Codes-for-Deep-Transfer-Learning-Based-Downlink-Channel-Prediction-for-FDD-Massive-MIMO-Systems

Data-Generation-and-Codes-for-Deep-Transfer-Learning-Based-Downlink-Channel-Prediction-for-FDD-Massive-MIMO-Systems

Notice

In order to use the datasets/codes or any (modified) part of them, please cite

1. The corresponding paper: Yang, Y., Gao, F., Zhong, Z., Ai, B., A. Alkhateeb. (2019). Deep Transfer Learning Based Downlink Channel Prediction for FDD Massive MIMO Systems.
2. The Remcom Wireless InSite website: Remcom, Wireless insite.

Unfold to see bibtex codes

@article{yang2019deep,
  title={Deep Transfer Learning Based Downlink Channel Prediction for FDD Massive MIMO Systems},
  author={Y. Yang and F. Gao and Z. Zhong and B. Ai  and A. Alkhateeb},
  journal={arXiv preprint arXiv:1912.12265},
  year={2019}
}
@unpublished{timmurphy,
title={Remcom Wireless InSite},
note = {\url{https://www.remcom.com/wireless-insite-em-propagation-software}}
}

Dependencies

This code requires the following: python 3.*, TensorFlow v1.4+

Data Generation

To access the datasets (i.e., samples_target64_1036_2.mat and samples_source64_1552_2.mat), please click here.
Although the orginal output files (i.e., *.p2m) are too large to be uploaded, we provide the DataPreProcess.py to demonstrate the datasets generation for readers' convenience。
More questions about the data generation, please contact: yyw18@mails.tsinghua.edu.cn.

Usage instructions

They also provide the foundation to reproduce the other results

Reproduce the Demo Result

Reproducing The Figure:

Related sources

Not finished yet, please wait for updates!