/SCSS_DNN_Comparison

Primary LanguageJupyter NotebookMIT LicenseMIT

Single-Channel Source Separation: Preliminary Test of Neural Network Architectures

Using the demodulation challenge with a QPSK single-carrier channel as the signal-of-interest (SOI), and an OFDM (NFFT=64, Number of subcarriers=56 16-QAM symbols, Cyclic Prefix Length=16; similar to 802.11b specifications).

   

Link to download trained model weights: Dropbox Link


Based on MIT-USAF AIA RF Challenge for Single-Channel Source Separation

Click here for the Single-Channel RF Challenge Github Repository

Click here for details on the challenge setup

Acknowledgements

Research was sponsored by the United States Air Force Research Laboratory and the United States Air Force Artificial Intelligence Accelerator and was accomplished under Cooperative Agreement Number FA8750-19-2-1000. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the United States Air Force or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.

The authors acknowledge the MIT SuperCloud and Lincoln Laboratory Supercomputing Center for providing HPC resources that have contributed to the research results reported within this paper.