Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal

Seng Kah Phooi; L. M. Ang

doi:10.1109/ICCIAS.2006.294170

Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal

Seng Kah Phooi, L. M. Ang

University of Nottingham

Research output: Book chapter/Published conference paper › Conference paper › peer-review

16 Citations (Scopus)

Abstract

This paper presents an improved adaptive radial basis function neural network (RBF NN) for nonlinear and nonstationary signal. The proposed method possesses distinctive properties of Lyapunov Theorybased Adaptive Filtering (LAF) in [1]-[2]. This method is different from many RBF NN training methods using gradient search techniques. A new Lyapunov function of the error between the desired output and the RBF NN output is first defined. The output asymptotically converges to the desired output by proper design of the weight adaptation law in Lyapunov sense. In this paper, we have proved that the design is independent of statistic properties of the input and output signals. The proposed method has better tracking capability compared with the LAF in [1]-[2]. The performance of the proposed technique is illustrated through the nonlinear adaptive prediction of nonstationary speech signals.

Original language	English
Title of host publication	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006
Pages	433-436
Number of pages	4
Volume	1
DOIs	https://doi.org/10.1109/ICCIAS.2006.294170
Publication status	Published - 01 Dec 2007
Event	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006 - Guangzhou, China Duration: 03 Oct 2006 → 06 Oct 2006

Conference

Conference	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006
Country/Territory	China
City	Guangzhou
Period	03/10/06 → 06/10/06

Access to Document

10.1109/ICCIAS.2006.294170

Cite this

@inproceedings{7d4e3c7ad004413ea9d04c1c2906e198,

title = "Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal",

abstract = "This paper presents an improved adaptive radial basis function neural network (RBF NN) for nonlinear and nonstationary signal. The proposed method possesses distinctive properties of Lyapunov Theorybased Adaptive Filtering (LAF) in [1]-[2]. This method is different from many RBF NN training methods using gradient search techniques. A new Lyapunov function of the error between the desired output and the RBF NN output is first defined. The output asymptotically converges to the desired output by proper design of the weight adaptation law in Lyapunov sense. In this paper, we have proved that the design is independent of statistic properties of the input and output signals. The proposed method has better tracking capability compared with the LAF in [1]-[2]. The performance of the proposed technique is illustrated through the nonlinear adaptive prediction of nonstationary speech signals.",

author = "Phooi, {Seng Kah} and Ang, {L. M.}",

year = "2007",

month = dec,

day = "1",

doi = "10.1109/ICCIAS.2006.294170",

language = "English",

isbn = "1424406056",

volume = "1",

pages = "433--436",

booktitle = "2006 International Conference on Computational Intelligence and Security, ICCIAS 2006",

note = "2006 International Conference on Computational Intelligence and Security, ICCIAS 2006 ; Conference date: 03-10-2006 Through 06-10-2006",

}

Phooi, SK & Ang, LM 2007, Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal. in 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006. vol. 1, 4072123, pp. 433-436, 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006, Guangzhou, China, 03/10/06. https://doi.org/10.1109/ICCIAS.2006.294170

TY - GEN

T1 - Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal

AU - Phooi, Seng Kah

AU - Ang, L. M.

PY - 2007/12/1

Y1 - 2007/12/1

N2 - This paper presents an improved adaptive radial basis function neural network (RBF NN) for nonlinear and nonstationary signal. The proposed method possesses distinctive properties of Lyapunov Theorybased Adaptive Filtering (LAF) in [1]-[2]. This method is different from many RBF NN training methods using gradient search techniques. A new Lyapunov function of the error between the desired output and the RBF NN output is first defined. The output asymptotically converges to the desired output by proper design of the weight adaptation law in Lyapunov sense. In this paper, we have proved that the design is independent of statistic properties of the input and output signals. The proposed method has better tracking capability compared with the LAF in [1]-[2]. The performance of the proposed technique is illustrated through the nonlinear adaptive prediction of nonstationary speech signals.

AB - This paper presents an improved adaptive radial basis function neural network (RBF NN) for nonlinear and nonstationary signal. The proposed method possesses distinctive properties of Lyapunov Theorybased Adaptive Filtering (LAF) in [1]-[2]. This method is different from many RBF NN training methods using gradient search techniques. A new Lyapunov function of the error between the desired output and the RBF NN output is first defined. The output asymptotically converges to the desired output by proper design of the weight adaptation law in Lyapunov sense. In this paper, we have proved that the design is independent of statistic properties of the input and output signals. The proposed method has better tracking capability compared with the LAF in [1]-[2]. The performance of the proposed technique is illustrated through the nonlinear adaptive prediction of nonstationary speech signals.

UR - http://www.scopus.com/inward/record.url?scp=38549148658&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38549148658&partnerID=8YFLogxK

U2 - 10.1109/ICCIAS.2006.294170

DO - 10.1109/ICCIAS.2006.294170

M3 - Conference paper

SN - 1424406056

SN - 9781424406050

VL - 1

SP - 433

EP - 436

BT - 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006

T2 - 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006

Y2 - 3 October 2006 through 6 October 2006

ER -

Adaptive RBF neural network training algorithm for nonlinear and nonstationary signal

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this