Colloquium lecture: Transmission failure testing with a Digital Twin of a real-world optical network element, April 14, 2026, 10:15 a.m. (Tutor: Mahdi Dibaei)

Bild Besprechungsraum 04.137

As Optical Networks grow in scale and complexity, the network elements making up those networks are developed into highly complex central nodes. To ensure functionality and stability in production use, testing and validation have to be performed. Testing on real network elements is limited by the availability and flexibility of physical devices, while tests on purely simulated systems have to be designed by hand and may not account for all edge cases.

To create simulated test conditions identical to the real world this thesis investigates Digital Twins. A Digital Twin System that mirrors the defect state of a real Network Element into the digital realm is designed and implemented.
Two different approaches are investigated and implemented to transfer the defect state to the Digital Twin, one utilizing a Network Analyzer, the other transferring the state directly. Additionally options for storing and replaying the state are investigated.
By supplying the state to multiple Digital Twins tests can be executed in parallel, enabling a higher utilization of the existing hardware.

To verify the functionality of the Digital Twin, three validation scenarios of inserting errors and defects are tested. They show that both systems can transfer defect states to the Digital Twin, but the timing precision and the error propagation differ between the approaches.
The successful propagation of the defect state to a Digital Twin demonstrates the feasibility of the examined approaches, with performance measurements enabling a quantitative analysis of the designed system.
Further tests show, that parallel testing on the same state is feasible with the direct state transfer system, with multiple Digital Twins successfully receiving the synchronized state.

Time: 10:15 a.m.

Room 04.137, Martensstr. 3, Erlangen

or

Zoom-Meeting:
https://fau.zoom-x.de/j/68350702053?pwd=UkF3aXY0QUdjeSsyR0tyRWtLQ0hYUT09

Meeting-ID: 683 5070 2053
Kenncode: 647333