite fusion, tracking, split/early-exit inference, and cross-layer co-design with constellation networking. We provide a system-level view and a taxonomy that unifies collaboration architectures, temporal mechanisms, and trust models. To support community development and keep this review actionable over time, we continuously curate relevant papers and resources at https://github.com/ziyangwang007/AI4Space.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This survey organizes on-orbit space AI into three paradigms with a taxonomy and a GitHub repo for updates, making it a practical reference rather than a source of new results.

read the letter

This paper is a survey that frames AI for satellite constellations around federated learning for training and aggregation, multi-agent methods for planning and control, and collaborative sensing for fusion and distributed inference. It adds a taxonomy that groups collaboration architectures, temporal mechanisms, and trust models, plus a GitHub link for continuous curation of papers and resources. The abstract lays out the shift from single-satellite work to constellation-scale issues like dynamic links, radiation faults, non-IID data, and safety constraints. That framing is the main contribution. It pulls scattered literature into one system-level view, which can help researchers see connections across Earth monitoring or resilient infrastructure applications. The repo is a concrete step that keeps the review from going stale quickly. Soft spots are the usual ones for a survey. The taxonomy's usefulness depends on how well the selected papers represent the field and whether the categories avoid forcing fits on future work. There are no new derivations, experiments, or code to verify, so soundness rests on accurate citation and logical grouping. The stress-test note is right that no internal contradictions show up in the description, but a full read would still need to check for coverage gaps. This is for people working at the AI-space intersection who want an organized entry point or help setting research priorities. A reader looking for a structured overview rather than a single new algorithm would get value. I would send it for peer review. It does the synthesis job cleanly and the update mechanism adds real utility without overclaiming.

Referee Report

0 major / 2 minor

Summary. This survey consolidates the emerging field of on-orbit space AI for satellite constellations by organizing existing literature into three complementary paradigms: (i) federated learning for cross-satellite training, personalization, and secure aggregation; (ii) multi-agent algorithms for cooperative planning, resource allocation, scheduling, formation control, and collision avoidance; and (iii) collaborative sensing and distributed inference for multi-satellite fusion, tracking, split/early-exit inference, and cross-layer co-design with constellation networking. It supplies a system-level view together with a taxonomy of collaboration architectures, temporal mechanisms, and trust models, and maintains a GitHub repository for continuous curation of papers and resources.

Significance. If the coverage is representative and the taxonomy is actionable, the paper supplies a coherent organizational framework for constellation-scale autonomy under constraints such as dynamic inter-satellite links, SWaP-C limits, radiation faults, non-IID data, and safety requirements. The explicit, ongoing GitHub curation mechanism is a concrete strength that increases the work's utility as a living reference for the community.

minor comments (2)

[Abstract] Abstract: the curly-brace notation around paradigm names (e.g., {federated learning}) is a LaTeX artifact and should be removed in the camera-ready version.
[Taxonomy] Taxonomy section: a single summary table or diagram that cross-references the three paradigms against the dimensions of architectures, temporal mechanisms, and trust models would improve immediate usability for readers.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive and constructive review of our survey on on-orbit space AI for satellite constellations. We are pleased that the referee accurately captured the paper's organization into the three paradigms of federated learning, multi-agent algorithms, and collaborative sensing, along with the unifying taxonomy and the GitHub curation effort. We appreciate the recommendation to accept.

Circularity Check

0 steps flagged

No significant circularity in this survey paper

full rationale

This is a literature survey consolidating external works under three paradigms (federated learning, multi-agent algorithms, collaborative sensing) plus a taxonomy of architectures, temporal mechanisms, and trust models. No equations, derivations, fitted parameters, predictions, or self-referential reductions exist. The GitHub curation link supports ongoing updates but does not bear any load-bearing claim or create definitional circularity. All cited literature is external; the organizational claim remains independent of any internal construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

As a literature survey, the paper introduces no new free parameters, axioms, or invented entities; it relies on standard concepts from federated learning, multi-agent systems, and distributed inference already established in the cited literature.

pith-pipeline@v0.9.0 · 5513 in / 1200 out tokens · 56608 ms · 2026-05-10T12:57:44.711388+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

170 extracted references · 4 canonical work pages

[1]

Large constellations of small satellites: A survey of near future challenges and missions,

G. Curzi, D. Modenini, and P . Tortora, “Large constellations of small satellites: A survey of near future challenges and missions,” Aerospace7, 133 (2020)

