Technology Category
Machine Learning: Calibration & Uncertainty QuantificationKnowledge Representation and Reasoning: Diagnosis and Abductive ReasoningPlanning, Routing, and Scheduling: Activity and Plan Recognition
An Unsupervised Deep XAI Framework for Localization of Concurrent Replay Attacks in Nuclear Reactor Signals
🤖 AI Summary
Addressing the challenge of detecting concurrent replay attacks in multivariate, nonstationary, and nonlinear time-series data from advanced nuclear reactors, this paper proposes an unsupervised deep explainable AI framework. The method integrates an autoencoder with an enhanced window-based SHAP (wSHAP) algorithm to achieve label-free attack detection, source identification, precise temporal localization, and attack-type classification—demonstrated for the first time on real-world nuclear operational data from the PUR-1 research reactor. It overcomes key limitations of conventional approaches, including reliance on synthetic datasets, Gaussian noise assumptions, and linear time-invariant modeling, thereby enabling full characterization of attack signatures and interpretable predictions under complex dynamic process conditions. Experimental results show >95% detection accuracy for concurrent replay attacks across six sensor channels, with accurate identification of attack count, exact onset/offset timestamps, and originating measurement channels.
Technology Category
Application Category
📝 Abstract
Next generation advanced nuclear reactors are expected to be smaller both in size and power output, relying extensively on fully digital instrumentation and control systems. These reactors will generate a large flow of information in the form of multivariate time series data, conveying simultaneously various non linear cyber physical, process, control, sensor, and operational states. Ensuring data integrity against deception attacks is becoming increasingly important for networked communication and a requirement for safe and reliable operation. Current efforts to address replay attacks, almost universally focus on watermarking or supervised anomaly detection approaches without further identifying and characterizing the root cause of the anomaly. In addition, these approaches rely mostly on synthetic data with uncorrelated Gaussian process and measurement noise and full state feedback or are limited to univariate signals, signal stationarity, linear quadratic regulators, or other linear-time invariant state-space which may fail to capture any unmodeled system dynamics. In the realm of regulated nuclear cyber-physical systems, additional work is needed on characterization of replay attacks and explainability of predictions using real data. Here, we propose an unsupervised explainable AI framework based on a combination of autoencoder and customized windowSHAP algorithm to fully characterize real-time replay attacks, i.e., detection, source identification, timing and type, of increasing complexity during a dynamic time evolving reactor process. The proposed XAI framework was benchmarked on several real world datasets from Purdue's nuclear reactor PUR-1 with up to six signals concurrently being replayed. In all cases, the XAI framework was able to detect and identify the source and number of signals being replayed and the duration of the falsification with 95 percent or better accuracy.
Problem
Research questions and friction points this paper is trying to address.
Detect and localize concurrent replay attacks in nuclear reactor signals
Address lack of explainability in unsupervised anomaly detection methods
Handle dynamic, multivariate time series data with complex noise patterns
Innovation
Methods, ideas, or system contributions that make the work stand out.
Unsupervised deep XAI framework for replay attacks
Autoencoder and windowSHAP for attack characterization
Real-time detection with 95% accuracy
🔎 Similar Papers
No similar papers found.
