Digital Twin Surveillance for CPS Manipulation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cyber-physical systems are vulnerable to manipulations and attacks, with existing detection methods focusing primarily on information flow anomalies and control logic modeling, failing to adequately address physical damage risks and manual attacks on sensors and actuators.
Innovation Solution
A computer-implemented method and surveillance arrangement using a digital twin simulation of cyber-physical systems to detect deviations in sensor/actor signal information, identifying manipulations in real-time by comparing actual and replicated signals, and applying root-cause analysis to localize and counteract threats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital twin simulation is used to detect manipulations, then detection accuracy and coverage improve, but system complexity increases
Solution Approach 1:
The patent creates a digital twin (virtual copy) of the cyber-physical system that replicates its behavior and state. This copy is used to detect manipulations by comparing actual system signals with simulated signals from the digital twin, thereby improving detection accuracy without directly modifying the physical system.
Solution Approach 2:
The surveillance arrangement is divided into separate functional modules: a digital twin unit that generates simulated signals, a comparison unit that detects deviations, and a manipulation identification unit that localizes threats. This modular segmentation manages system complexity by organizing functions into independent, manageable components.
2Loss of time
If real-time detection is implemented, then response time improves, but computational load increases
Solution Approach 1:
The system performs cyclic comparisons between actual sensor/actor signals and simulated signals from the digital twin at regular intervals. This periodic detection approach enables real-time manipulation identification while managing computational load by processing data in discrete cycles rather than continuously.
Solution Approach 2:
The digital twin pre-computes expected system behavior and signals before actual manipulation detection is needed. By having the virtual model ready and running in parallel, the system can immediately compare actual signals with pre-simulated signals, reducing response time without proportionally increasing real-time computational burden.
3Reliability
If comprehensive signal monitoring is applied, then manipulation detection capability improves, but false alarm rate increases
Solution Approach 1:
The system continuously compares actual signals with simulated signals from the digital twin and uses the deviations as feedback to identify manipulations. This feedback mechanism improves detection capability by systematically analyzing signal differences while the digital twin's accurate modeling helps distinguish true manipulations from normal variations, reducing false alarms.
Data Source
AI summary
To identify manipulations of cyber-physical-systems in real-time to avoid or prevent damages to the cyber-physical systems, it is proposed with regard to (i) a cyber-physical-system with an embedded, distributed and complex system structure and providing sensor/actor-signal-information depicting a behavior of the cyber-physical-system during operation or commissioning, and (ii) a Digital-Twin-Unit, which in the course of “Model-based Digital-Twin-Representation” of the cyber-physical-system creates and executes a digital twin replicating the behavior of the cyber-physical-system and consequently producing replicated sensor/actor-signal-information by simulating the cyber-physical-system, and when the cyber-physical-system and the Digital-Twin-Unit are run in parallel, to detect cyclically a deviation in the behavior of the cyber-physical-system by comparing information by information the sensor/actor-signal-information with the replicated sensor/actor-signal-information, to identify a manipulation of the cyber-physical-system if for each detection cycle the sensor/actor-signal-information and the replicated sensor/actor-signal-information are different and consequently the deviation is detected, and the detected deviation exceeds a threshold or tolerance value.