Complex Empirical Mode Decomposition for Multichannel Information Fusion

  • Danilo Mandic
  • George Souretis
  • Wai Yie Leong
  • David Looney
  • Marc M. Van Hulle
  • Toshihisa Tanaka

Information “fusion” via signal “fission” is addressed in the framework of empirical mode decomposition (EMD). In this way, a general nonlinear and non-stationary signal is first decomposed into its oscillatory components (fission); the components of interest are then combined in an ad hoc or automated fashion to provide greater knowledge about a process in hand (fusion). The extension to the field of complex numbers C is particularly important for the analysis of phase-dependent processes, such as those coming from sensor arrays. This allows us to combine the data driven nature of EMD with the power of complex algebra to model amplitude-phase relationships within multichannel data. The analysis shows that the extensions of EMD to C are not straightforward and that they critically depend on the criterion for finding local extrema within a complex signal. For rigour, convergence of EMD is addressed within the framework of fixed point theory. Simulation examples on information fusion for brain computer interface (BCI) support the analysis.


Empirical Mode Decomposition Instantaneous Frequency Information Fusion Intrinsic Mode Function Brain Computer Interface 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Altaf, M.U.B., Gautama, T., Tanaka, T., Mandic, D.P.: Rotation invariant complex empirical mode decomposition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 3, pp. 1009–1012 (2007)Google Scholar
  2. 2.
    Cohen, L.: Instantaneous anything. In: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 5, pp. 105–108 (1993)Google Scholar
  3. 3.
    Delechelle, E., Lemoine, J., Niang, O.: Empirical mode decomposition: An analytical approach for sifting process. IEEE Signal Processing Letters 12, 764–767 (2005)CrossRefGoogle Scholar
  4. 4.
    Goh, S.L., Chen, M., Popovic, D.H., Aihara, K., Obradovic, D., Mandic, D.P.: Complex-valued forecasting of wind profile. Renewable Energy 31(1733–1750), tba (2006)Google Scholar
  5. 5.
    Hall, D.L., Llinas, J.: An introduction to multisensor data fusion. Proceedings of the IEEE 85(1), 6–23 (1997)CrossRefGoogle Scholar
  6. 6.
    Hu, J., Yang, Y.D., Kihara, D.: EMD: an ensemble algorithm for discovering regulatory motifs in DNA sequences. BMC Bioinformatics 7, 342 (2006)CrossRefGoogle Scholar
  7. 7.
    Huang, N.E., Shen, Z., Long, S.R., Wu, M.L., Shih, H.H., Quanan, Z., Yen, N.C., Tung, C.C., Liu, H.H.: The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society A 454, 903–995 (1998)MATHCrossRefGoogle Scholar
  8. 8.
    Leibovic, K.N.: Contraction mapping with application to control processes. Journal of Electronics and Control pp. 81–95 (1963)Google Scholar
  9. 9.
    Looney, D., Li, L., Rutkowski, T.M., Mandic, D.P., Cichocki, A.: Ocular artifacts removal from eeg using emd. In: Proceedings of the 1st International Conference on Cognitive Neurodynamics (ICCN) (2007)Google Scholar
  10. 10.
    Mandic, D.P., Chambers, J.A.: Recurrent Neural Networks for Prediction: Architectures, Learning Algorithms and Stability. Wiley (2001)Google Scholar
  11. 11.
    Mandic, D.P., Goh, S.L., Aihara, K.: Sequential data fusion via vector spaces: Fusion of heterogeneous data in the complex domain. International Journal of VLSI Signal Processing Systems 48(1–2), 98–108 (2007)Google Scholar
  12. 12.
    Mandic, D.P., Obradovic, D., Kuh, A., Adali, T., Trutschell, U., Golz, M., Wilde, P.D., Barria, J., Constantinides, A., Chambers, J.: Data fusion for modern engineering applications: An overview. In: Proceedings of the IEEE International Conference on Artificial Neural Networks (ICANN’05), pp. 715–721 (2005)Google Scholar
  13. 13.
    Smith, L.S., Mtetwa, N.: A tool for synthesizing spike trains with realistic interference. Journal of Neuroscience Methods 159(1), 170–180 (2006)CrossRefGoogle Scholar
  14. 14.
    Souretis, G., Mandic, D.P., Griselli, M., Tanaka, T., Hulle, M.M.V.: Blood volume signal analysis with empirical mode decomposition. In: Proceedings of the 15th International DSP Conference, pp. 147–150 (2007)Google Scholar
  15. 15.
    Tanaka, T., Mandic, D.P.: Complex empirical mode decomposition. IEEE Signal Processing Letters 14(2), 101–104 (Feb 2007)CrossRefGoogle Scholar
  16. 16.
    Wald, L.: Some terms of reference in data fusion. IEEE Transactions on Geosciences and Remote Sensing 37(3), 1190–1193 (1999)CrossRefGoogle Scholar
  17. 17.
    Waltz, E., Llinas, J.: Multisensor Data Fusion. Artech House (1990)Google Scholar
  18. 18.
    Weng, B., Barner, K.E.: Optimal and bidirectional optimal empirical mode decomposition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 3, pp. 1501–1504 (2007)Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  • Danilo Mandic
    • 1
  • George Souretis
    • 1
  • Wai Yie Leong
    • 2
  • David Looney
    • 1
  • Marc M. Van Hulle
    • 3
  • Toshihisa Tanaka
    • 4
  1. 1.Imperial College LondonLondonUK
  2. 2.Agency for Science, Technology and ResearchSingapore
  3. 3.Katholieke Universiteit LeuvenBelgium
  4. 4.Tokyo University of Agriculture and TechnologyJapan

Personalised recommendations