Artificial Intelligence in Rail Transit Wireless Communication Systems: Status, Challenges and Solutions

Junhui Zhao,Xu Gao,Zhengyuan Wu,Qingmiao Zhang,Haitao Han

DOI: https://doi.org/10.1016/j.phycom.2024.102484

IF: 2.379

2024-01-01

Physical Communication

Abstract:With the continuous evolution of communication technologies such as 5G/6G and the continuous development of artificial intelligence (AI), rail transit wireless communication systems have seen unprecedented growth opportunities. However, this is accompanied by a series of challenges, including the accuracy of channel estimation in high-speed mobile environment, the complexity of resource management, and edge collaborative optimization. The aim of this paper is to explore these issues in depth and propose corresponding solutions. Firstly, we integrate AI with rail transit wireless communication to build relevant architectures and summarize the solutions to the rail transit wireless communication problems based on AI algorithms and the related research progress. Secondly, we apply AI algorithms to improving the stability of channel estimation in complex and changing channel environments, so as to enhance the communication quality. Finally, to meet the demands of rail transit wireless communication, we introduce a resource management and edge collaborative optimization model, and explore the prospects of the wide application of multiple AI algorithms in these fields. In this paper, significant progress has been made in channel estimation, resource management and edge collaborative optimization through in-depth research and innovation combined with AI algorithms. This lays the foundation for introducing more efficient and reliable communication solutions for intelligent rail transit systems.