Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (11): 3419-3426.DOI: 10.11772/j.issn.1001-9081.2023111636

• Data science and technology • Previous Articles     Next Articles

Multivariate time series anomaly detection based on multi-domain feature extraction

Pei ZHAO, Yan QIAO(), Rongyao HU, Xinyu YUAN, Minyue LI, Benchu ZHANG   

  1. Anhui Province Key Laboratory of Industry Safety and Emergency Technology (Hefei University of Technology),Hefei Anhui 230601,China
  • Received:2023-11-27 Revised:2024-03-04 Accepted:2024-03-18 Online:2024-03-22 Published:2024-11-10
  • Contact: Yan QIAO
  • About author:ZHAO Pei, born in 1999, M. S. candidate. His research interests include anomaly detection, information security.
    HU Rongyao, born in 1999, M. S. candidate. His research interests include anomaly detection, information security.
    YUAN Xinyu, born in 1999, M. S. candidate. His research interests include computer networks, pattern recognition.
    LI Minyue, born in 1999, M. S. candidate. Her research interests include natural language processing.
    ZHANG Benchu, born in 2000, M. S. candidate. His research interests include anomaly detection, network security.
  • Supported by:
    Anhui Natural Science Foundation(2008085MF203);Fundamental Research Funds for Central Universities(PA2024GDSK0095)

基于多域特征提取的多变量时间序列异常检测

赵培, 乔焰(), 胡荣耀, 袁新宇, 李敏悦, 张本初   

  1. 工业安全与应急技术安徽省重点实验室(合肥工业大学),合肥 230601
  • 通讯作者: 乔焰
  • 作者简介:赵培(1999—),男,安徽滁州人,硕士研究生,主要研究方向:异常检测、信息安全
    胡荣耀(1999—),男,安徽淮南人,硕士研究生,主要研究方向:异常检测、信息安全
    袁新宇(1999—),男,安徽合肥人,硕士研究生,主要研究方向:计算机网络、模式识别
    李敏悦(1999—),女,安徽合肥人,硕士研究生,主要研究方向:自然语言处理
    张本初(2000—),男,安徽六安人,硕士研究生,主要研究方向:异常检测、网络安全。
  • 基金资助:
    安徽省自然科学基金资助项目(2008085MF203);中央高校基本科研业务费专项资金资助项目(PA2024GDSK0095)

Abstract:

Due to the high dimensionality and the complex variable distribution of Multivariate Time Series (MTS) data, the existing anomaly detection models generally suffer from high error rates and training difficulties when dealing with MTS datasets. Moreover, most models only consider the spatial-temporal features of time series samples, which are not sufficient to learn the features of time series. To solve the above problems, a multivariate Time Series anomaly detection model based on Multi-domain Feature Extraction (MFE-TS) was proposed. Firstly, starting from the original data domain, the Long Short-Term Memory (LSTM) network and the Convolutional Neural Network (CNN) were used to extract the temporal correlation and spatial correlation features of the MTS respectively. Secondly, Fourier transform was used to convert the original time series into frequency domain space, and Transformer was used to learn the amplitude and phase features of the data in frequency domain space. Multi-domain feature learning was able to model time series features more comprehensively, thereby improving anomaly detection performance of the model to MTS. In addition, the masking strategy was introduced to further enhance the feature learning ability of the model and make the model have a certain degree of noise resistance. Experimental results show that MFE-TS has superior performance on multiple real MTS datasets, while it still maintain good detection accuracy on datasets with noise.

Key words: Multivariate Time Series (MTS), anomaly detection, unsupervised learning, multi-domain feature extraction

摘要:

多变量时间序列(MTS)数据具有高维性,且分布复杂多变,现有的异常检测模型在面对MTS数据集时普遍存在误判率高、训练困难等问题,且多数模型仅考虑时间序列样本的时空特征,对时间序列特征的学习并不全面。为了解决以上问题,提出一种基于多域特征提取的MTS异常检测模型(MFE-TS)。首先,从原始数据域出发,使用长短期记忆(LSTM)网络与卷积神经网络(CNN)分别提取MTS的时间相关性和空间相关性特征。其次,用傅里叶变换将原始时间序列转换到频域空间,并利用Transformer学习数据在频域空间的幅度与相位特征。多域特征学习能更全面地建模时间序列特征,从而提高模型对MTS的异常检测性能。此外,引入掩码策略,进一步增强模型的特征学习能力,并使模型具备一定的抗噪性。实验结果表明,MFE-TS在多个真实MTS数据集上展现了优越的性能,同时在含有噪声的数据集中仍能保持较好的检测效果。

关键词: 多变量时间序列, 异常检测, 无监督学习, 多域特征提取

CLC Number: