This dataset has been developed within the framework of the project AI-NET-PROTECT (KIS8CEL010, FKZ 16KIS1282), funded by the German Ministry of Education and Research (BMBF), and the project B5G-OPEN (grant agreement 101016663), funded by the European Commission.
Overview
The presented dataset is a set of experimental traces generated in a laboratory testbed. All samples were collected from laboratory experiments carried out within the systems laboratory of the Photonic Networks and Systems Department at Fraunhofer HHI. Each sample is comprised of an experiment from a straight-line coherent optical setup, where the configuration is defined by a permutation of the parameters in Table 1. The dataset contains QoT information, constellation diagrams and other relevant data from the physical layer such as digital signal processing (DSP) results and spectral profiles for each distinct scenario.
Table 1: Configuration table
| Parameters | Values |
|---|---|
| Modulation Format | 4-QAM and 16-QAM |
| Symbol Rate (GBd) | 28 and 32 |
| Distance (km) | b2b, 80, 160, and 240 |
| Start Center Frequency of the WDM Grid (THz) | 192.1 |
| End Center Frequency of the WDM Grid (THz) | 196.05 |
| Maximum Number of WDM Channels | 80 |
| Channel Spacing (GHz) | 50 |
| Channel Under Test Number | 1, 10,20, 30, 40, 50, 60, 70, 80 |
| Distinct random channel allocation profiles | 88 |
Tasks
The dataset can be used for the task of ML-based QoT estimation, optical spectrum processing, constellation diagram analysis, and more.
Files
The dataset files are available for download on this webpage. If you are interested in receiving the dataset, please follow the link.
Subject Keywords
Quality of Transmission, Machine Learning Estimation, Optical Spectra, Constellation Diagram, Digital Twin
Conflicts of Interest
The authors have no conflicts of interest to declare.
Related Resources
If you are interested in the preparation pipeline of these datasets, you may consider watching the recording of one of our relevant demonstrations at ECOC 2022.