Delayed AMUSE – A Tool for Blind Source Separation and Denoising

  • Ana R. Teixeira
  • Ana Maria Tomé
  • Elmar W. Lang
  • Kurt Stadlthanner
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3195)


In this work we propose a generalized eigendecomposition (GEVD) of a matrix pencil computed after embedding the data into a high-dim feature space of delayed coordinates. The matrix pencil is computed like in AMUSE but in the feature space of delayed coordinates. Its GEVD yields filtered versions of the source signals as output signals. The algorithm is implemented in two EVD steps. Numerical simulations study the influence of the number of delays and the noise level on the performance.


Heart Rate Variability Source Signal Sensor Signal Blind Source Separation Singular Spectrum Analysis 
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.
    Cichocki, A., Amari, S.-I.: Adaptive Blind Signal and Image Processing. Wiley, Chichester (2002)CrossRefGoogle Scholar
  2. 2.
    Gharieb, R., Cichocki, A.: Second-order statistics based blind source separation using a bank of subband filters. Digital Signal Processing 13, 252–274 (2003)CrossRefGoogle Scholar
  3. 3.
    Schreiber, T., Kantz, H.: Nonlinear projective filtering II: Application to real time series. In: Proceedings of Nolta (1998)Google Scholar
  4. 4.
    Vetter, R., Vesin, J., Celka, P., Renevey, P., Krauss, J.: Automatic nonlinear noise reduction using local principal component analysis and MDL parameter selection. In: IASTED Internation Conference on Signal Processing, Pattern Recognition& Applications, Crete, Greece, pp. 290–294 (2002)Google Scholar
  5. 5.
    Gruber, P., Theis, F.J., Tomé, A.M., Lang, E.W.: Automatic denoising using local independent component analysis. In: Proceed. EIS 2004, Madeira, Portugal (2004)Google Scholar
  6. 6.
    Moskvina, V., Schmidt, K.M.: Approximate projectors in singular spectrum analysis. SIAM Journal Mat. Anal. Appl. 24(4), 932–942 (2003)zbMATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Ghil, M., Allen, M., Dettinger, M.D., Ide, K., et al.: Advanced spectral methods for climatic time series. Reviews of Geophysics 40(1), 3.1–3.41 (2002)Google Scholar
  8. 8.
    Tomé, A.M., Ferreira, N.: On-line source separation of temporally correlated signals. In: European Signal Processing Conference, EUSIPCO 2002, Toulouse, France (2002)Google Scholar
  9. 9.
    Tomé, A.M.: Separation of a mixture of signals using linear filtering and second order statistics. In: 10th European Symposium on Artificial Neural Networks, Brugges, pp. 307–312 (2002)Google Scholar
  10. 10.
    Stone, J.V.: Blind source separation using temporal predictibility. Neural Computation 13(7), 1557–1574 (2001)CrossRefGoogle Scholar
  11. 11.
    Teixeira, A.R., Rocha, A.P., Almeida, R., Tomé, A.M.: The analysis of heart rate variability using independent component signals. In: Second International Conference on Biomedical Engineering, Innsbruck, Austria. IASTED, pp. 240–243 (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Ana R. Teixeira
    • 1
  • Ana Maria Tomé
    • 1
  • Elmar W. Lang
    • 2
  • Kurt Stadlthanner
    • 2
  1. 1.Departamento de Electrónica e Telecomunicações/IEETAUniversidade de AveiroAveiroPortugal
  2. 2.Institute of Biophysics, Neuro- and Bioinformatics GroupUniversity of RegensburgRegensburgGermany

Personalised recommendations