Abstract
The growing integration of Digital Twins (DTs) in Industry 4.0 environments exposes the physical-virtual communication layer as a critical vector for cyber vulnerabilities; while most studies focus on complex and resource-intensive security mechanisms, this work demonstrates that the inherently predictable nature of DT communications allows simple statistical metrics-such as the μ+3σ threshold-to provide robust, interpretable, and computationally efficient anomaly detection. Using a Docker-based simulation, we emulate Denial-of-Service (DoS), Man-in-the-Middle (MiTM), and intrusion attacks, showing that each generates a distinct statistical signature (e.g., a 50-fold increase in packet rate during DoS). The results confirm that data rate monitoring offers a viable, non-intrusive, and cost-effective first line of defense, thereby enhancing the resilience of IIoT-based Digital Twins.