A CNN based end to end communication systems
Updated: 07/02/2019.
This repository contains source code necessary to reproduce the results presented in the following paper:
A CNN-Based End-to-End Learning Framework Towards Intelligent Communication Systems
by Nan Wu, Xudong Wang, Bin Lin, and Kaiyao Zhang, accepted to IEEE access.
Dependency
- Python (3.7.0)
- Numpy (1.15.4)
- Keras (2.2.4)
- Tensorflow (1.13.1)
AWGN channel
- use model_LBC_AWGN.py to train model at a fixed Eb/N0
- use test_model_LBC_AWGN.py to test the model at a range of Eb/N0
Rayleigh fading channel
- use model_LBC_Rayleigh.py to train model at a fixed Eb/N0
- use test_model_LBC_Rayleigh.py to test the model at a range of Eb/N0
Bursty AWGN channel
- use model_LBC_Bursty_AWGN.py to train model at a fixed Eb/N0
- use test_model_LBC_Bursty_AWGN.py to test the model at a range of Eb/N0
Differential Version
- use model_DLBC_Rayleigh.py to train model at a fixed Eb/N0
- use test_model_DLBC_Rayleigh.py to test the model at a range of Eb/N0
The differential version currently only supports n=1, adding n involves complex multiplication in high-dimensional space,and is under construction
Questions?
if you have any questions, please e-mail(zky2682810462@163.com).