FractalNet-Based Heterogeneous Federated Learning for Orbital Edge Intelligence in Satellite Mega-Constellations: A Wildfire Case Study

• URL: http://arxiv.org/abs/2609.00875v1
• Date: Tue, 01 Sep 2026 08:07:39 GMT
• Title: FractalNet-Based Heterogeneous Federated Learning for Orbital Edge Intelligence in Satellite Mega-Constellations: A Wildfire Case Study
• Abstract summary: mega-constellations are emerging as large-scale sensing, communication, and computation fabrics.<n>We propose a heterogeneous federated learning method based on the FractalNet architecture for orbital edge intelligence.<n> Experiments on simulated mega-constellations validate the approach across convergence, communication efficiency, energy adaptation, scheduled-pooling savings, robustness, and latency.
• Score: 0.3093890460224435
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Satellite mega-constellations are emerging as large-scale sensing, communication, and computation fabrics, yet their learning architectures remain largely inherited from terrestrial federated learning and ground-centric mission operations--- ill-suited to satellites that differ by orders of magnitude in Size, Weight, Power, and Cost (SWAP-C), radiation tolerance, link availability, and propagation delay. We propose a heterogeneous federated learning method based on the FractalNet architecture for orbital edge intelligence. We formalize contact-window-constrained, depth-heterogeneous federated optimization and introduce a distributed path scheduler that assigns model depth as a function of SWAP-C constraints, predicted inter-satellite contacts, and training statistics. To reduce message overhead and energy consumption, each tier pools updates periodically rather than at every contact opportunity, and a three-tier agentic control plane governs in-space scheduling, anomaly escalation, and policy-governed autonomy. As a case study, we apply the framework to wildfire detection, where each orbital shell naturally learns a different semantic level of situational awareness: pixel-scale thermal anomalies at low Earth orbit (LEO), regional fire-front dynamics at medium Earth orbit (MEO), and larger-scale risk propagation at geostationary or high Earth orbit (GEO/HEO). Experiments on simulated mega-constellations validate the approach across convergence, communication efficiency, energy adaptation, scheduled-pooling savings, robustness, and latency.

Related papers

• Reinforcement Learning-Guided Evolutionary Policy Optimization for Preference-Adjustable Heterogeneous Agile Earth Observation Satellite Scheduling [50.176885352114674]
  Heterogeneous agile Earth observation satellite (AEOS) scheduling requires task selection, satellite assignment, and observation sequencing.<n>This paper proposes an evolutionary policy optimization framework for heterogeneous AEOS scheduling with preference-adjustable weighted objectives.
  arXiv  Detail & Related papers  (2026-08-25T12:16:28Z)
• Space Network of Experts: Architecture and Expert Placement [53.89437186709926]
  Space data centers are envisioned as a promising platform for executing energy-intensive large language models (LLMs)<n>One key challenge, however, is the efficient distributed deployment of a large-scale LLM in a satellite network.<n>We present the Space Network of Experts (Space-XNet) framework targeting the distributed execution of a popular mixture-of-experts (MoE) model in space.
  arXiv  Detail & Related papers  (2026-05-01T08:40:31Z)
• Star-Fusion: A Multi-modal Transformer Architecture for Discrete Celestial Orientation via Spherical Topology [0.0]
  We present Star-Fusion, a multi-modal architecture that reformulates orientation estimation as a discrete topological classification task.<n>The proposed architecture employs a SwinV2-Tiny transformer backbone for photometric feature extraction, a tripartite convolutional heatmap branch for spatial grounding, and a coordinate-based for geometric anchoring.<n> Experimental evaluations on a synthetic Hipparcos-derived dataset demonstrate that Star-Fusion achieves a Top-1 accuracy of 93.4% and a Top-3 accuracy of 97.8%.
  arXiv  Detail & Related papers  (2026-04-29T12:01:41Z)
• OptiVote: Non-Coherent FSO Over-the-Air Majority Vote for Communication-Efficient Distributed Federated Learning in Space Data Centers [68.73273027298625]
  megaconstellations are driving the long-term vision of space data centers (SDCs)<n>AirComp is an in-network aggregation framework for learning free-space (FSO)<n>AirVote integrates sign gradient (SGD) with a majority-signposition modulation (PPM), where each satellite conveys local gradient by activating PPM time slots.<n>OptiVote mitigates phase-sensitive field superposition into phase-agnostic optical intensity combining.
  arXiv  Detail & Related papers  (2025-12-30T16:40:02Z)
• Bringing Federated Learning to Space [3.058685580689604]
  Federated learning offers a promising framework to conduct collaborative model training across satellite networks.<n>We introduce a comprehensive "space-ification" framework that adapts terrestrial algorithms to operate under orbital constraints.<n>Our analysis demonstrates that space-adapted FL algorithms efficiently scale to constellations of up to 100 satellites.
  arXiv  Detail & Related papers  (2025-11-18T20:16:07Z)
• Joint AoI and Handover Optimization in Space-Air-Ground Integrated Network [48.485907216785904]
  Low Earth orbit (LEO) satellite constellations offer promising solutions with global coverage and reduced latency.<n>Yet struggle with intermittent coverage and intermittent communication windows due to orbital dynamics.<n>Our three-layer design employs hybrid free-space optical (FSO) links for high-capacity satellite-to-ground communication and reliable radio frequency (RF) links for HAP-to-ground transmission.
  arXiv  Detail & Related papers  (2025-09-16T06:16:56Z)
• Accelerating Privacy-Preserving Federated Learning in Large-Scale LEO Satellite Systems [57.692181589325116]
  Large-scale low-Earth-orbit (LEO) satellite systems are increasingly valued for their ability to enable rapid and wide-area data exchange.<n>Due to privacy concerns and regulatory constraints, raw data collected at remote clients cannot be centrally aggregated.<n>Federated learning offers a privacy-preserving alternative by training local models on distributed devices and exchanging only model parameters.<n>We propose a discrete temporal graph-based on-demand scheduling framework that dynamically allocates communication resources to accelerate federated learning.
  arXiv  Detail & Related papers  (2025-09-05T03:33:42Z)
• TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observation [65.74990259650984]
  We introduce TerraFM, a scalable self-supervised learning model that leverages globally distributed Sentinel-1 and Sentinel-2 imagery.<n>Our training strategy integrates local-global contrastive learning and introduces a dual-centering mechanism.<n>TerraFM achieves strong generalization on both classification and segmentation tasks, outperforming prior models on GEO-Bench and Copernicus-Bench.
  arXiv  Detail & Related papers  (2025-06-06T17:59:50Z)
• FedSN: A Federated Learning Framework over Heterogeneous LEO Satellite Networks [18.213174641216884]
  A large number of Low Earth Orbit (LEO) satellites have been launched and deployed successfully in space by commercial companies, such as SpaceX. Due to multimodal sensors equipped by the LEO satellites, they serve not only for communication but also for various machine learning applications, such as space modulation recognition, remote sensing image classification, etc. We propose FedSN as a general FL framework to tackle the above challenges, and fully explore data diversity on LEO satellites.
  arXiv  Detail & Related papers  (2023-11-02T14:47:06Z)
• Federated learning for LEO constellations via inter-HAP links [0.0]
  Low Earth Obit (LEO) satellite constellations have seen a sharp increase of deployment in recent years. To apply machine learning (ML) in such applications, the traditional way of downloading satellite data such as imagery to a ground station (GS) is not desirable. 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.
  arXiv  Detail & Related papers  (2022-05-15T08:22:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.