Context-aware Rate Adaptation for Predictive Flying Networks using Contextual Bandits

2025-04-08

AI Summary

In high-mobility aerial networks, conventional rate adaptation (RA) fails due to rapid node movement and abrupt channel variations. To address this, we propose Linear Upper Confidence Bound Rate Adaptation (LinRA), the first contextual bandit-based RA framework for predictive air-to-ground communication. LinRA jointly leverages pre-known flight trajectories, real-time link-state measurements, and obstacle-aware environmental information to enable proactive channel adaptation. By formulating the problem as a linear upper confidence bound (UCB) optimization and incorporating trajectory-driven predictive control, LinRA achieves both low computational complexity and high scalability. Evaluation shows that LinRA accelerates convergence by 5.2× compared to state-of-the-art baselines and improves throughput by 80% in non-line-of-sight (NLoS) scenarios, approaching the theoretical performance upper bound.

Technology Category

• Machine Learning: Online Learning & Bandits
• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

The increasing complexity of wireless technologies, such as Wi-Fi, presents significant challenges for Rate Adaptation (RA) due to the large configuration space of transmission parameters. While extensive research has been conducted on RA for low-mobility networks, existing solutions fail to adapt in flying networks, where high mobility and dynamic wireless conditions introduce additional uncertainty. We propose Linear Upper Confidence Bound for RA (LinRA), a novel Contextual Bandit-based approach that leverages real-time link context to optimize transmission rates. Designed for predictive flying networks, where future trajectories are known, LinRA proactively adapts to obstacles affecting channel quality. Simulation results demonstrate that LinRA converges $mathbf{5.2 imes}$ faster than state-of-the-art benchmarks and improves throughput by 80% in Non Line-of-Sight (NLoS) conditions, matching the performance of ideal algorithms. With low time complexity, LinRA is a scalable and efficient RA solution for predictive flying networks.

Problem

• Optimizing transmission rates in high-mobility flying networks
• Adapting to dynamic wireless conditions with contextual bandits
• Improving throughput and convergence speed in NLoS scenarios

Innovation

• Contextual Bandit-based rate adaptation
• Leverages real-time link context
• Proactively adapts to obstacles