Advertisement

Wavelet matrix transform for time-series similarity measurement

  • Zhi-kun Hu (胡志坤)Email author
  • Fei Xu (徐 飞)
  • Wei-hua Gui (桂卫华)
  • Chun-hua Yang (阳春华)
Article

Abstract

A time-series similarity measurement method based on wavelet and matrix transform was proposed, and its anti-noise ability, sensitivity and accuracy were discussed. The time-series sequences were compressed into wavelet subspace, and sample feature vector and orthogonal basics of sample time-series sequences were obtained by K-L transform. Then the inner product transform was carried out to project analyzed time-series sequence into orthogonal basics to gain analyzed feature vectors. The similarity was calculated between sample feature vector and analyzed feature vector by the Euclid distance. Taking fault wave of power electronic devices for example, the experimental results show that the proposed method has low dimension of feature vector, the anti-noise ability of proposed method is 30 times as large as that of plain wavelet method, the sensitivity of proposed method is 1/3 as large as that of plain wavelet method, and the accuracy of proposed method is higher than that of the wavelet singular value decomposition method. The proposed method can be applied in similarity matching and indexing for lager time series databases.

Key words

wavelet transform singular value decomposition inner product transform time-series similarity 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    AGRAWALL R, FALOUTSOS C, SWAMI A. Efficient similarity search in sequence databases [C]// Proceeding of the 4th International Conference on Foundations of Data Organization and Algorithms. New York: Springer, 1993: 69–84.CrossRefGoogle Scholar
  2. [2]
    RAFIEI D, MENDELZON A. Efficient retrieval of similar time sequences using DFT [C]// Proceedings of the 15th International Conference on Foundation of Data Organizations and Algorithms. New York: Springer, 1998: 249–257.Google Scholar
  3. [3]
    LI Qiu-dan, CHI Zhong-xian, SUN Rui-chao. A novel similar matching algorithm for time series data [J]. Control and Decision, 2004, 19(8): 915–919. (in Chinese)Google Scholar
  4. [4]
    FRANKY K C, ADA W, CLEMENT Y. Haar wavelets for efficient similarity search of time-series: With and without time warping [J]. IEEE Transaction on Knowledge and Data Engineering, 2003, 15(3): 686–705.CrossRefGoogle Scholar
  5. [5]
    IVAN P, RANEE J M. Similarity search over time-series data using wavelets [C]// Proceedings of the 18th International Conference on Data Engineering. Washington DC: IEEE Computer Society, 2002: 212–221.Google Scholar
  6. [6]
    ZHANG Hai-qin, CAI Qing-sheng. Time series similar pattern matching based on wavelet transform [J]. Chinese Journal of Computers, 2003, 26(3): 373–377. (in Chinese)MathSciNetGoogle Scholar
  7. [7]
    LIU Jian-guo, LI Zhi-shun, LIU Dong. Features of underwater echo extraction based on SWT and SVD [J]. Acta Acustica, 2006, 31(2): 167–172. (in Chinese)Google Scholar
  8. [8]
    KONSSTANTINIDES K, YAO K. Statistical analysis of effective singular values in matrix rank determination [J]. IEEE Transaction on Acoustics, Speech and Signal Process, 1988, 36(5): 757–763.CrossRefGoogle Scholar
  9. [9]
    GROUTAGE D, BENNINKK D. Feature sets for non-stationary signals derived from moments of singular value decomposition of Cohen-Posch distributions [J]. IEEE Transactions on Signal Processing, 2006, 48(5): 1498–1503.CrossRefGoogle Scholar
  10. [10]
    MALLAT S. A theory for multi resolution signal decomposition: The wavelet representation [J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1989, 11(7): 674–693.CrossRefGoogle Scholar
  11. [11]
    HU Han-hui, YANG Hong, TAN Qing, YI Nian-en. Sintering fan faults diagnosis based on wavelet analysis [J]. Journal of Central South University: Science and Technology, 2007, 38(6): 1139–1173. (in Chinese)Google Scholar
  12. [12]
    PENG Xiao-qi, SONG Yan-po, TANG Ying, ZHANG Jian-zhi. Approach based on wavelet analysis for detecting and a mending anomalies in dataset [J]. Journal of Central South University of Technology, 2006, 13(5): 491–495.CrossRefGoogle Scholar
  13. [13]
    AKRITAS A G, MALASCHONOK G I. Application of singular value decomposition [J]. Mathematics and Computers in Simulation, 2004, 67(1): 15–31.MathSciNetCrossRefGoogle Scholar
  14. [14]
    VRABIE V D, MARS J I, LACOUME J L. Modified singular value decomposition by means of independent component analysis [J]. Signal Processing, 2004, 84(3): 645–652.CrossRefGoogle Scholar
  15. [15]
    ZHAO Xue-zhi, CHEN Tong-jian, YE Bang-yan. Processing of milling force signal and isolation of state information of milling machine based on singular value decomposition [J]. Chinese Journal of Mechanical Engineering, 2007, 43(6): 169–174. (in Chinese)CrossRefGoogle Scholar

Copyright information

© Central South University Press and Springer Berlin Heidelberg 2009

Authors and Affiliations

  • Zhi-kun Hu (胡志坤)
    • 1
    Email author
  • Fei Xu (徐 飞)
    • 1
  • Wei-hua Gui (桂卫华)
    • 2
  • Chun-hua Yang (阳春华)
    • 2
  1. 1.School of Physics Science and TechnologyCentral South UniversityChangshaChina
  2. 2.School of Information Science and EngineeringCentral South UniversityChangshaChina

Personalised recommendations