Nonlinear signal separation for multinonlinearity constrained mixing model

Pei Gao; W. L. Woo; S. S. Dlay

doi:10.1109/TNN.2006.873288

Nonlinear signal separation for multinonlinearity constrained mixing model

Pei Gao^*, W. L. Woo, S. S. Dlay

^*Corresponding author for this work

Computer and Information Sciences Department

Research output: Contribution to journal › Article › peer-review

20 Citations (Scopus)

Abstract

In this letter, a new type of nonlinear mixture is derived and developed into a multinonlinearity constrained mixing model. The proposed signal separation solution integrates the Theory of Series Reversion with a polynomial neural network whereby the hidden neurons are spanned by a set of mutually reversed activation functions. Simulations have been undertaken to support the theory of the proposed scheme and the results indicate promising performance.

Original language	English
Pages (from-to)	796-802
Number of pages	7
Journal	IEEE Transactions on Neural Networks
Volume	17
Issue number	3
DOIs	https://doi.org/10.1109/TNN.2006.873288
Publication status	Published - 8 May 2006

Keywords

Adaptive signal estimation
Blind separation
Independent component analysis (ICA)
Series reversion

Access to Document

10.1109/TNN.2006.873288

Cite this

@article{87f8e2b7b7714d0080dbc945779db693,

title = "Nonlinear signal separation for multinonlinearity constrained mixing model",

abstract = "In this letter, a new type of nonlinear mixture is derived and developed into a multinonlinearity constrained mixing model. The proposed signal separation solution integrates the Theory of Series Reversion with a polynomial neural network whereby the hidden neurons are spanned by a set of mutually reversed activation functions. Simulations have been undertaken to support the theory of the proposed scheme and the results indicate promising performance.",

keywords = "Adaptive signal estimation, Blind separation, Independent component analysis (ICA), Series reversion",

author = "Pei Gao and Woo, {W. L.} and Dlay, {S. S.}",

year = "2006",

month = may,

day = "8",

doi = "10.1109/TNN.2006.873288",

language = "English",

volume = "17",

pages = "796--802",

journal = "IEEE Transactions on Neural Networks",

issn = "1045-9227",

publisher = "IEEE",

number = "3",

}

TY - JOUR

T1 - Nonlinear signal separation for multinonlinearity constrained mixing model

AU - Gao, Pei

AU - Woo, W. L.

AU - Dlay, S. S.

PY - 2006/5/8

Y1 - 2006/5/8

N2 - In this letter, a new type of nonlinear mixture is derived and developed into a multinonlinearity constrained mixing model. The proposed signal separation solution integrates the Theory of Series Reversion with a polynomial neural network whereby the hidden neurons are spanned by a set of mutually reversed activation functions. Simulations have been undertaken to support the theory of the proposed scheme and the results indicate promising performance.

AB - In this letter, a new type of nonlinear mixture is derived and developed into a multinonlinearity constrained mixing model. The proposed signal separation solution integrates the Theory of Series Reversion with a polynomial neural network whereby the hidden neurons are spanned by a set of mutually reversed activation functions. Simulations have been undertaken to support the theory of the proposed scheme and the results indicate promising performance.

KW - Adaptive signal estimation

KW - Blind separation

KW - Independent component analysis (ICA)

KW - Series reversion

U2 - 10.1109/TNN.2006.873288

DO - 10.1109/TNN.2006.873288

M3 - Article

C2 - 16722182

AN - SCOPUS:33646517887

SN - 1045-9227

VL - 17

SP - 796

EP - 802

JO - IEEE Transactions on Neural Networks

JF - IEEE Transactions on Neural Networks

IS - 3

ER -

Nonlinear signal separation for multinonlinearity constrained mixing model

Abstract

Keywords

Access to Document

Fingerprint

Cite this