Abstract

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 (FNs), 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 in predictable FNs, where future trajectories are known. Simulation results demonstrate that LinRA converges 5.2× faster than benchmarks and improves throughput by 80% in Non Line-of-Sight conditions, matching the performance of ideal algorithms. © 2025 Elsevier B.V., All rights reserved.

Abstract
Despite the trend towards ubiquitous wireless connectivity, there are scenarios where the communications infrastructure is damaged and wireless coverage is insufficient or does not exist, such as in natural disasters and temporary crowded events. Flying networks, composed of Unmanned Aerial Vehicles (UAV), have emerged as a flexible and cost-effective solution to provide on-demand wireless connectivity in these scenarios. UAVs have the capability to operate virtually everywhere, and the growing payload capacity makes them suitable platforms to carry wireless communications hardware. The state of the art in the field of flying networks is mainly focused on the optimal positioning of the flying nodes, while the wireless link parameters are configured with default values. On the other hand, current link adaptation algorithms are mainly targeting fixed or low mobility scenarios. We propose a novel rate adaptation approach for flying networks, named Trajectory Aware Rate Adaptation (TARA), which leverages the knowledge of flying nodes’ movement to predict future channel conditions and perform rate adaptation accordingly. Simulation results of 100 different trajectories show that our solution increases throughput by up to 53% and achieves an average improvement of 14%, when compared with conventional rate adaptation algorithms such as Minstrel-HT. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024.

Abstract

Abstract
Several research works have applied Reinforcement Learning (RL) algorithms to solve the Rate Adaptation (RA) problem in Wi-Fi networks. The dynamic nature of the radio link requires the algorithms to be responsive to changes in link quality. Delays in the execution of the algorithm due to implementional details may be detrimental to its performance, which in turn may decrease network performance. These delays can be avoided to a certain extent. However, this aspect has been overlooked in the state of the art when using simulated environments, since the computational delays are not considered. In this paper, we present an analysis of computational delays and their impact on the performance of RL-based RA algorithms, and propose a methodology to incorporate the experimental computational delays of the algorithms from running in a specific target hardware, in a simulation environment. Our simulation results considering the real computational delays showed that these delays do, in fact, degrade the algorithm’s execution and training capabilities which, in the end, has a negative impact on network performance. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024.