Two-Layer Federated Learning With Heterogeneous Model Aggregation for 6G Supported Internet of Vehicles

Abstract

The vision of the upcoming 6G technologies that have fast data rate, low latency, and ultra-dense network, draws great attentions to the Internet of Vehicles (IoV) and Vehicle-to-Everything (V2X) communication for intelligent transportation systems. There is an urgent need for distributed machine learning techniques that can take advantages of massive interconnected networks with explosive amount of heterogeneous data generated at the network edge. In this study, a two-layer federated learning model is proposed to take advantages of the distributed end-edgecloud architecture typical in 6G environment, and to achieve a more efficient and more accurate learning while ensuring data privacy protection and reducing communication overheads. A novel multi-layer heterogeneous model selection and aggregation scheme is designed as a part of the federated learning process to better utilize the local and global contexts of individual vehicles and road side units (RSUs) in 6G supported vehicular networks. This context-aware distributed learning mechanism is then developed and applied to address intelligent object detection, which is one of the most critical challenges in modern intelligent transportation systems with autonomous vehicles. Evaluation results showed that the proposed method, which demonstrates a higher learning accuracy with better precision, recall and F1 score, outperforms other state-of-the-art methods under 6G network configuration by achieving faster convergence, and scales better with larger numbers of RSUs involved in the learning process. Index Terms-Federated learning, End-edge-cloud computing, Internet of vehicles, Heterogeneous data, 6G technology I. INTRODUCTION EAMLESS and ubiquitous communication infrastructure has been a key enabler of modern intelligent services. While

2D Federated Learning for Personalized Human Activity Recognition in Cyber-Physical-Social Systems

“…In addition, different learning schemes have been integrated into the federated learing framework, to pursue an efficient distributed training process. Zhou et al [22] introduced a hierarchical federated learning structure, in which a two-layer model aggregation scheme was designed, targeting modern vehicular network infrastructure in 6G environments. Zhan et al [23] focused on the improvement of model training during the federated learning processes, based on an incentive mechanism.…”

Machine learning for next‐generation intelligent transportation systems: A survey

“…278 Furthermore, integration of ML with 6G can motivate and enable various technologies. For example, federated learning with in-network computing 279 was applied to autonomous driving 280,281 while considering security issues in 6G. In a conclusion, research about 6G ITS has just begun, and this is one of the current and future research hotspots.…”

Green Internet of Vehicles (IoV) in the 6G Era: Toward Sustainable Vehicular Communications and Networking

“…The concept of 'edge AI' has been proposed recently which pushes the network intelligence at the edge devices, and enables the AI-based learning algorithms to run at distributed edge devices [25]. The federated learning fits the edge computing architecture well due to the distributed learning model, data privacy protection function and the alleviation of communication overhead [10], [26]. However, in a vehicular edge computing network, the mobility of vehicles becomes an issue in keeping continuous information sharing between the edge and the cloud server.…”