Attack detection for industrial networks

Protection through intelligent anomaly detection

Our motivation

Production networks are increasingly growing with the implementation of IoT applications. For example, widespread use of IoT devices enables ever-increasing data transparency and the application of artificial intelligence in production. In this landscape of machine communication data, manipulation and attacks on such networks of diverse devices are a high risk! Systems that monitor heterogeneous networks and include the connected production processes are not available.

 

Our Idea

The project "Hybrid AI Intrusion Prevention for Industrial Control Systems - HAIP" focuses on precisely such systems. Beyond the mere monitoring of communication on production networks, the transmitted process data is searched for anomalies. Such anomalies can be assigned to different techniques of infiltration, exploration or manipulation of networks. Using a wide variety of methods from the fields of artificial intelligence, machine learning and the integration of expert knowledge, the funded project is researching how individual techniques of cyber attacks can be identified automatically. From these techniques, logical reasoning, statistics and expert knowledge will be used to automatically decide whether a cyber attack is present.

 

In the further course of the project, techniques will be developed with which such attacks can be automatically countered by the network. In this way, a system is developed that functions independently of the manufacturers of the network components.

 

Your benefit

HAIP offers a system to secure their production, which will be further developed with partners in the long term. The required basis is developed in an experimental project framework. Our know-how in machine learning and cybersecurity in production is evaluated on real use cases.

 

The partners

Experienced partners from research and industry are involved in the HAIP project. Rhebo GmbH from Leipzig is a company that deals with the reliability of industrial control systems and offers ready-made solutions. Nexocraft GmbH is a company in the consortium that contributes experience in the integration of ML-based software in companies.

The University of Bielefeld as a research partner under the leadership of Prof. Barbara Hammer provides expertise in the field of artificial intelligence. Together with Fraunhofer IOSB-INA from Lemgo, the research partners thus create the application-oriented research in the field of cybersecurity and AI, which is then implemented in a demonstrator.

 

Visit the HAIP project on our online presence and see what the partners have to say about this project.

Andreas Bunte

Contact Press / Media

Dr.-Ing. Andreas Bunte

Gruppenleiter Symbolische Verfahren und Optimierung

Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB
Campusallee 1
32657 Lemgo, Germany

Phone +49 5261 942 90 - 22

Felix Specht

Contact Press / Media

M.Sc. Felix Specht

Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB
Campusallee 1
32657 Lemgo, Germany

Phone +49 5261 94290-34