• DocumentCode
    2042443
  • Title

    Information theory based estimator of the number of sources in a sparse linear mixing model

  • Author

    Balan, Radu

  • Author_Institution
    Dept. of Math., Maryland Univ., College Park, MD
  • fYear
    2008
  • fDate
    19-21 March 2008
  • Firstpage
    269
  • Lastpage
    273
  • Abstract
    In this paper we present an Information theoretic estimator for the number of sources mutually disjoint in a linear mixing model. The approach follows the Minimum Description Length prescription and is roughly equal to the sum of negative normalized maximum log-likelihood and the logarithm of number of sources. Preliminary numerical evidence supports this approach and compares favorably to both the Akaike (AIC) and Bayesian (BIC) Information Criteria.
  • Keywords
    Bayes methods; blind source separation; maximum likelihood estimation; sparse matrices; Akaike information criteria; Bayesian information criteria; blind source separation; information theoretic estimator; minimum description length prescription; negative normalized maximum log-likelihood estimation; sparse linear mixing model; Estimation theory; Hydrogen; Information theory; Mathematical model; Mathematics; Maximum likelihood estimation; Signal analysis; Statistics; Time frequency analysis; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Sciences and Systems, 2008. CISS 2008. 42nd Annual Conference on
  • Conference_Location
    Princeton, NJ
  • Print_ISBN
    978-1-4244-2246-3
  • Electronic_ISBN
    978-1-4244-2247-0
  • Type

    conf

  • DOI
    10.1109/CISS.2008.4558534
  • Filename
    4558534