Time-frequency cross-fusion-guided dual-stream clustering for anomaly detection
LIU Xiaohui
ZHONG Chao
YU Jia
XIONG Chen
MA Zhaoyang
LI Da
BAI Yunfei
MIAO Xiangtai
WANG Jing
Abstract：To address the challenges in time series anomaly detection tasks,including label scarcity,high false positive and false negative rates,and model performance vulnerability to seasonal and long-term trend variations,this paper proposes a Time-Frequency Cross-Fusion-Guided Dual-Stream Clustering Anomaly Detection Method(TFCDC).First,time-domain and frequency-domain feature vectors are extracted separately,and a cross-fusion mechanism is employed to enable complementary modeling of multi-scale information,yielding a joint time-frequency feature representation.Second,the time-frequency features are encoded using a Long Short-Term Memory(LSTM)network and a linear transformation layer to produce two latent variables.Third,a two-stage clustering strategy is ap-plied in the latent spaces to aggregate normal samples while distinguishing anomalous ones,which ef-fectively reduces false positive and false negative rates.Finally,comparative experiments are con-ducted against state-of-the-art baseline models on six benchmark anomaly detection datasets.The ex-perimental results demonstrate that the proposed TFCDC model outperforms mainstream baselines on multiple dataset,achieving an average F1-score improvement of 14.5%across the six datasets and a notable 32.8%improvement on the ultra-high-dimensional,ultra-long-sequence WADI dataset com-pared to the baseline.These findings confirm that TFCDC exhibits superior accuracy and robustness,effectively mitigating the interference from long-term trend variations and anomalous data.
Keywords：deep learningtime-seriesanomaly detectiontime-frequency fusionclustering
Publication Date：2025-12-30
Online Publishing Date：2026-02-02(First online date of this platform, not the publication date of the document)
Pages：9( 156-164 )