An Adversarial Filtering Framework With Time–Frequency Cross-Domain Consistency for Multivariate Time-Series Anomaly Detection
Authors:
- 12 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (11) , 24464-24476
Abstract
Most data collected in real-world scenarios is contaminated, adversely affecting the performance of existing anomaly detection methods that typically rely on clean training data to learn normal patterns. While some approaches have made progress in addressing data contamination, they still face the following challenges: 1) most methods use pixel-level constraints between the contaminated and reconstructed data as the training loss, and the contaminated data is preferentially learned due to the larger error; and 2) current anomaly detection methods based on the frequency domain do not fully utilize the complex spectral characteristics nor the cross-domain consistency of time-series data, which limits the further improvement of the anomaly detection performance. This article proposes a filter-enhanced reconstruction network with domain-adversarial training for multivariate time-series anomaly detection (FERDA), which enables the model to extract normal patterns and high-level semantic information more accurately by training with the error of filtered and reconstructed data. FERDA designs a dual-domain contamination filtering module consisting of the multihead memory module and the complex linear network. The module is trained to learn normal patterns and cross-domain information through adversarial learning of intradomain semantic consistency and interdomain distinguishability. Moreover, a two-stage contamination suppression strategy is proposed. The contamination degree of data is quantified by the difference between pre- and postfiltering, and the impact of contamination is further mitigated by two-stage training with different losses. The proposed method outperforms 23 state-of-the-art baselines in experiments on five benchmark datasets across different domains.Keywords
Funding Information
- Science and Technology Foundation of State Grid Corporation of China (5216A025000M-057-ZN)
- Beijing University of Posts and Telecommunications
This publication has 52 references indexed in Scilit:
- VAEAT: Variational AutoeEncoder with adversarial training for multivariate time series anomaly detectionInformation Sciences, 2024
- Rectifying inaccurate unsupervised learning for robust time series anomaly detectionInformation Sciences, 2024
- ATFA: Adversarial Time–Frequency Attention network for sensor-based multimodal human activity recognitionExpert Systems with Applications, 2023
- Anomaly detection for multivariate times series through the multi-scale convolutional recurrent variational autoencoderExpert Systems with Applications, 2023
- Robust anomaly detection for multivariate time series through temporal GCNs and attention-based VAEKnowledge-Based Systems, 2023
- Time-series anomaly detection with stacked Transformer representations and 1D convolutional networkEngineering Applications of Artificial Intelligence, 2023
- Contrastive autoencoder for anomaly detection in multivariate time seriesInformation Sciences, 2022
- A survey of transformersAI Open, 2022
- KfreqGAN: Unsupervised detection of sequence anomaly with adversarial learning and frequency domain informationKnowledge-Based Systems, 2022
- Random forest in remote sensing: A review of applications and future directionsISPRS Journal of Photogrammetry and Remote Sensing, 2016