Blind Source Separation

  • Fengyu CongEmail author


Blind source separation algorithms have been widely used in the EEG signal processing. This chapter introduces the EEG model basis of blind source separation and details of three mainstream algorithms, i.e., principal component analysis (PCA), independent component analysis (ICA), and tensor decomposition, to provide a comprehensive review on this growing topic. The main focus will be on basic principles of applying ICA on continuous EEG data to remove artifacts, PCA, and tensor decomposition on ERP data to conduct group analysis. The introduction of current softwares specialized in PCA and ICA on EEG signal processing will also be covered.


Blind source separation Independent component analysis EEG Event-related potentials Back projection 

Supplementary material (66.9 mb)
Code (ZIP 68531 kb)


  1. Cattell RB. The procrustes program: producing direct rotation to test a hypothesized factor structure. Comput Behav Sci. 1976;7:258–62.Google Scholar
  2. Chaumette E, Comon P, Muller D. ICA-based technique for radiating sources estimation: application to airport surveillance. IEEE Proc Radar Sonar Navig. 1994;142(4):211.. 92363Google Scholar
  3. Cichocki A. Tensor decompositions: a new concept in brain data analysis? arXiv Preprint arXiv: 1305.0395. 2013;50(2011):507–17.Google Scholar
  4. Cichocki A, Zdunek R, Amari S. Nonnegative matrix and tensor factorization. IEEE Signal Process Mag. 2009;25(1):142–5.CrossRefGoogle Scholar
  5. Cois J. Blind signal separation: statistical principles. Proc IEEE. 2009;86(10):2009–25.Google Scholar
  6. Cong F, Kalyakin I, Chang Z, Ristaniemi T. Analysis on subtracting projection of extracted independent components from EEG recordings. Biomed Tech (Berl). 2011a;56(4):223–34.CrossRefGoogle Scholar
  7. Cong F, Kalyakin I, Ristaniemi T. Can back-projection fully resolve polarity indeterminacy of independent component analysis in study of event-related potential? Biomed Signal Process Control. 2011b;6(4):422–6.CrossRefGoogle Scholar
  8. Cong F, Phan AH, Zhao Q, Huttunen-Scott T, Kaartinen J, Ristaniemi T, et al. Benefits of multi-domain feature of mismatch negativity extracted by non-negative tensor factorization from Eeg collected by low-density array. Int J Neural Syst. 2012;22(6):1250025.CrossRefGoogle Scholar
  9. Cong F, He Z, Hämäläinen J, Leppänen PHT, Lyytinen H, Cichocki A, Ristaniemi T. Validating rationale of group-level component analysis based on estimating number of sources in EEG through model order selection. J Neurosci Methods. 2013;212(1):165–72.CrossRefGoogle Scholar
  10. Cong F, Puoliväli T, Alluri V, Sipola T, Burunat I, Toiviainen P, et al. Key issues in decomposing fMRI during naturalistic and continuous music experience with independent component analysis. J Neurosci Methods. 2014;223:74–84.CrossRefGoogle Scholar
  11. Cong F, Lin Q-H, Kuang L-D, Gong X-F, Astikainen P, Ristaniemi T. Tensor decomposition of EEG signals: a brief review. J Neurosci Methods. 2015a;248:59–69.CrossRefGoogle Scholar
  12. Cong F, Ristaniemi T, Lyytinen H. Advanced signal processing on event related potentials (ERPs): World Scientific; 2015b. p. 131–87.Google Scholar
  13. De Lathauwer L, De Moor B, Vandewalle J. Fetal electrocardiogram extraction by blind source. IEEE Trans Biomed Eng. 2000;47(5):567–72.CrossRefGoogle Scholar
  14. Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods. 2004;134(1):9–21.CrossRefGoogle Scholar
  15. Dien J. Addressing misallocation of variance in principal components analysis of event-related. Potentials. 1998;11(1):43–55.Google Scholar
  16. Dien J. Differential lateralization of trait anxiety and trait fearfulness: evoked potential correlates. Personal Individ Differ. 1999;26:1998–2000.Google Scholar
  17. Dien J, Beal DJ, Berg P. Optimizing principal components analysis of event-related potentials: matrix type, factor loading weighting, extraction, and rotations. Clin Neurophysiol. 2005;116:1808–25.CrossRefGoogle Scholar
