GLNN

This repository is the Python implementation of paper "Robust Beamforming with Gradient-based Liquid Neural Network".

Blog

English version : NotImplemented.

Chinese version : Zhihu.

Files in this repo

main.py: The main function. Can be directly run to get the results.

utils.py: This file contains the util functions. It also contains definition of system params.

net.py: This file defines and declares the GLNN and the params.

CSIdyn64.mat: An example of dataset for trial run.

requirements.txt: The requirement of the recommended environment.

Reference

Should you find this work beneficial, kindly grant it a star!

To keep abreast of our research, please consider citing:

Xinquan Wang, Fenghao Zhu, Chongwen Huang, Ahmed Alhammadi, Faouzi Bader, Zhaoyang Zhang, Chau Yuen, Merouane Debbah, "Robust Beamforming with Gradient-based Liquid Neural Network," IEEE Wireless communications Letters.

@article{glnn,
      title={Robust Beamforming with Gradient-based Liquid Neural Network},
      author={Xinquan Wang and Fenghao Zhu and Chongwen Huang and Ahmed Alhammadi and Faouzi Bader and Zhaoyang Zhang and Chau Yuen and M{\'e}rouane Debbah},
      journal={IEEE Wirel. Commun. Lett.},
      year={2024}
}

Due to the size limitation of the files, bigger dataset is available here.

More than GLNN...

We are excited to announce a novel method that utilizes Manifold Learning to optimize spectrum efficiency in beamforming in RIS-aided MIMO systems.

Compared to baseline, it can speed up the convergence by 23 times and achieves a stronger robustness!

See GMML for more information!