Anna Arestova, M. Sc.
Design of a hybrid genetic algorithm for time-sensitive networking
20th International GI/ITG Conference on Measurement, Modelling and Evaluation of Computing Systems, MMB 2020 (Saarbrücken, 2020-03-16 - 2020-03-18)
In: Holger Hermanns, Holger Hermanns (ed.): Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2020
, , :
Engineering of next-generation Train Control and Management Systems (TCMS)
(Third Party Funds Single)Term: 2018-10-01 - 2021-09-30
Funding source: Siemens AGWith the developing technologies and methods in the field of real-time communication and the constantly increasing amount of data to be transmitted, the railway industry has jumped on the bandwagon of modernizing its processes.
In the field of railway applications, various manufacturers still provide different and mostly incompatible solutions. These solutions are specified for a certain constellation of a train, but in most cases they are not able to offer the correct functionality if the constellation of the train changes. In order to separate safety and time-critical areas from non-critical areas that e.g. offer services for passengers such as wireless LAN, separate networks with their own infrastructure must also be set up. That means more weight and costs for the train and its manufacturers.
In the area of real-time communications, time-sensitive networking (TSN) has proven to be a possible solution to overcome the problems mentioned above. It provides methods and mechanisms for Ethernet technology that enrich it with aspects of determinism and reliability.
With Time-Sensitive Networks (TSN), the safety and time critical domains can be merged with non-critical areas, so that the safety and time critical domains can still be guaranteed sufficient reliability and determinism and the needs of passengers are satisfied.
The aim of this research project is to test the suitability of TSN in the railway sector. The primary goal of the project is to analyse whether the requirements of safety and time-critical applications can be met with respect to deterministic network communication and bounded latencies and at the same time to prove that the fulfilment of the requirements of critical applications does not lead to a significant impairment of non-critical applications.