Model structure selection in convolutive mixtures

Mads Dyrholm; Scott Makeig; Lars Kai Hansen

Model structure selection in convolutive mixtures

Mads Dyrholm, Scott Makeig, Lars Kai Hansen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

424 Downloads (Pure)

Abstract

The CICAAR algorithm (convolutive independent component analysis with an auto-regressive inverse model) allows separation of white (i.i.d) source signals from convolutive mixtures. We introduce a source color model as a simple extension to the CICAAR which allows for a more parsimoneous representation in many practical mixtures. The new filter-CICAAR allows Bayesian model selection and can help answer questions like: 'Are we actually dealing with a convolutive mixture?'. We try to answer this question for EEG data.

Original language	English
Title of host publication	6th International Conference on Independent Component Analysis and Blind Source Separation
Publication date	2006
Publication status	Published - 2006

Access to Document

Imm3880

http://www2.imm.dtu.dk/pubdb/p.php?3880

OpenUrl availability

Full text

Cite this

@inproceedings{3e72d75d352d40139667991d8a1db0cc,

title = "Model structure selection in convolutive mixtures",

abstract = "The CICAAR algorithm (convolutive independent component analysis with an auto-regressive inverse model) allows separation of white (i.i.d) source signals from convolutive mixtures. We introduce a source color model as a simple extension to the CICAAR which allows for a more parsimoneous representation in many practical mixtures. The new filter-CICAAR allows Bayesian model selection and can help answer questions like: 'Are we actually dealing with a convolutive mixture?'. We try to answer this question for EEG data.",

keywords = "ica convolutive maximum likelihood contextual eeg model selection",

author = "Mads Dyrholm and Scott Makeig and Hansen, {Lars Kai}",

year = "2006",

language = "English",

booktitle = "6th International Conference on Independent Component Analysis and Blind Source Separation",

}

TY - GEN

T1 - Model structure selection in convolutive mixtures

AU - Dyrholm, Mads

AU - Makeig, Scott

AU - Hansen, Lars Kai

PY - 2006

Y1 - 2006

N2 - The CICAAR algorithm (convolutive independent component analysis with an auto-regressive inverse model) allows separation of white (i.i.d) source signals from convolutive mixtures. We introduce a source color model as a simple extension to the CICAAR which allows for a more parsimoneous representation in many practical mixtures. The new filter-CICAAR allows Bayesian model selection and can help answer questions like: 'Are we actually dealing with a convolutive mixture?'. We try to answer this question for EEG data.

AB - The CICAAR algorithm (convolutive independent component analysis with an auto-regressive inverse model) allows separation of white (i.i.d) source signals from convolutive mixtures. We introduce a source color model as a simple extension to the CICAAR which allows for a more parsimoneous representation in many practical mixtures. The new filter-CICAAR allows Bayesian model selection and can help answer questions like: 'Are we actually dealing with a convolutive mixture?'. We try to answer this question for EEG data.

KW - ica convolutive maximum likelihood contextual eeg model selection

M3 - Article in proceedings

BT - 6th International Conference on Independent Component Analysis and Blind Source Separation

ER -