Gradient Algorithms for Complex Non-Gaussian Independent Component/Vector Extraction, Question of Convergence

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F19%3A00006125" target="_blank" >RIV/46747885:24220/19:00006125 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/67985556:_____/19:00500102

Výsledek na webu

<a href="https://asap.ite.tul.cz/wp-content/uploads/sites/3/2018/12/ICEvXII.pdf" target="_blank" >https://asap.ite.tul.cz/wp-content/uploads/sites/3/2018/12/ICEvXII.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TSP.2018.2887185" target="_blank" >10.1109/TSP.2018.2887185</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Gradient Algorithms for Complex Non-Gaussian Independent Component/Vector Extraction, Question of Convergence

Popis výsledku v původním jazyce

We revise the problem of extracting one inde-pendent component from an instantaneous linear mixture of signals. The mixing matrix is parameterized by two vectors: one column of the mixing matrix, and one row of the de-mixing matrix. The separation is based on the non-Gaussianity of the source of interest, while the remaining background signalsare assumed to be Gaussian. Three gradient-based estimation algorithms are derived using the maximum likelihood principleand are compared with the Natural Gradient algorithm for Independent Component Analysis and with One-Unit FastICA based on negentropy maximization. The ideas and algorithms are also generalized to the extraction of a vector component when the extraction proceeds jointly from a set of instantaneous mixtures. Throughout this paper, we address the problem concerning the size of the region of convergence for which the algorithms guarantee the extraction of the desired source. We show that the size is influenced by the Signal-to-Interference Ratio onthe input channels. Simulations comparing several algorithms confirm this observation. They show a different size of the region of convergence under a scenario in which the source of interestis dominant or weak. Here, our proposed modifications of the gradient methods, taking into account the dominance/weakness of the source, show improved global convergence.

Název v anglickém jazyce

Gradient Algorithms for Complex Non-Gaussian Independent Component/Vector Extraction, Question of Convergence

Popis výsledku anglicky

We revise the problem of extracting one inde-pendent component from an instantaneous linear mixture of signals. The mixing matrix is parameterized by two vectors: one column of the mixing matrix, and one row of the de-mixing matrix. The separation is based on the non-Gaussianity of the source of interest, while the remaining background signalsare assumed to be Gaussian. Three gradient-based estimation algorithms are derived using the maximum likelihood principleand are compared with the Natural Gradient algorithm for Independent Component Analysis and with One-Unit FastICA based on negentropy maximization. The ideas and algorithms are also generalized to the extraction of a vector component when the extraction proceeds jointly from a set of instantaneous mixtures. Throughout this paper, we address the problem concerning the size of the region of convergence for which the algorithms guarantee the extraction of the desired source. We show that the size is influenced by the Signal-to-Interference Ratio onthe input channels. Simulations comparing several algorithms confirm this observation. They show a different size of the region of convergence under a scenario in which the source of interestis dominant or weak. Here, our proposed modifications of the gradient methods, taking into account the dominance/weakness of the source, show improved global convergence.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10102 - Applied mathematics

Projekt

<a href="/cs/project/GA17-00902S" target="_blank" >GA17-00902S: Pokročilé metody slepé separace podprostorů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

IEEE Transactions on Signal Processing

ISSN

1053-587X

e-ISSN

—

Svazek periodika

67

Číslo periodika v rámci svazku

4

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

15

Strana od-do

1050-1064

Kód UT WoS článku

000455720600015

EID výsledku v databázi Scopus

2-s2.0-85058892691