Latency-aware Multimodal Federated Learning over UAV Networks

2025-10-02

AI Summary

To address high end-to-end latency and low training efficiency in multi-modal federated learning (FL) systems over unmanned aerial vehicle (UAV) networks, this paper proposes a latency-aware joint optimization framework. The framework deeply integrates multi-modal sensing with FL and jointly optimizes UAV sensing scheduling, transmit power, 3D trajectory, wireless resource allocation, and base station (BS) resource management. To tackle the resulting non-convex, tightly coupled optimization problem, we design a low-complexity iterative algorithm based on block coordinate descent (BCD) and successive convex approximation (SCA), and rigorously prove its convergence. Experimental results under diverse data heterogeneity scenarios demonstrate that the proposed method significantly reduces end-to-end latency by 32.7% on average, accelerates model convergence by 1.8×, and improves final test accuracy by 2.4%, validating its effectiveness and practicality.

Technology Category

• Machine Learning: Multimodal Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

This paper investigates federated multimodal learning (FML) assisted by unmanned aerial vehicles (UAVs) with a focus on minimizing system latency and providing convergence analysis. In this framework, UAVs are distributed throughout the network to collect data, participate in model training, and collaborate with a base station (BS) to build a global model. By utilizing multimodal sensing, the UAVs overcome the limitations of unimodal systems, enhancing model accuracy, generalization, and offering a more comprehensive understanding of the environment. The primary objective is to optimize FML system latency in UAV networks by jointly addressing UAV sensing scheduling, power control, trajectory planning, resource allocation, and BS resource management. To address the computational complexity of our latency minimization problem, we propose an efficient iterative optimization algorithm combining block coordinate descent and successive convex approximation techniques, which provides high-quality approximate solutions. We also present a theoretical convergence analysis for the UAV-assisted FML framework under a non-convex loss function. Numerical experiments demonstrate that our FML framework outperforms existing approaches in terms of system latency and model training performance under different data settings.

Problem

• Minimizing latency in UAV-assisted multimodal federated learning systems
• Optimizing resource allocation and trajectory planning for UAV networks
• Providing convergence analysis under non-convex loss functions

Innovation

• UAVs collect multimodal data for federated learning
• Joint optimization of sensing, power, trajectory and resources
• Iterative algorithm with convergence analysis for latency minimization