Software-defined time sensitive networks (SD-TSN) for industrial automation
Loading...
Files
Accepted Version
Date
2022-01-13
Authors
Seliem, Mohamed
Pesch, Dirk
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published Version
Abstract
Industrial networks interconnect devices to support industrial processes and manufacturing systems. Industrial applications results in time-critical and/or mission-critical data traffic, which require high reliability, guarantees on the delay bounds and robust time synchronization with high precision. Time Sensitive Networking (TSN)is considered as a suitable multi-service network technology for industrial systems and applications. TSN provides greater timing accuracy, enhanced performance in reducing packet delay variation, and adds more networking determinism. However, the configurations of TSN switches presents a challenge (e.g. time-aware traffic shaping schedules). Each output port of a TSN switch has to follow timed schedule to achieve the desired performance. The schedules have to be distributed at the network initialization and have to be maintained upon new devices joining or obsolete device being removed. In this paper, we use the Software Defined Networking (SDN) paradigms to overcome the aforementioned challenges. We propose the Software-Defined Time Sensitive Networks (SD-TSN), where end devices sign contracts with the network controller, to check the availability of the network resources and to select an optimized path for the data transmission. Therefore, It achieves overall improved network resources management, while meeting QoS requirement with high reliability. Our performance analysis show that SD-TSN accurately schedule the critical time traffic resulting in a bounded end-to-end latency with very low jitter. Even in the presence of different types of data traffic, the critical data traffic meets the quality of service requirements with a maximum jitter of 6\ \mus. Moreover, SD-TSN takes 10 seconds run-time, compared to 2 minutes run-time when traditional TSN scheduling is used, to calculate the schedules of 100 TSN critical flows.
Description
Keywords
Industrial automation , Mininet , SDN , Software-defined networking , Time-sensitive network , TSN
Citation
Seliem, M. and Pesch, D. (2022) 'Software-defined time sensitive networks (SD-TSN) for industrial automation', 2022 14th International Conference on Computational Intelligence and Communication Networks (CICN), Al-Khobar, Saudi Arabia, 4-6 December. doi: 10.1109/CICN56167.2022.10008262