2020
[2]

Satellite constellations – 2024 survey, trends and economic sustainability,

E. Kulu, “Satellite constellations – 2024 survey, trends and economic sustainability,” in75th International Astronautical Congress (IAC 2024),(Milan, Italy, 2024). Paper IAC-24.E6.1.13

2024
[3]

Survey of inter-satellite communica- tion for small satellite systems: Physical layer to network layer view,

R. Radhakrishnan, W. W. Edmonson, F. Afghah,et al., “Survey of inter-satellite communica- tion for small satellite systems: Physical layer to network layer view,” IEEE Commun. Surv. & Tutorials18, 2442–2473 (2016)

2016
[4]

Enhancing leo mega-constellations with inter-satellite links: Vision and challenges,

C. Wu, S. Han, Q. Chen,et al., “Enhancing leo mega-constellations with inter-satellite links: Vision and challenges,” IEEE Wirel. Commun. (2025)

2025
[5]

Software defined intelligent satellite-terrestrial inte- grated networks: Insights and challenges,

S. Yuan, M. Peng, Y. Sun, and X. Liu, “Software defined intelligent satellite-terrestrial inte- grated networks: Insights and challenges,” Digit. Commun. Networks (2022)

2022
[6]

Φsat: Artificial intelligence for earth observation,

European Space Agency, “ Φsat: Artificial intelligence for earth observation,” Web page (2020). Accessed 2026-03-03

2020
[7]

Φ-sats programme,

European Space Agency Phi-lab, “Φ-sats programme,” Web page (2020). Accessed 2026-03- 03

2020
[8]

The ϕ-sat-1 mission: The first on-board deep neural network demonstrator for satellite earth observation,

G. Giuffrida, L. Fanucci, G. Meoni,et al., “The ϕ-sat-1 mission: The first on-board deep neural network demonstrator for satellite earth observation,” IEEE Transactions on Geosci. Remote. Sens.60, 1–14 (2021)

2021
[9]

Ops-sat,

European Space Agency, “Ops-sat,” Web page (2019). Accessed 2026-03-03

2019
[10]

Opportunities and challenges of on-board ai- based image recognition for small satellite earth observation missions,

B. Chintalapati, A. Precht, S. Hanra,et al., “Opportunities and challenges of on-board ai- based image recognition for small satellite earth observation missions,” Adv. space research 75, 6734–6751 (2025)

2025
[11]

A comprehensive survey of orbital edge computing: Systems, applications, and algorithms,

Y. Zengshan, W. Changhao, G. Chongbin,et al., “A comprehensive survey of orbital edge computing: Systems, applications, and algorithms,” Chin. J. Aeronaut.38, 103316 (2025)

2025
[12]

A comprehensive survey on orbital edge computing: Systems, applications, and algorithms,

C. Wu, Y. Li, M. Xu,et al., “A comprehensive survey on orbital edge computing: Systems, applications, and algorithms,” arXiv preprint (2023)

2023
[13]

Space ai: Leveraging artificial intelligence for space to improve life on earth,

Z. Wang, “Space ai: Leveraging artificial intelligence for space to improve life on earth,” arXiv preprint arXiv:2512.22399 (2025)

work page arXiv 2025
[14]

Current ai technology in space,

J. Goodwill, C. Wilson, and J. MacKinnon, “Current ai technology in space,” inPrecision Medicine for Long and Safe Permanence of Humans in Space,(Elsevier, 2025), pp. 239–250. 20

2025
[15]

Analyzing the single event upset vulnera- bility of binarized neural networks on sram fpgas,

I. Souvatzoglou, A. Papadimitriou, A. Sari,et al., “Analyzing the single event upset vulnera- bility of binarized neural networks on sram fpgas,” in2021 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT),(IEEE, 2021), pp. 1–6

2021
[16]

Federated learning in satellite constella- tions,

B. Matthiesen, N. Razmi, I. Leyva-Mayorga,et al., “Federated learning in satellite constella- tions,” IEEE Netw.38, 232–239 (2024)

2024
[17]

Towards satellite non-iid imagery: A spectral clustering-assisted federated learning approach,

L. Zou, Y. M. Park, C. M. Thwal,et al., “Towards satellite non-iid imagery: A spectral clustering-assisted federated learning approach,” arXiv preprint (2024)

2024
[18]

