Adaptive UAV-Assisted Hierarchical Federated Learning: Optimizing Energy, Latency, and Resilience for Dynamic Smart IoT Networks

2025-03-08

AI Summary

Frequent disconnections of unmanned aerial vehicles (UAVs) in dynamic intelligent Internet-of-Things (IoT) environments exacerbate training costs, degrade energy efficiency, and undermine robustness in hierarchical federated learning (HFL). Method: This paper proposes a disconnection-resilient HFL co-optimization framework. It formulates a global training cost minimization model explicitly accounting for UAV disconnection risk; designs a device-UAV affinity scoring mechanism integrating data heterogeneity, channel distance, and resource states; and introduces a Twin-Delayed Deep Deterministic Policy Gradient (TD3)-driven dynamic association algorithm coupled with a two-stage greedy UAV redeployment strategy. Convergence is guaranteed via augmented Lagrangian optimization and KL-divergence-based modeling. Results: Experiments on real-world datasets demonstrate significant reduction in total training cost, while maintaining high model accuracy and system resilience under communication disruptions.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Hierarchical Federated Learning (HFL) introduces intermediate aggregation layers, addressing the limitations of conventional Federated Learning (FL) in geographically dispersed environments with limited communication infrastructure. An application of HFL is in smart IoT systems, such as remote monitoring, disaster response, and battlefield operations, where cellular connectivity is often unreliable or unavailable. In these scenarios, UAVs serve as mobile aggregators, providing connectivity to the terrestrial IoT devices. This paper studies an HFL architecture for energy-constrained UAVs in smart IoT systems, pioneering a solution to minimize global training cost increased caused by UAV disconnection. In light of this, we formulate a joint optimization problem involving learning configuration, bandwidth allocation, and device-to-UAV association, and perform global aggregation in time before UAV drops disconnect and redeployment of UAVs. The problem explicitly accounts for the dynamic nature of IoT devices and their interruptible communications and is unveiled to be NP-hard. To address this, we decompose it into three subproblems. First, we optimize the learning configuration and bandwidth allocation using an augmented Lagrangian function to reduce training costs. Second, we propose a device fitness score, integrating data heterogeneity (via Kullback-Leibler divergence), device-to-UAV distances, and IoT device resources, and develop a twin-delayed deep deterministic policy gradient (TD3)-based algorithm for dynamic device-to-UAV assignment. Third, We introduce a low-complexity two-stage greedy strategy for finding the location of UAVs redeployment and selecting the appropriate global aggregator UAV. Experiments on real-world datasets demonstrate significant cost reductions and robust performance under communication interruptions.

Problem

• Optimize energy and latency in UAV-assisted HFL for IoT networks.
• Address UAV disconnection and dynamic IoT device communication challenges.
• Develop algorithms for learning configuration, bandwidth allocation, and UAV redeployment.

Innovation

• UAVs as mobile aggregators in HFL
• TD3-based algorithm for device-UAV assignment
• Two-stage greedy strategy for UAV redeployment