Industrial Endpoint Event Correlation for TSN Behavior Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mechanisms for tracking events in networked industrial systems, such as those using Time-Sensitive Networking (TSN), are intrusive and not applicable for end-points communicating over a network, necessitating improved methods for efficient event logging and analysis.
Innovation Solution
A method and control node that collect timestamped event log entries from end-points within a time window, identify correlated events, and provide a mapping between events and the end-points where they occur, enabling efficient tracking and analysis of events in industrial systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Tracealyzer is used to track events in embedded systems, then event tracking capability is improved, but it becomes intrusive and inapplicable to networked embedded systems
Solution Approach 1:
The system segments event tracking functionality into separate components: end-points generate and send event log entries independently, while the control node performs the analysis. This segmentation allows the solution to work in networked distributed systems without being intrusive to individual embedded components.
Solution Approach 2:
A control node acts as an intermediary that collects event log entries from multiple end-points via the network, correlates events across different end-points, and performs analysis centrally. This intermediary approach enables event tracking in networked systems without modifying the end-point systems themselves.
2Loss of information
If event log entries are collected from multiple end-points, then comprehensive event analysis is improved, but network traffic and processing load increase
Solution Approach 1:
The system collects event log entries within a defined time window rather than continuously, and focuses analysis on correlated event tuples. This partial action approach provides sufficient event analysis comprehensiveness while reducing the overall volume of data transmitted and processed.
Solution Approach 2:
End-points prepare and send event log entries with timestamps in advance, allowing the control node to perform time-window-based filtering and correlation. This preliminary preparation at the source reduces the processing burden on the control node and optimizes network traffic patterns.
Data Source
AI summary
A mechanisms for analyzing event based behavior of end-points of a network in an industrial system and a method performed by a control node, the method includes collecting, from the end-points, timestamped event log entries as created by the end-points within a time window. Each respective timestamped event log entry pertains to a respective event performed by one of the end-points. The method includes identifying tuples of correlated events by comparing the timestamped event log entries to each other, thereby providing a mapping between the events and those end-points where the events of each tuple of correlated events are performed.