Automated Root Cause Analysis in Time-Sensitive Networks Based on Fault Models

Time-Sensitive networking plays a major role in the convergence of IT and OT in the use cases of Industry 4.0. The available mechanisms of TSN, such as Frame Preemption (IEEE 802.1Q), Time Synchronization (IEEE 802.1AS), and Enhancements for Scheduled Traffic (IEEE 802.1Q), make devices and networks more complex when they first start up, run, or fail. Fault detection and diagnosis require experience and expert knowledge to find the root cause of faults and troubleshoot them. However, unlike other communication technologies, there is no information about possible faults or errors, how to recognize errors, or how to handle errors in time-sensitive mechanisms. Therefore, a fully automated approach to identifying the underlying cause of a malfunction is required to aid network administrators in the event of a malfunction, thereby minimizing downtime. How can an automated root cause analysis system in time-sensitive networking be realized, and how can faulty configuration of scheduled traffic be automatically detected? This work describes a concept for automated root cause analysis in time-sensitive networks based on fault models (SARCAI-TSN), and investigates the possible symptoms of faulty Frame Preemption and Scheduled Traffic configuration with a test setup. Furthermore, it presents a scheduled traffic anomaly detection algorithm for the detection of faulty scheduled traffic configurations. This research provides assistance to both vendors and users in fault detection and diagnosis (FDD) in Time-Sensitive Networking.