Cross-layer anomaly detection in industrial cyber-physical systems

Abstract

Within the frame of the fourth industrial revolution, also known as Industry 4.0, industrial cyber-physical production systems (ICPS) have experienced a significant progress. To this end, Industry 4.0 has brought upon an improved, flexible, and cost-efficient system architecture that can sustain the development of innovative applications and services. Nonetheless, this technological advancement also exposed ICPS to significant cyber threats. This paper contributes to the development of a cross-layer anomaly detection system (ADS) for ICPS by defining a lightweight detection methodology that leverages Dempster-Shafer's Theory of Evidence in order to: infer the system's state; fuse evidence from a wide range of monitored parameters; and deliver a comprehensive and scalable detection system. The proposed approach is validated in the context of a real natural gas transportation installation.