ML Security PapersDecentralized Trust for Space AI: Blockchain-Based Federated Learning Across Multi-Vendor LEO Satellite Networks
Mohamed Elmahallawy , Asma Jodeiri Akbarfam
Published on arXiv
2512.08882
Data Poisoning Attack
OWASP ML Top 10 — ML02
Key Finding
OrbitChain reduces federated satellite learning convergence time by up to 30 hours versus single-vendor baselines while finalizing 1,000+ blockchain blocks with sub-second latency under a proof-of-authority ledger.
OrbitChain
Novel technique introduced
The rise of space AI is reshaping government and industry through applications such as disaster detection, border surveillance, and climate monitoring, powered by massive data from commercial and governmental low Earth orbit (LEO) satellites. Federated satellite learning (FSL) enables joint model training without sharing raw data, but suffers from slow convergence due to intermittent connectivity and introduces critical trust challenges--where biased or falsified updates can arise across satellite constellations, including those injected through cyberattacks on inter-satellite or satellite-ground communication links. We propose OrbitChain, a blockchain-backed framework that empowers trustworthy multi-vendor collaboration in LEO networks. OrbitChain (i) offloads consensus to high-altitude platforms (HAPs) with greater computational capacity, (ii) ensures transparent, auditable provenance of model updates from different orbits owned by different vendors, and (iii) prevents manipulated or incomplete contributions from affecting global FSL model aggregation. Extensive simulations show that OrbitChain reduces computational and communication overhead while improving privacy, security, and global model accuracy. Its permissioned proof-of-authority ledger finalizes over 1000 blocks with sub-second latency (0.16,s, 0.26,s, 0.35,s for 1-of-5, 3-of-5, and 5-of-5 quorums). Moreover, OrbitChain reduces convergence time by up to 30 hours on real satellite datasets compared to single-vendor, demonstrating its effectiveness for real-time, multi-vendor learning. Our code is available at https://github.com/wsu-cyber-security-lab-ai/OrbitChain.git
Key Contributions
- OrbitChain: a permissioned proof-of-authority blockchain framework that provides transparent, auditable provenance of model updates and rejects manipulated contributions in multi-vendor federated satellite learning
- Offloads consensus from resource-constrained LEO satellites to high-altitude platforms (HAPs), achieving sub-second block finalization (0.16s–0.35s across quorum sizes)
- Demonstrates up to 30-hour convergence improvement over single-vendor baselines on real satellite datasets while improving privacy preservation and security
🛡️ Threat Analysis
OrbitChain explicitly defends against falsified or manipulated model updates injected by adversaries (including via cyberattacks on inter-satellite or satellite-ground links) that would corrupt the global FSL model — a direct Byzantine/data-poisoning defense in federated learning. The blockchain ledger enforces auditable provenance and prevents poisoned contributions from entering global aggregation.