Review of Network Anomaly Detection in the High-speed Railway Signal System Based on Artificial Intelligence

Liu, Siyuan; Wang, Jing

doi:10.1109/ccai57533.2023.10201304

Cited by 1 publication

(1 citation statement)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI methods have huge potential to solve practical problems in the railway industry related to train automation, the maintenance of track and catenary systems, the monitoring of civil structures, etc. [6][7][8][9]. There is an increasing number of contributions in the literature applying AI methods in railway operations, such as using renewable energy and smart grid systems [10], digital twins for smart railways [11,12], automatic train operation [13,14], trackside predictive maintenance operations [15][16][17], the dynamic forecasting of train dwell time [18], transport system modelling [19], etc.…”

Section: Introductionmentioning

confidence: 99%

Railway Cloud: Management and Orchestration Functionality Designed as Microservices

Atanasov,

Pencheva,

Trifonov

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

The cloudification and virtualisation of railway functions have the potential to improve railway operation efficiency, reliability, safety, and security, as well as to enhance passenger experience by offering innovative services. This paper considers issues related to the management and orchestration of railway clouds that host cloudified railway functions. A microservices-based approach to the design of railway cloud management and orchestration functionality is proposed. The basic railway cloud concepts were defined, and functionality related to the basic orchestration of the railway cloud and deployments is analysed in order to derive the requirements of platform resources and workload management. This functionality is further designed in the form of microservices, meaning that they could possibly be used in orchestration applications to enable improvements in scalability, fault isolation, and data security. The design of microservices follows the principles of the Representational State of Transfer (REST) application programming interface (API) as a set of interlinked resources. Resources related to railway cloud orchestration are identified with their associated data, relationships to other resources, and applicable methods. The resources’ methods are used in railway applications to implement the required orchestration functionality and to maintain the state of railway cloud orchestration processes. To verify the synthesised microservices, the common orchestration application logic and microservices’ logic were modelled, and it was proved that the orchestration processes, which run concurrently, expose equivalent behaviour. The proposed approach was validated using a simulation, aiming to evaluate injected latency as a key performance indicator for the reliability and safety of railway operations. Additionally, some safety and security issues related to railway cloud management and orchestration are considered.

show abstract