EPSRC PETRAS 2

Integrity Checking at the Edge

(2019-2021)

Edge Computing Explainable AI Industrial IoT Vulnerability Analysis SCADA

Internet of Things (IoT) Infrastructure / Systems (IS) Edge/Fog Computing (E/FC) Data Science (DS)

Critical infrastructure systems such as future factories and water treatment plants increasingly rely on interconnected edge devices, cloud platforms, and legacy industrial control systems. This growing complexity introduces new attack surfaces and vulnerabilities that traditional perimeter-based security approaches struggle to address. The Integrity Checking at the Edge (ICE) project investigates how explainable artificial intelligence deployed on edge nodes can protect these cyber-physical environments. By analysing interactions and data flows between SCADA controllers, programmable logic controllers, and modern IoT components, the project maps potential vulnerability chains across heterogeneous industrial networks. Explainable AI techniques provide visual analytics and human-centric assurance workflows that allow operators to understand and trust the automated reasoning behind security decisions. A user-centred design methodology ensures that the tools developed are accessible to practitioners who may not have deep machine learning expertise.

The Bristol Critical Infrastructures Testbed serves as the primary evaluation environment, providing realistic demonstrators that combine legacy and modern equipment. This testbed allows the research team to validate their approach against real-world industrial configurations, testing how edge-deployed AI handles the complexity of mixed SCADA, PLC, and IoT ecosystems.

Funded by EPSRC through the PETRAS National Centre of Excellence for IoT Systems Cybersecurity, ICE contributes towards building resilient, transparent, and auditable security mechanisms for the next generation of industrial IoT deployments. The project outcomes inform best practices for securing critical national infrastructure as it undergoes digital transformation.