Skip to main content Skip to main navigation

Publication

Enhancing Security in IIoT-Edge: A Resilient DDoS Defence Framework for Future Networks

Ihab Alzalam; Afnan Aghai; Christoph Lipps; Hans Dieter Schotten
In: IEEE Future Networks World Forum 2026. IEEE Future Networks World Forum (FNWF-2026), October 26-28, Lisbon, Portugal, IEEE, 10/2026.

Abstract

Fifth Generation (5G) and Beyond 5G (B5G) networks enable the deployment of distributed edge-cloud architectures that support latency-critical industrial applications and large-scale Industrial Internet of Things (IIoT) ecosystems. While these architectures improve operational efficiency and scalability, their distributed nature and exposure to heterogeneous traffic patterns increase the attack surface for cyber threats, particularly Distributed-Denial-of-Service (DDoS) attacks. This work presents a practical framework for real-time DDoS detection and mitigation in IIoT edge-cloud environments. A number of machine learning (ML) models are trained and evaluated on the Edge-IIoTset dataset. XGBoost is identified as the most suitable model for deployment and is then deployed in a custom OpenStack-based testbed, where simple and non-invasive TCP, ICMP, and HTTP DDoS attacks are emulated. The model is linked with a window-based mitigation mechanism. Real-time results show that the framework provides stable and reliable detection for the emulated TCP and ICMP flood attacks, with consistent detection and blocking actions against malicious traffic. For the emulated HTTP flood traffic, the system correctly identifies attack hosts; however, application-layer traffic proves more difficult to differentiate from legitimate traffic, resulting in lower flow-level recall and higher resource consumption.

Projects