Machine learning techniques for Intrusion Detection in SCADA Systems

In this work, we aim at developing a NIDS (Network Intrusion Detection System) that detects attacks targeting SCADA systems, in a concrete industrial used case scenario.In order to achieve this goal, various machine learning approaches, such as Support Vector Machines (SVMs) and Random Forest (RF), are used to build this intrusion detection system along with a deep learning algorithm called Long Short Term Memory (LSTM).

These pre-processing files were implemented to work with a Gas Pipeline Control System dataset (IanArffDataset.arff). The dataset is hosted in the Industrial Control System (ICS) Cyber Attack Datasets website:

https://sites.google.com/a/uah.edu/tommy-morris-uah/ics-data-sets http://www.ece.uah.edu/~thm0009/icsdatasets/IanArffDataset.arff

If you want to cite our work, please use the following BibTex entry:

@inproceedings{lopez:2018:inproceedings:scada-ml,
  author    = {Lopez Perez, Rocio and Adamsky, Florian and Soua, Ridha and Engel Thomas},
  title     = {{Machine Learning for Reliable Network Attack Detection in SCADA Systems}}
  publisher = {{IEEE}},
  year      = {2018},
  booktitle = {Proceedings of the 17th IEEE International Conference On Trust, Security And Privacy In Computing And Communications (IEEE TrustCom-18)}
}