Concept drift detection in image data stream: a survey on current literature, limitations and future directions,

Q.-T. Tran, N.-A. Le-Khac, and M. Bertolotto, “Concept drift detection in image data stream: a survey on current literature, limitations and future directions,” Artif. Intell. Rev. (2026)

2026
[19]

Delay-tolerant networking: an approach to interplanetary internet,

S. Burleigh, A. Hooke, L. Torgerson,et al., “Delay-tolerant networking: an approach to interplanetary internet,” IEEE Commun. Mag.41, 128–136 (2003)

2003
[20]

Practical secure aggregation for privacy-preserving machine learning,

K. Bonawitz, V . Ivanov, B. Kreuter,et al., “Practical secure aggregation for privacy-preserving machine learning,” inproceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security,(2017), pp. 1175–1191

2017
[21]

Machine learning with ad- versaries: Byzantine tolerant gradient descent,

P . Blanchard, E. M. El Mhamdi, R. Guerraoui, and J. Stainer, “Machine learning with ad- versaries: Byzantine tolerant gradient descent,” inAdvances in Neural Information Processing Systems (NeurIPS),(2017)

2017
[22]

The Federated Satellite Systems paradigm: Concept and business case evaluation,

A. Golkar and I. L. i Cruz, “The Federated Satellite Systems paradigm: Concept and business case evaluation,” Acta Astronaut.111, 230–248 (2015)

2015
[23]

Review on hardware devices and software techniques enabling neural network inference onboard satellites,

L. Diana and P . Dini, “Review on hardware devices and software techniques enabling neural network inference onboard satellites,” Remote. Sens.16, 3957 (2024)

2024
[24]

Auction-based and distributed optimization approaches for scheduling obser- vations in satellite constellations with exclusive orbit portions,

G. Picard, “Auction-based and distributed optimization approaches for scheduling obser- vations in satellite constellations with exclusive orbit portions,” inProceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022),(Inter- national Foundation for Autonomous Agents and Multiagent Systems, Online, 2022)

2022
[25]

A distributed autonomous mission planning method for the low-orbit imaging constellation,

Q. Yang, B. Song, Y. Chen,et al., “A distributed autonomous mission planning method for the low-orbit imaging constellation,” Algorithms16, 475 (2023)

2023
[26]

Distributed space debris tracking with consensus labeled random finite set filtering,

B. Wei and B. Nener, “Distributed space debris tracking with consensus labeled random finite set filtering,” Sensors18, 3005 (2018)

2018
[27]

Cooperative space object tracking using consensus-based filters,

B. Jia, K. D. Pham, E. Blasch,et al., “Cooperative space object tracking using consensus-based filters,” in17th International Conference on Information Fusion (FUSION 2014), Salamanca, Spain, July 7–10, 2014,(IEEE, 2014), pp. 1–8

2014
[28]

Consensus labeled random finite set filtering for distributed multi-object tracking,

C. Fantacci, B.-N. Vo, B.-T. Vo,et al., “Consensus labeled random finite set filtering for distributed multi-object tracking,” arXiv preprint arXiv:1501.01579 (2015)

work page arXiv 2015
[29]

Towards Federated Satellite Systems and Internet of Satellites: The federation deployment control protocol,

J. A. R. de Azua, N. Garzaniti, A. Golkar,et al., “Towards Federated Satellite Systems and Internet of Satellites: The federation deployment control protocol,” Remote. Sens.13, 982 (2021)

2021
[30]

Satflow: Scalable network planning for leo mega- constellations,

S. Cen, Q. Pan, Y. Zhu, and B. Li, “Satflow: Scalable network planning for leo mega- constellations,” in2024 IEEE 32nd International Conference on Network Protocols (ICNP),(IEEE, 2024), pp. 1–11

2024
[31]

Intelligent mission planning for au- tonomous distributed satellite systems,

S. Hilton, K. Thangavel, A. Gardi, and R. Sabatini, “Intelligent mission planning for au- tonomous distributed satellite systems,” Acta Astronaut.225, 857–869 (2024)

2024
[32]

Artificial intelligence for trusted autonomous satellite operations,

K. Thangavel, R. Sabatini, A. Gardi,et al., “Artificial intelligence for trusted autonomous satellite operations,” Prog. Aerosp. Sci.144, 100960 (2024)

2024
[33]

