As a major category of unsupervised anomaly detection methods for multivariate time series, autoregression-based methods train a predictor to model the normal pattern from only normal time series, and then detect anomalies by prediction error. However, we find that the discrepancy of input data between the training and inference stages is a crucial challenge, which may lead to volatile results when detectors take abnormal time series as input. Furthermore, the correlations among multiple sensors are intricate, where irrelevant sensors may bring noise dependencies. This article proposes an autoregression-based time series anomaly detection method named DUal Masked self-Attention (DUMA). First, we propose a block-mask mechanism to enhance the robustness of the predictor for abnormal input data. Then a max-mask self-attention is proposed to reduce the noise dependencies between irrelevant sensors. Experiments on three cyber-physical systems (CPSs) datasets show that DUMA outperforms the state-of-the-art baseline methods.
|Number of pages||10|
|Journal||IEEE Sensors Journal|
|State||Published - Feb 1 2023|
Bibliographical noteFunding Information:
This work was supported in part by the NSFC under Grant 62136002, in part by the National Key Research and Development Program of China under Grant 2021YFC3340700, and in part by the Shanghai Trusted Industry Internet Software Collaborative Innovation Center.
© 2001-2012 IEEE.
- Anomaly detection
- cyber-physical systems (CPSs)
- multivariate time series
ASJC Scopus subject areas
- Electrical and Electronic Engineering