Research Paper#Federated Learning, Space Data Centers, Free-Space Optical Communication🔬 ResearchAnalyzed: Jan 3, 2026 15:52
OptiVote: FSO-Based Federated Learning for Space Data Centers
Published:Dec 30, 2025 16:40
•1 min read
•ArXiv
Analysis
This paper addresses the challenge of enabling efficient federated learning in space data centers, which are bandwidth and energy-constrained. The authors propose OptiVote, a novel non-coherent free-space optical (FSO) AirComp framework that overcomes the limitations of traditional coherent AirComp by eliminating the need for precise phase synchronization. This is a significant contribution because it makes federated learning more practical in the challenging environment of space.
Key Takeaways
- •Proposes OptiVote, a non-coherent FSO AirComp framework for federated learning in space data centers.
- •Utilizes signSGD, majority-vote aggregation, and PPM for communication efficiency and robustness.
- •Eliminates the need for precise phase synchronization, addressing a key challenge in space environments.
- •Introduces a CSI-free dynamic power control scheme to mitigate aggregation bias.
- •Provides theoretical analysis and convergence guarantees.
Reference
“OptiVote integrates sign stochastic gradient descent (signSGD) with a majority-vote (MV) aggregation principle and pulse-position modulation (PPM), where each satellite conveys local gradient signs by activating orthogonal PPM time slots.”