ITSC 2024 Paper Abstract

Close

Paper ThBT13.11

Liu, Lei (Beijing Jiaotong University), Zhao, Hongli (Beijing Jiaotong University)

Enhancing Handoff Performance in 5G Based High-Speed Rail Wireless Networks through Attentive Hierarchical Reinforcement Learning

Scheduled for presentation during the Poster Session "Railway systems and applications" (ThBT13), Thursday, September 26, 2024, 14:30−16:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Keywords Rail Traffic Management, Theory and Models for Optimization and Control, Transportation Security

Abstract

High-speed rail systems necessitate a robust and reliable communication network to ensure the efficient operation of train control systems. In the era of 5G technology, the demand for a high-performing network that can cater to the precise needs of railway operations is paramount. Existing rule-based handoff decision algorithms often fall short in addressing the dynamic and high-speed environment of train movements, leading to suboptimal handover success rates that are inadequate for the stringent requirements of train control systems. To overcome these limitations, we propose a cutting-edge framework that integrates an attention-enhanced hierarchical reinforcement learning network. This innovative approach is tailored to discern the intricate historical patterns of signal strength variations experienced by high-speed trains and to predict future Reference Signal Received Power (RSRP) levels with high accuracy. By doing so, it can proactively determine the optimal timing for initiating handovers, ensuring that the process is seamless and reducing the likelihood of communication failures. Our predictive model focuses on the unique challenges faced by high-speed railway communication networks, such as rapid fluctuations in signal quality and the need for uninterrupted connectivity for train control. By leveraging the predictive insights generated by our model, we can enable a more intelligent and anticipatory handover process, thus significantly improving the reliability and efficiency of the 5G networks designed for high-speed rail control systems. This approach represents a significant step forward in meeting the evolving demands of high-speed railway communications and ensuring the safety and reliability of train operations.

 

 

All Content © PaperCept, Inc.


This site is protected by copyright and trademark laws under US and International law.
All rights reserved. © 2002-2024 PaperCept, Inc.
Page generated 2024-12-26  06:29:31 PST  Terms of use