Dissipativity analysis of discrete-time delayed neural networks

Zhiguang Feng; Wei Xing Zheng

Dissipativity analysis of discrete-time delayed neural networks

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

Abstract

The objective of this paper to analyze dissipativity of discrete-time neural networks with time-varying delay. The main idea is to introduce the concept of extended dissipativity for discrete-time neural networks with a view to unifying several performance measures such as the Hâˆž performance, passivity, l2-lâˆž performance and dissipativity. The reciprocally convex approach together with a Lyapunov function involving a triple-summable term is applied to develop the extended dissipativity criterion for discrete-time neural networks with time-varying delay. In addition, the new criterion also ensures the stability of the neural networks. The improved results are validated through a numerical example in comparison with the existing results.

Original language	English
Title of host publication	Proceedings of the 5th Australian Control Conference (AUCC), November 5-6, 2015, Gold Coast, Australia
Publisher	Engineers Australia
Pages	134-137
Number of pages	4
ISBN (Print)	9781467395526
Publication status	Published - 2015
Event	Australian Control Conference - Duration: 5 Nov 2015 → …

Conference

Conference	Australian Control Conference
Period	5/11/15 → …

Keywords

Lyapunov stability
discrete-time systems
neural networks (computer science)

Cite this

@inproceedings{d930af3fbf884f71ba50574f61f74f40,

title = "Dissipativity analysis of discrete-time delayed neural networks",

abstract = "The objective of this paper to analyze dissipativity of discrete-time neural networks with time-varying delay. The main idea is to introduce the concept of extended dissipativity for discrete-time neural networks with a view to unifying several performance measures such as the H{\^a}ˆ{\v z} performance, passivity, l2-l{\^a}ˆ{\v z} performance and dissipativity. The reciprocally convex approach together with a Lyapunov function involving a triple-summable term is applied to develop the extended dissipativity criterion for discrete-time neural networks with time-varying delay. In addition, the new criterion also ensures the stability of the neural networks. The improved results are validated through a numerical example in comparison with the existing results.",

keywords = "Lyapunov stability, discrete-time systems, neural networks (computer science)",

author = "Zhiguang Feng and Zheng, {Wei Xing}",

year = "2015",

language = "English",

isbn = "9781467395526",

pages = "134--137",

booktitle = "Proceedings of the 5th Australian Control Conference (AUCC), November 5-6, 2015, Gold Coast, Australia",

publisher = "Engineers Australia",

note = "Australian Control Conference ; Conference date: 05-11-2015",

}

TY - GEN

T1 - Dissipativity analysis of discrete-time delayed neural networks

AU - Feng, Zhiguang

AU - Zheng, Wei Xing

PY - 2015

Y1 - 2015

N2 - The objective of this paper to analyze dissipativity of discrete-time neural networks with time-varying delay. The main idea is to introduce the concept of extended dissipativity for discrete-time neural networks with a view to unifying several performance measures such as the Hâˆž performance, passivity, l2-lâˆž performance and dissipativity. The reciprocally convex approach together with a Lyapunov function involving a triple-summable term is applied to develop the extended dissipativity criterion for discrete-time neural networks with time-varying delay. In addition, the new criterion also ensures the stability of the neural networks. The improved results are validated through a numerical example in comparison with the existing results.

AB - The objective of this paper to analyze dissipativity of discrete-time neural networks with time-varying delay. The main idea is to introduce the concept of extended dissipativity for discrete-time neural networks with a view to unifying several performance measures such as the Hâˆž performance, passivity, l2-lâˆž performance and dissipativity. The reciprocally convex approach together with a Lyapunov function involving a triple-summable term is applied to develop the extended dissipativity criterion for discrete-time neural networks with time-varying delay. In addition, the new criterion also ensures the stability of the neural networks. The improved results are validated through a numerical example in comparison with the existing results.

KW - Lyapunov stability

KW - discrete-time systems

KW - neural networks (computer science)

UR - http://handle.uws.edu.au:8081/1959.7/uws:33529

UR - http://www.aucc.org.au/AUCC2015/

M3 - Conference Paper

SN - 9781467395526

SP - 134

EP - 137

BT - Proceedings of the 5th Australian Control Conference (AUCC), November 5-6, 2015, Gold Coast, Australia

PB - Engineers Australia

T2 - Australian Control Conference

Y2 - 5 November 2015

ER -

Dissipativity analysis of discrete-time delayed neural networks

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this