Energy Autoencoder

Deep energy autoencoder for noncoherent communications based on the following paper.

• T. V. Luong, Y. Ko, M. Matthaiou, N. A. Vien and H. Q. Ngo, “Deep energy autoencoder for noncoherent multicarrier MU-SIMO systems”, IEEE Trans. Wireless Commun., Feb. 2020. https://ieeexplore.ieee.org/document/9036067

Three main files:

• nc_ea.py: single-user NC-EA (noncoherent energy autoencoder system)
• nc_eama_ul_2users.py: NC-EAMA uplink with 2 users (MA = multiple access)
• nc_eama_dl_2users.py: NC-EAMA downlink with 2 users

Requirements

• Tensorflow 2.0
• Keras 2.3.1
• Numpy 1.18.1