  18. Durso G, Prieur P. Blind identification methods applied to Electricity C France’s civil works and power plants monitoring. The 1997 IEEE Signal Processing Workshop on Higher-Order Statistics; 1997. p. 82–86.Google Scholar
  19. Eichele T, Rachakonda S, Brakedal B, Eikeland R, Calhoun VD. EEGIFT: group independent component analysis for event-related EEG data. Comput Intell Neurosci. 2011;2011:1–9.CrossRefGoogle Scholar
  20. Harman HH. Modern factor analysis. Chicago: University of Chicago Press; 1976.Google Scholar
  21. Himberg J, Hyvärinen A, Esposito F. Validating the independent components of neuroimaging time series via clustering and visualization. NeuroImage. 2004;22(3):1214–22.CrossRefGoogle Scholar
  22. Hyvarinen A. Fast and robust fixed-point algorithm for independent component analysis. IEEE Trans Neural Netw Learn Syst. 1999;10:626–34.CrossRefGoogle Scholar
  23. Hyvärinen A, Karhunenen J, Oja E. Independent component analysis. Neural Comput. 2001;13(7):504.CrossRefGoogle Scholar
  24. Kaiser H. The Varimax criterion for analytic rotation in factor analysis. Psychometrika. 1958;23(3):187–200.CrossRefGoogle Scholar
  25. Kolda TG, Bader BW. Tensor decompositions and applications. SIAM Rev. 2009;51(3):455–500.CrossRefGoogle Scholar
  26. Luck SJ. An introduction to the event-related potential technique. Monogr Soc Res Child Dev. 2014;78(3):388.Google Scholar
  27. Makeig S, Bell AJ, Jung T-P, Sejnowski TJ. Independent component analysis of electroencephalographic data. In: Advances in neural information processing systems; 1996. p. 145–51.Google Scholar
  28. Mathew G, Reddy VU. Blind separation of multiple co-channel bpsk signals arriving at an antenna array. IEEE Signal Process Lett. 1995;2(9):176–8.CrossRefGoogle Scholar
  29. Mocks J. Topographic components model for event-related potentials and some biophysical considerations. IEEE Trans Biomed Eng. 1988;35:482.CrossRefGoogle Scholar
  30. Mode D, Discovery N. Multi-subject independent component analysis of fMRI: a decade of intrinsic networks, default mode, and neurodiagnostic discovery. IEEE Rev Biomed Eng. 2012;5:60–73.CrossRefGoogle Scholar
  31. Sejnowski TJ, Bell AJ. Information-maximization approach to blind separation and blind deconvolution. Technology. 1995;1159(1994):1129–59.Google Scholar
  32. Spencer KM, Dien J, Donchin E. Spatiotemporal analysis of the late ERP responses to deviant stimuli. Psychophysiology. 2001;38(2):343–58.CrossRefGoogle Scholar
  33. Squires NK, Squires KC, Hillyard SA. Two varieties of long – latency positive waves evoke d by unpredictable auditory stimuli in man. Electroencephalogr Clin Neurophysiol. 1975;38(4):387–401.CrossRefGoogle Scholar
  34. Swindlehurst AL, Goris MJ, Ottersten B. Some experiments with array data collected in actual urban and suburban environments. In: First IEEE signal processing workshop on signal processing advances in wireless communications: IEEE; 1997. p. 301–4.Google Scholar
  35. van der Veen AJ. Algebraic methods for deterministic blind beamforming. Proc IEEE. 1998;86(10):1987–2008.CrossRefGoogle Scholar
  36. Zhang Q, Hu G, Tian L, Ristaniemi T, Wang H, Chen H, Wu J, Cong F. Examining stability of independent component analysis based on coefficient and component matrices for voxel-based morphometry of structural magnetic resonance imaging. Cogn Neurodyn. 2018;12:461–70.CrossRefGoogle Scholar
  37. Zhao Q, Zhou G, Adali T, Zhang L, Cichocki A. Kernelization of tensor-based models for multiway data analysis: processing of multidimensional structured data. IEEE Signal Process Mag. 2013;30(4):137–48.CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.School of Biomedical Engineering, Faculty of Electronic Information and Electrical EngineeringDalian University of TechnologyDalianChina

Personalised recommendations