Dataset Title:  Interference Classification in TN-NTN Integrated Networks

Dataset Description: Received signal realizations in frequency domain considering 3 classes:
Class 0 - No interference, Class 1 - NTN interference and Class 2 - TN interference.
The resulting dataset contains nearly 30k labeled signal chunks 

This dataset has been used in this work (please cite this reference in your work if you make use of this dataset):

• N. Abishaik-Ravindran, S. Briseno-Ramirez, F. Ortiz, E. Lagunas, S. Chatzinotas, "Interference Classification in TN-NTN Dynamic Spectrum Sharing Environment", submitted to IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN)

Detailed description:

The simulator generated signal data for the three different LEO satellite configurations for all three interference scenarios. Signal Data is in the X.npy file with the following shape: (33033, 8192, 3). Each row in the numpy array is a three channel 8192 bin FFT of the discretized signal chunk of 0.25s for a LEO satellite downlink containing either no interference, terrestrial interference or non-terrestrial interference from another LEO satellite. For training, the type of interference can be identified by the corresponding label in the Y.npy which is one-hot encoded:

Class 0: No interference

Class 1: NTN interference

Class 2: TN interference

The three channels of the FFT are:

Magnitude

Sine of Phase

Cosine of Phase

DOWNLOAD DATA