Communication-efficient learning of deep networks from decentralized data,

H. B. McMahan, E. Moore, D. Ramage,et al., “Communication-efficient learning of deep networks from decentralized data,” inProceedings of AISTATS,(2017)

2017
[34]

A survey and critique of multiagent deep reinforcement learning,

P . Hernandez-Leal, B. Kartal, and M. E. Taylor, “A survey and critique of multiagent deep reinforcement learning,” Auton. Agents Multi-Agent Syst.33, 750–797 (2019)

2019
[35]

Consensus-based decentralized auctions for robust task allocation,

H.-L. Choi, L. Brunet, and J. P . How, “Consensus-based decentralized auctions for robust task allocation,” IEEE Transactions on Robotics25, 912–926 (2009)

2009
[36]

Branchynet: Fast inference via early exiting from deep neural networks,

S. Teerapittayanon, B. McDanel, and H. T. Kung, “Branchynet: Fast inference via early exiting from deep neural networks,” arXiv preprint (2017)

2017
[37]

Spaceexit: Enabling efficient adaptive computing in space with early exits,

J. Liu, X. Zhu, T. Xu,et al., “Spaceexit: Enabling efficient adaptive computing in space with early exits,” in2025 USENIX Annual Technical Conference (USENIX ATC),(Boston, MA, USA, 2025)

2025
[38]

Multi-agent deep reinforcement learning: a survey,

S. Gronauer and K. Diepold, “Multi-agent deep reinforcement learning: a survey,” Artif. Intell. Rev.55, 895–943 (2022)

2022
[39]

Space-air-ground integrated network: A survey,

J. Liu, Y. Shi, Z. M. Fadlullah, and N. Kato, “Space-air-ground integrated network: A survey,” IEEE Commun. Surv. & Tutorials20, 2714–2741 (2018). 21

2018
[40]

Revolutionizing future connectivity: A contemporary survey on ai- empowered satellite-based non-terrestrial networks in 6g,

S. Mahboob and L. Liu, “Revolutionizing future connectivity: A contemporary survey on ai- empowered satellite-based non-terrestrial networks in 6g,” IEEE Commun. Surv. & Tutorials 26, 1279–1321 (2024)

2024
[41]

Artificial intelligence for satellite communication: A survey,

G. Fontanesi, F. Ortíz, E. Lagunas,et al., “Artificial intelligence for satellite communication: A survey,” IEEE Commun. Surv. & Tutorials (2025). Early access / accepted manuscript

2025
[42]

Space-air-ground integrated network (sagin): A survey,

J. Chen, H. Zhang, and Z. Xie, “Space-air-ground integrated network (sagin): A survey,” arXiv preprint (2023)

2023
[43]

A survey of next-generation computing technologies in space- air-ground integrated networks,

Z. Shen, J. Jin, C. Tan,et al., “A survey of next-generation computing technologies in space- air-ground integrated networks,” ACM Comput. Surv.56, 1–40 (2023)

2023
[44]

Advancing earth observation: a survey on ai-powered image processing in satellites,

A. Duggan, B. Andrade, and H. Afli, “Advancing earth observation: a survey on ai-powered image processing in satellites,” Eur. J. Remote. Sens.58(2025)

2025
[45]

Delay is not an option: Low latency routing in space,

M. Handley, “Delay is not an option: Low latency routing in space,” inProceedings of the 17th ACM Workshop on Hot Topics in Networks (HotNets ’18),(ACM, 2018), pp. 85–91

2018
[46]

Selecting and scheduling observations of agile satellites,

M. Lemaître, G. Verfaillie, S. Jouhaud,et al., “Selecting and scheduling observations of agile satellites,” Aerosp. Sci. Technol.6, 367–381 (2002)

2002
[47]

Satellite scheduling problems: A survey of applications in earth and outer space observation,

B. Ferrari, J.-F. Cordeau, M. Delorme,et al., “Satellite scheduling problems: A survey of applications in earth and outer space observation,” Comput. & Oper. Res.173, 106875 (2025)

2025
[48]

Delay-Tolerant Networking Architecture,

L. Torgerson, S. C. Burleigh, H. Weiss,et al., “Delay-Tolerant Networking Architecture,” RFC 4838 (2007)

2007
[49]

Contact graph routing,

S. C. Burleigh, “Contact graph routing,” NASA Tech Briefs (2011). NTRS Document ID 20120006508; Report No. NPO-45488

2011
[50]

