iADCPS: Time Series Anomaly Detection for Evolving Cyber-physical Systems via Incremental Meta-learning
This addresses the need for adaptive anomaly detection in evolving cyber-physical systems, which is crucial for identifying faults and attacks, though it appears incremental in its method improvements.
The paper tackles the problem of anomaly detection in evolving cyber-physical systems by proposing iADCPS, an incremental meta-learning approach that adapts to data distribution shifts, achieving F1-Scores of 99.0%, 93.1%, and 78.7% on three datasets and outperforming state-of-the-art methods.
Anomaly detection for cyber-physical systems (ADCPS) is crucial in identifying faults and potential attacks by analyzing the time series of sensor measurements and actuator states. However, current methods lack adaptation to data distribution shifts in both temporal and spatial dimensions as cyber-physical systems evolve. To tackle this issue, we propose an incremental meta-learning-based approach, namely iADCPS, which can continuously update the model through limited evolving normal samples to reconcile the distribution gap between evolving and historical time series. Specifically, We first introduce a temporal mixup strategy to align data for data-level generalization which is then combined with the one-class meta-learning approach for model-level generalization. Furthermore, we develop a non-parametric dynamic threshold to adaptively adjust the threshold based on the probability density of the abnormal scores without any anomaly supervision. We empirically evaluate the effectiveness of the iADCPS using three publicly available datasets PUMP, SWaT, and WADI. The experimental results demonstrate that our method achieves 99.0%, 93.1%, and 78.7% F1-Score, respectively, which outperforms the state-of-the-art (SOTA) ADCPS method, especially in the context of the evolving CPSs.