Monaural ICA of white noise mixtures is hard

Lars Kai Hansen, Kaare Brandt Petersen

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    102 Downloads (Pure)

    Abstract

    Separation of monaural linear mixtures of `white' source signals is fundamentally ill-posed. In some situations it is not possible to find the mixing coefficients for the full `blind' problem. If the mixing coefficients are known, the structure of the source prior distribution determines the source reconstruction error. If the prior is strongly multi-modal source reconstruction is possible with low error, while source signals from the typical `long tailed' distributions used in many ICA settings can not be reconstructed. We provide a qualitative discussion of the limits of monaural blind separation of white noise signals and give a set of `no go' cases.
    Original languageEnglish
    Title of host publicationProceedings of ICA'2003 Fourth Int. Symp on Independent Component Analysis and Blind Signal Separation, Nara Japan, April 4,
    Publication date2003
    Pages815-820
    Publication statusPublished - 2003
    EventICA'2003 Fourth Int. Symp on Independent Component Analysis and Blind Signal Separation : April 4 - Nara Japan
    Duration: 1 Jan 2003 → …

    Conference

    ConferenceICA'2003 Fourth Int. Symp on Independent Component Analysis and Blind Signal Separation : April 4
    CityNara Japan
    Period01/01/2003 → …

    Fingerprint Dive into the research topics of 'Monaural ICA of white noise mixtures is hard'. Together they form a unique fingerprint.

    Cite this