Decentralized Trust for Space AI: Blockchain-Based Federated Learning Across Multi-Vendor LEO Satellite Networks

TL;DR

The paper tackles the challenge of trustworthy, cross-vendor federated learning in heterogeneous LEO satellite networks, where intermittent connectivity and competitive governance raise trust and provenance concerns. It introduces OrbitChain, a two-layer framework that offloads consensus to high-altitude platform validators via a Proof-of-Authority ledger and employs a cross-vendor model optimization layer with age- and reputation-weighted secure aggregation to accelerate convergence while preserving privacy. Key contributions include a ledger-backed, tamper-evident provenance mechanism (OrbitLedger), hierarchical in-space aggregation, and sub-second finality demonstrated through simulated experiments showing up to 30 hours faster convergence than single-vendor setups. The work provides a practical, scalable path for secure multi-vendor space AI collaboration, with open-source code to enable real-world adoption and experimentation.

Abstract

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

Figures (8)

• Figure 1: An illustration of the OrbitChain Framework.
• Figure 2: Average latency comparison across quorum modes (1-of-5, 3-of-5, 5-of-5).
• Figure 3: PoA performance under 1-of-5 quorum mode. Each point represents the consensus finalization time for a block.
• Figure 4: PoA performance under 3-of-5 quorum mode.
• Figure 5: PoA performance under 5-of-5 quorum mode.