Learning Based Dynamic Cluster Reconfiguration for UAV Mobility Management with 3D Beamforming

This repository is accompanying the paper "Learning Based Dynamic Cluster Reconfiguration for UAV Mobility Management with 3D Beamforming" (I. Meer, K.-L. Besser, M. Ozger, D. Schupke, V. Poor, C. Cavdar. In Proceedings of the 2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN), pp. 486-491, May 2024), doi:10.1109/ICMLCN59089.2024.10625071, arXiv:2402.00224.

The idea is to give an interactive version of the calculations and presented concepts to the reader. One can also change different parameters and explore different behaviors on their own.

File List

The following files are provided in this repository:

baseline.py: Python module that implements the baseline algorithms for comparison
beamforming.py: Python module that implements the 3D beamforming
data_logger.py: Python module that contains the callback to store data during training
environment.py: Python module that contains the gym environment of the considered scenario
main_training.py: Main Python script that starts the training and saves the model
movement.py: Python module that implements the UAV movement based on SDEs
plot_test.py: Python script that plots the test results
reliability.py: Python module that contains the calculations of the outage probability
test.py: Python script that runs the test phase
util.py: Python module that contains utility functions.
requirements.txt: File listing all the required libraries

Usage

Running it online

You can use services like CodeOcean or Binder to run the scripts online.

Local Installation

If you want to run it locally on your machine, Python3 and Jupyter are needed. The present code was developed and tested with the following versions:

Python 3.10
numpy 1.24
scipy 1.10
matplotlib 3.7
tensorflow 2.13
tensorboard 2.13
stable-baselines3 1.7.0

Make sure you have Python3 installed on your computer. You can then install the required packages by running

pip3 install -r requirements.txt

This will install all the needed packages which are listed in the requirements file. You can then run the training and testing by issuing

python3 main_training.py
python3 test.py

Acknowledgements

This research was supported by the German Research Foundation (DFG) under grant BE 8098/1-1, by the CELTIC-NEXT Project, 6G for Connected Sky (6G-SKY), with funding received from Vinnova, Swedish Innovation Agency, and by the U.S. National Science Foundation under Grants CNS-2128448 and ECCS-2335876.

License and Referencing

This program is licensed under the GPLv3 license. If you in any way use this code for research that results in publications, please cite our original article listed above.

You can use the following BibTeX entry

@inproceedings{Meer2024icmlcn,
  author = {Meer, Irshad A. and Besser, Karl-Ludwig and Ozger, Mustafa and Schupke, Dominic and Poor, H. Vincent and Cavdar, Cicek},
  title = {Learning Based Dynamic Cluster Reconfiguration for UAV Mobility Management With 3D Beamforming},
  booktitle = {2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN)},
  year = {2024},
  month = {5},
  publisher = {IEEE},
  venue = {Stockholm, Sweden},
  pages = {486--491},
  doi = {10.1109/ICMLCN59089.2024.10625071},
  archiveprefix = {arXiv},
  eprint = {2402.00224},
  primaryclass = {cs.IT},
}