LightEdge-IDS: A Lightweight Real-Time Intrusion Detection System for Cyber-Physical Attacks in Water Distribution Systems
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
With the growing digitization of critical infrastructure, Water Distribution Systems have become vulnerable to complex cyber–physical attacks. Machine learning-based Intrusion Detection Systems, such as the Rectilinear Hybrid Belief Classifier, achieves strong offline accuracy. However, they are not suitable for real-time use because of their limited computing power and latency at the edge. This paper presents LightEdge-IDS, a real-time, lightweight, zone-adaptive IDS designed to secure WDS in embedded environments. The system models the WDS as a graph and uses a hybrid Graph Neural Network and Gated Recurrent Unit architecture on sensor-focused subgraphs to capture spatial and temporal dependencies. Lightweightzone-specific adapters perform localized inference and generate feature embeddings for global analysis, enabling scalable and efficient edge deployment. This design achieves high adaptability and resource efficiency, providing a practical basis for real-time anomaly detection in distributed water networks. This paper presents the architectural design of LightEdge-IDS, along with key implementation details and experimental results. A complete description of the full implementation and an extended evaluation will be presented in a future, extended version of the work.
Fields of Study & Subjects
Network Security and Intrusion Detection
Water Systems and Optimization
Smart Grid Security and Resilience
Publication details
- Journal / source
- International Conference on Computing, Communication, Control and Cyber-Physical Systems (I5CPS)
- Pages
- 1–6
- Publication date
- 26 February 2026
- Language
- EN