Contact graph routing in DTN space networks: overview, enhancements and performance,

G. Araniti, N. Bezirgiannidis, E. Birrane,et al., “Contact graph routing in DTN space networks: overview, enhancements and performance,” IEEE Commun. Mag.53, 38–46 (2015)

2015
[51]

Analysis of the contact graph routing algorithm: Bounding interplanetary paths,

E. Birrane, S. Burleigh, and N. Kasch, “Analysis of the contact graph routing algorithm: Bounding interplanetary paths,” Acta Astronaut.75, 108–119 (2012)

2012
[52]

Ai/ml for mission processing onboard satellites,

M. J. Veyette, K. Aylor, D. Stafford,et al., “Ai/ml for mission processing onboard satellites,” inAIAA SCITECH 2022 Forum,(2022)

2022
[53]

Benchmarking space mission applications on the snapdragon processor onboard the iss,

J. Swope, F. Mirza, E. Dunkel,et al., “Benchmarking space mission applications on the snapdragon processor onboard the iss,” J. Aerosp. Inf. Syst.20, 807–816 (2023)

2023
[54]

Optical communications downlink from a low-earth orbiting 1.5u cubesat,

T. S. Rose, D. W. Rowen, S. D. LaLumondiere,et al., “Optical communications downlink from a low-earth orbiting 1.5u cubesat,” Opt. Express27, 24382–24392 (2019)

2019
[55]

Characterization of soft errors caused by single event upsets in cmos processes,

T. Karnik, P . Hazucha, and J. Patel, “Characterization of soft errors caused by single event upsets in cmos processes,” IEEE Transactions on Dependable Secur. Comput.1, 128–143 (2004)

2004
[56]

Evaluation and mitigation of radiation-induced soft errors in graphics processing units,

D. A. G. de Oliveira, L. L. Pilla, T. Santini, and P . Rech, “Evaluation and mitigation of radiation-induced soft errors in graphics processing units,” IEEE Transactions on Comput. 65, 791–804 (2016)

2016
[57]

Mars attacks!: Software protection against space radiation,

H. Wang, S. Myint, V . Verma,et al., “Mars attacks!: Software protection against space radiation,” inProceedings of the 22nd ACM Workshop on Hot Topics in Networks (HotNets ’23), (2023), pp. 245–253

2023
[58]

Autonomous collision avoidance system (autoca),

ESA Proceedings Database, “Autonomous collision avoidance system (autoca),” Web page (2020). Accessed 2026-03-03

2020
[59]

Spacecraft autonomous decision-planning for collision avoidance maneu- vers,

N. Bourriezet al., “Spacecraft autonomous decision-planning for collision avoidance maneu- vers,” arXiv preprint (2023)

2023
[60]

Advances and open problems in federated learning,

P . Kairouz, H. B. McMahan, B. Aventet al., “Advances and open problems in federated learning,” Foundations Trends Mach. Learn.14, 1–210 (2021)

2021
[61]

Nasa spacecraft conjunction assessment and collision avoidance best practices handbook,

National Aeronautics and Space Administration, “Nasa spacecraft conjunction assessment and collision avoidance best practices handbook,” NASA/SP-20230002470 Rev 1 (2023)

2023
[62]

Hisatfl: A hierarchical federated learning framework for satellite networks with cross-domain privacy adaptation,

L. Li, L. Zhu, and W. Li, “Hisatfl: A hierarchical federated learning framework for satellite networks with cross-domain privacy adaptation,” Electronics14, 3237 (2025)

2025
[63]

Can decentralized algorithms outperform centralized algorithms? a case study for decentralized parallel stochastic gradient descent,

X. Lian, C. Zhang, H. Zhang,et al., “Can decentralized algorithms outperform centralized algorithms? a case study for decentralized parallel stochastic gradient descent,” inAdvances in Neural Information Processing Systems,vol. 30 (2017), pp. 5330–5340

2017
[64]

Sfl-leo: Asynchronous split-federated learning design for leo satellite-ground network framework,

J. Wu, J. Zhang, Z. Lin,et al., “Sfl-leo: Asynchronous split-federated learning design for leo satellite-ground network framework,” arXiv preprint (2025)

2025
[65]

Event-triggered communication network with limited- bandwidth constraint for multi-agent reinforcement learning,

G. Hu, Y. Zhu, D. Zhao,et al., “Event-triggered communication network with limited- bandwidth constraint for multi-agent reinforcement learning,” IEEE Transactions on Neural Networks Learn. Syst.34, 3966–3978 (2023)

