• DocumentCode
    3096267
  • Title

    The general source separation problem

  • Author

    Lacoume, J. ; Gaeta, M.

  • fYear
    1990
  • fDate
    10-12 Oct. 1990
  • Firstpage
    154
  • Lastpage
    158
  • Abstract
    Presents a maximum likelihood approach of the separation problem of independent sources linearly combined. This separation cannot be successful using only the spectral matrix. But it can be done using higher order statistics and in particular the fourth order cumulant of the signals. The general linear case can be solved in frequency domain. In this domain the authors introduce the notion of circular complex variables, and give the maximum likelihood estimator of the source model in the complex data case.<>
  • Keywords
    signal processing; statistical analysis; circular complex variables; complex data; fourth order cumulant; frequency domain; general source separation; higher order statistics; independent sources; maximum likelihood approach; maximum likelihood estimator; signal vector; source model; spectral matrix; Additive noise; Bioreactors; Gaussian noise; Higher order statistics; Matrix decomposition; Maximum likelihood detection; Maximum likelihood estimation; Nonlinear filters; Source separation; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Spectrum Estimation and Modeling, 1990., Fifth ASSP Workshop on
  • Conference_Location
    Rochester, NY, USA
  • Type

    conf

  • DOI
    10.1109/SPECT.1990.205565
  • Filename
    205565