The data is used in our paper "Robustness Enhancement of Machine Fault Diagnostic Models for Railway
Applications through Data Augmentation". 
A preprint version of our paper can be found at https://www.researchgate.net/publication/344119566_REVISED--Robustness_enhancement_of_machine_fault_diagnostic_models_for_railway_applications_through_data_augmentation 
The reviewd version will be published on Mechanical Systems and Signal Processing.

The code for this research can be found at https://github.com/quickhdsdc/Robustness-Enhancement-of-Machine-Fault-Diagnostic-Models 
The data contains four folders. The raw time series sorted according to the speed ranges and datasets data can be found in "All_speeds".
"ESZUEG" refers to the M1 dataset described in the paper. "BOSCH" refers to the M2 dataset. "LEILA" refers to the M3 dataset. "Entgleis" refers to the M4 dataset.
The extracted features can be in "All_featurespace". The input data for augmentation can be found in "Augmentation". The calculated MMD distance can be found in "MMD_env".

More details can be found in our paper and github respository.