Nautilus: A Verifiable Hierarchical Federated Learning Framework for Vehicular-Edge-Cloud Systems
2026-06-22 • Distributed, Parallel, and Cluster Computing
Distributed, Parallel, and Cluster Computing
AI summaryⓘ
The authors address challenges in Federated Learning for connected vehicles, where different devices have varying resources. They created Nautilus, a system that smartly assigns tasks based on each vehicle's capacity to make learning faster and more efficient. To ensure that these assignments are fair and that participants follow instructions without revealing private data, they use a Zero-Knowledge Proof method. Their tests show Nautilus reduces communication needs and speeds up training while keeping the process trustworthy.
Federated LearningInternet of Vehicles (IoV)Resource HeterogeneityDynamic SchedulingZero-Knowledge Proof (ZKP)Compression RatioModel ConvergencePrivacy PreservationEdge ComputingTraining Efficiency
Authors
Linyang Wu, Linpeng Jia, Hanwen Zhang, Tiantian Duan, Yi Sun
Abstract
Federated Learning (FL) enables privacy-preserving collaborative learning for Internet of Vehicles (IoV) scenarios, but the extreme heterogeneity of vehicular-edge-cloud resources severely limits system efficiency. While dynamic scheduling strategies can mitigate this issue, they introduce new trust concerns: how to verify that scheduling decisions are fair, and whether clients faithfully execute optimization instructions without disclosing private data? This paper proposes Nautilus, a verifiable and efficient federated learning framework. First, we design a multi-dimensional resource-aware scheduling algorithm that dynamically allocates compression ratios and training tasks based on vehicles bandwidth, latency, and computing power, significantly improving system training efficiency. Second, to address the trust deficit in the scheduling process, we introduce a Zero-Knowledge Proof (ZKP) mechanism that ensures the fairness of scheduling strategy generation and the compliance of client execution while preserving privacy. Experimental results demonstrate that the framework effectively reduces communication overhead and accelerates model convergence while maintaining system integrity.