ML Propagation Loss Model

The Machine Learning based Propagation Loss (MLPL) model is a module for ns-3 that uses ML to train a propagation loss model to reproduce the physical conditions of an experimental testbed
0/5 (0 reviews)

About MLPL

The Machine Learning based Propagation Loss (MLPL) model is a module for the network simulator 3 (ns-3) that uses machine learning to train a propagation loss model to reproduce the physical conditions of an experimental testbed. The propagation loss model is trained with network traces collected in the experimental testbed that shall be reproduced.

Copyright (C) 2020-2023, INESC TEC. This project is licensed under the terms of the European Union Public Licence (EUPL) v1.2

Authors

This project is authored by:

  • Eduardo Nuno Almeida [INESC TEC, Portugal] [eduardo.n.almeida (at) inesctec.pt ; enmsa (at) outlook.pt]
  • Mohammed Rushad, Sumanth Reddy Kota, Akshat Nambiar, Hardik L. Harti, Chinmay Gupta, Danish Waseem, Mohit P. Tahiliani [NITK Surathkal, India]
  • GonΓ§alo Santos, Helder Fontes, Rui Campos, Manuel Ricardo [INESC TEC, Portugal]

Releases

All stable releases of the MLPL module are available in the GitLab Releases page.

ML Propagation Loss Models

The MLPL module contains two models depending on the method to train and predict the propagation loss.

  • D-MLPL: Distance-based ML Propagation Loss Model
    • Train and predict the propagation loss according to the distance between the nodes.
  • P-MLPL: Position-based ML Propagation Loss Model
    • Train and predict the propagation loss according to the positions of the nodes.

They are both implemented in the same set of files. The term MLPL is used to refer the generic model, regardless of the specific sub-model used. The specific sub-models are referred by their acronyms.

Publications

The list of publications in international conferences and journals about the MLPL model is available in the following page:

List of Publications

Cite This Module

If you would like to use this module, please cite it as follows:

Eduardo Nuno Almeida, Helder Fontes, Rui Campos, and Manuel Ricardo. 2023. Position-Based Machine Learning Propagation Loss Model Enabling Fast Digital Twins of Wireless Networks in ns-3. In Proceedings of the 2023 Workshop on ns-3 (WNS3 '23). ACM, 69–77. https://doi.org/10.1145/3592149.3592150

πŸ“Š Presentation Slides 🎞️ Presentation Video

πŸ† Best Paper Award πŸ† ACM Artifacts Available Badge

BibTex ``` @inproceedings{almeida2023position, author = {Almeida, Eduardo Nuno and Fontes, Helder and Campos, Rui and Ricardo, Manuel}, title = {Position-Based Machine Learning Propagation Loss Model Enabling Fast Digital Twins of Wireless Networks in ns-3}, year = {2023}, isbn = {9798400707476}, publisher = {ACM}, booktitle = {Proceedings of the 2023 Workshop on ns-3}, pages = {69–-77}, doi = {10.1145/3592149.3592150}, } ```

1.0.2

Works with ns-3.41

Latest Build :

Build History : ML Propagation Loss Model 1.0.2

Release Notes

<p>Release date: 13 May 2024</p> <ul> <li>Fix compatibility issues with ns-3.41.</li> <li>General fixes and improvements.</li> </ul>

1.0.1

Works with ns-3.41

Latest Build :

Build History : ML Propagation Loss Model 1.0.1

Release Notes

<p>Release date: 16 November 2023</p> <ul> <li>Update installation instructions to use ns3-ai v1.2.0, while MLPL does not support the latest version of ns3-ai.</li> <li>Add Python Black and isort code formatting.</li> <li>General fixes and improvements.</li> </ul>

1.0

Works with ns-3.40

Latest Build :

Build History : ML Propagation Loss Model 1.0

Release Notes

<p>Release date: 19 October 2023</p> <ul> <li>First stable release of MLPL.<ul> <li>Implementation of the Position-based MLPL (P-MLPL) model. More information about this model on the <a href="https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/blob/master/doc/publications.md">paper</a> published in the Workshop on ns-3 (WNS3) 2023.</li> <li>Implementation of the Distance-based MLPL (D-MLPL) model. More information about this model on the <a href="https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/blob/master/doc/publications.md">paper</a> published in the Workshop on ns-3 (WNS3) 2022.</li> </ul> </li> <li>General fixes and improvements.</li> <li>Update documentation.</li> </ul>

Dependencies

The MLPL module depends on other ns-3 App Store modules. The list of dependencies and their installation instructions are explained in the [MLPL's README](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/blob/master/README.md#setup-ns-3-and-dependencies). # MLPL Setup The instructions to download and setup the MLPL module are described in the [MLPL's README](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/blob/master/README.md#mlpl-setup).

### Bug Reports Bug reports or feature suggestions can be made by creating an issue in the [issue tracker](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/issues). ### Contributions Contributions to this module are welcome. If you would like to contribute to this module, feel free to open a [merge request](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/merge_requests). Contributions to this module are subject to the [license](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/blob/master/LICENSE) terms defined in this repository.

### Repository This module is open-source and developed in the [MLPL GitLab repository](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model). ### Releases The official releases of the MLPL module are stored in the [MLPL GitLab Releases](https://gitlab.com/inesctec-ns3/ml-propagation-loss-model/-/releases) page.

This ns-3 extension is one or more contributed modules.

Version 1.0.2

Released May 13, 2024

Works with ns-3.41