2023
[66]

Ground-assisted federated learning 22 in leo satellite constellations,

N. Razmi, B. Matthiesen, A. Dekorsy, and P . Popovski, “Ground-assisted federated learning 22 in leo satellite constellations,” IEEE Wirel. Commun. Lett.11, 717–721 (2022)

2022
[67]

Asynchronous federated optimization,

C. Xie, S. Koyejo, and I. Gupta, “Asynchronous federated optimization,” arXiv preprint (2019)

2019
[68]

Federated learning with buffered asynchronous aggre- gation,

J. Nguyen, K. Malik, H. Zhan,et al., “Federated learning with buffered asynchronous aggre- gation,” inProceedings of The 25th International Conference on Artificial Intelligence and Statistics (AISTATS),vol. 151 ofProceedings of Machine Learning Research(PMLR, 2022), pp. 3581–3607

2022
[69]

Fedsn: A federated learning framework over heterogeneous leo satellite networks,

Z. Lin, Z. Chen, Z. Fang,et al., “Fedsn: A federated learning framework over heterogeneous leo satellite networks,” IEEE Transactions on Mob. Comput.24, 1293–1307 (2025)

2025
[70]

A collaborative inference algorithm in low-earth-orbit satellite network for unmanned aerial vehicle,

Z. Xu, P . Zhang, C. Li,et al., “A collaborative inference algorithm in low-earth-orbit satellite network for unmanned aerial vehicle,” Drones7, 575 (2023)

2023
[71]

Bundle Protocol Version 7,

S. Burleigh, K. Fall, and E. J. Birrane, “Bundle Protocol Version 7,” RFC 9171 (2022)

2022
[72]

Bundle Protocol Security (BPSec),

E. J. Birrane and K. McKeever, “Bundle Protocol Security (BPSec),” RFC 9172 (2022)

2022
[73]

Energy-aware federated learning in satellite constellations,

N. Razmi, B. Matthiesen, A. Dekorsy, and P . Popovski, “Energy-aware federated learning in satellite constellations,” arXiv preprint (2024)

2024
[74]

Fedleo: An offloading-assisted decentralized federated learning framework for low earth orbit satellite networks,

Z. Zhai, Q. Wu, S. Yu,et al., “Fedleo: An offloading-assisted decentralized federated learning framework for low earth orbit satellite networks,” IEEE Transactions on Mob. Comput.23, 5260–5279 (2024)

2024
[75]

Rafl: Reputation-aware federated learning with hierarchical aggre- gation in leo satellite networks,

X. Xu, Y. Li, and L. Lu, “Rafl: Reputation-aware federated learning with hierarchical aggre- gation in leo satellite networks,” J. Syst. Archit.168, 103565 (2025)

2025
[76]

Secure and efficient federated learning in leo constellations using decentralized key generation and on-orbit model aggregation,

M. Elmahallawy, T. Luo, and M. I. Ibrahem, “Secure and efficient federated learning in leo constellations using decentralized key generation and on-orbit model aggregation,” arXiv preprint (2023)

2023
[77]

Federated learning: Strategies for improving communication efficiency,

J. Koneˇ cný, H. B. McMahan, F. X. Yu,et al., “Federated learning: Strategies for improving communication efficiency,” arXiv preprint (2016)

2016
[78]

Fedpaq: A communication-efficient federated learning method with periodic averaging and quantization,

A. Reisizadeh, A. Mokhtari, H. Hassani,et al., “Fedpaq: A communication-efficient federated learning method with periodic averaging and quantization,” inProceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics (AISTATS),vol. 108 ofProceedings of Machine Learning Research(PMLR, 2020), pp. 2021–2031

2020
[79]

Qsgd: Communication-efficient sgd via gradient quanti- zation and encoding,

D. Alistarh, D. Grubic, J. Li,et al., “Qsgd: Communication-efficient sgd via gradient quanti- zation and encoding,” inAdvances in Neural Information Processing Systems,vol. 30 (2017)

2017
[80]

Deep gradient compression: Reducing the communication bandwidth for distributed training,

Y. Lin, S. Han, H. Mao,et al., “Deep gradient compression: Reducing the communication bandwidth for distributed training,” inInternational Conference on Learning Representations (ICLR),(2018)

2018

Showing first 80 references.