Flexible Timing Analysis Approach for Centralized and Distributed Configuration Models in Dynamic TSN Networks

Time-Sensitive Networking (TSN), which enables deterministic communication through a series of standards, has become essential in safety-critical domains. Recent advancements have driven the development of dynamic TSN networks, with IEEE 802.1Qcc providing both centralized and distributed configuration models to support dynamic traffic changes during operation. Even with these dynamic traffic changes, the real-time performance of time-critical Stream Reservation (SR) traffic must still be guaranteed, necessitating a specialized timing analysis method to calculate the worst-case end-to-end delay. However, conventional timing analysis methods, which depend on network topology and detailed traffic parameters like GCL scheduling tables and SR traffic routes, are unsuitable for dynamic TSN networks due to their high computational complexity and reliance on global network topology and traffic routes. Therefore, the motivation of this paper is to propose flexible methods for conducting timing analysis that adapts to dynamic traffic changes. Specifically, we present two flexible timing analysis methods, Network Calculus-based Centralized (NC-C) and Network Calculus-based Distributed (NC-D), for SR traffic in typical TSN networks that utilize the combined mechanisms of Time-Aware Shaping (TAS) and Credit-Based Shaping (CBS). The NC-C method is designed for centralized configuration models, offering fast timing analysis to enhance computational efficiency during dynamic traffic changes. The NC-D method is designed for distributed configuration models, providing localized timing analysis without relying on global network topology or traffic information to support dynamic traffic changes. Experimental results demonstrate that, compared to the state-of-the-art, the NC-C method reduces the worst-case end-to-end delay by an average of 40%, and the NC-D method increases the number of reserved SR flows by an average of 34%.