Distributed Real-Time Anomaly Detection in Networked Industrial Sensing Systems

Abstract

Abstract-Reliable real-time sensing plays a vital role in ensuring the reliability and safety of industrial Cyber-Physical Systems (CPSs) such as wireless sensor and actuator networks. For many reasons, such as harsh industrial environments, faultprone sensors, or malicious attacks, sensor readings may be abnormal or faulty. This could lead to serious system performance degradation or even catastrophic failure. Current anomaly detection approaches are either centralized and complicated, or restricted due to strict assumptions, which are not suitable for practical large-scale Networked Industrial Sensing Systems (NISSs) where sensing devices are connected via digital communications, such as wireless sensor networks or smart grid systems. In this paper, we introduce a fully distributed general-anomalydetection (GAD) scheme, which uses graph theory and exploits spatiotemporal correlations of physical processes to carry out real-time anomaly detection for general large-scale NISSs. We formally prove the scalability of our GAD approach and evaluate the performance of GAD for two industrial applications: building structure monitoring and smart grids. Extensive trace-driven simulations validate our theoretical analysis, and demonstrate that our approach can significantly outperform state-of-the-art approaches in terms of detection accuracy and efficiency.

Real-time fault diagnosis and fault-tolerant control

“…The selected papers [16][17][18] contribute to data-driven fault diagnosis. In [16] by Zhu et al, a novel fault classification mechanism is presented by developing probabilistic principal component analyser under hidden Markov model framework.…”

“…The real-time implementation of the proposed algorithm is also illustrated. The paper [18] authored by Chen et al, deals with the distributed real-time anomaly detection in networked industrial sensing systems. The proposed method utilizes graph theory and data-driven analysis tools such as principal component analysis.…”

“…Twelve selected papers pay attention on fault diagnosis methods and applications [10][11][12][13][14][15][16][17][18][19][20][21], and the other eleven papers are concentrated on real-time fault tolerant control and applications [22][23][24][25][26][27][28][29][30][31][32]. We are going to overview the selected papers following fault diagnosis techniques and fault-tolerant control techniques sequentially.…”

“…Therefore, knowledge-based diagnosis method is also called data-driven fault diagnosis methods (e.g., see the survey paper [10,11]). The selected research papers [12][13][14][15][16][17][18][19][20][21] fall into the respective categories readily (e.g., see Table 1). …”

Anomaly Detection Based on Zone Partition for Security Protection of Industrial Cyber-Physical Systems

“…T HE rapid growth of information and communication technologies have prompted the traditional industrial control systems creating a tighter integration between physical process and cyberspace, and lead to the emergency of Industrial Cyber-Physical Systems (ICPSs) [1], [2]. Integrating the cyber domain and physical domain provides significantly less isolation for physical system from the outside world than predecessor systems, creating a greater vulnerabilities and cyber security problems [3].…”