Machine learning based intrusion detection approaches for industrial IoT control systems : a review - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Machine learning based intrusion detection approaches for industrial IoT control systems : a review

John Georgakis
  • Fonction : Auteur

Résumé

Industrial control systems are subject to cyber attacks becoming more frequent and the consequences can be potentially disastrous. Intrusion detection is one of the tools used in the risk management process. Specific approaches have been developed for industrial systems. Most of them are designed to detect abnormalities between a reference model and observed behavior , one of the differences with IT systems is that this model may include one of the controlled physical system. The construction of this model is a key step and the use of learning techniques appears as an attractive solution. This article reviews the approaches proposed in the literature and discuss their possibilities and limitations.
Fichier non déposé

Dates et versions

hal-01978684 , version 1 (11-01-2019)

Identifiants

  • HAL Id : hal-01978684 , version 1

Citer

Jean-Marie Flaus, John Georgakis. Machine learning based intrusion detection approaches for industrial IoT control systems : a review. International Conference on Industrial Internet of Things and Smart Manufacturing, Sep 2018, Londres, United Kingdom. ⟨hal-01978684⟩
182 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More