Passive and exponential filter design for fuzzy neural networks

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

This paper proposes a new passive and exponential filter for Takagi-Sugeno fuzzy Hopfield neural networks, with time delay and external disturbance. Based on the Lyapunov-Krasovskii stability theory, Jensen's inequality, and linear matrix inequality (LMI), a new delay-dependent criterion is proposed such that the filtering error system becomes exponentially stable and passive from the external disturbance to the output error. The proposed filter can be obtained by solving the LMI, which can be easily facilitated using standard numerical packages. Two numerical examples are given to illustrate the effectiveness of the proposed filter.

Original languageEnglish
Pages (from-to)126-137
Number of pages12
JournalInformation Sciences
Volume238
DOIs
Publication statusPublished - 2013 Jul 20

Fingerprint

Fuzzy neural networks
Filter Design
Fuzzy Neural Network
Linear matrix inequalities
Filter
Hopfield neural networks
Matrix Inequality
Linear Inequalities
Disturbance
Delay-dependent Criteria
Time delay
Jensen's inequality
Hopfield Neural Network
Stability Theory
Lyapunov
Time Delay
Filtering
Numerical Examples
Output
Fuzzy neural network

Keywords

  • Exponential filter
  • Hopfield neural networks
  • Linear matrix inequality (LMI)
  • Lyapunov-Krasovskii stability theory
  • Passive filter
  • Takagi-Sugeno fuzzy

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science Applications
  • Information Systems and Management

Cite this

Passive and exponential filter design for fuzzy neural networks. / Ahn, Choon Ki.

In: Information Sciences, Vol. 238, 20.07.2013, p. 126-137.

Research output: Contribution to journalArticle

@article{211ca8bb09c643a59e49f3c517313776,
title = "Passive and exponential filter design for fuzzy neural networks",
abstract = "This paper proposes a new passive and exponential filter for Takagi-Sugeno fuzzy Hopfield neural networks, with time delay and external disturbance. Based on the Lyapunov-Krasovskii stability theory, Jensen's inequality, and linear matrix inequality (LMI), a new delay-dependent criterion is proposed such that the filtering error system becomes exponentially stable and passive from the external disturbance to the output error. The proposed filter can be obtained by solving the LMI, which can be easily facilitated using standard numerical packages. Two numerical examples are given to illustrate the effectiveness of the proposed filter.",
keywords = "Exponential filter, Hopfield neural networks, Linear matrix inequality (LMI), Lyapunov-Krasovskii stability theory, Passive filter, Takagi-Sugeno fuzzy",
author = "Ahn, {Choon Ki}",
year = "2013",
month = "7",
day = "20",
doi = "10.1016/j.ins.2013.03.004",
language = "English",
volume = "238",
pages = "126--137",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Passive and exponential filter design for fuzzy neural networks

AU - Ahn, Choon Ki

PY - 2013/7/20

Y1 - 2013/7/20

N2 - This paper proposes a new passive and exponential filter for Takagi-Sugeno fuzzy Hopfield neural networks, with time delay and external disturbance. Based on the Lyapunov-Krasovskii stability theory, Jensen's inequality, and linear matrix inequality (LMI), a new delay-dependent criterion is proposed such that the filtering error system becomes exponentially stable and passive from the external disturbance to the output error. The proposed filter can be obtained by solving the LMI, which can be easily facilitated using standard numerical packages. Two numerical examples are given to illustrate the effectiveness of the proposed filter.

AB - This paper proposes a new passive and exponential filter for Takagi-Sugeno fuzzy Hopfield neural networks, with time delay and external disturbance. Based on the Lyapunov-Krasovskii stability theory, Jensen's inequality, and linear matrix inequality (LMI), a new delay-dependent criterion is proposed such that the filtering error system becomes exponentially stable and passive from the external disturbance to the output error. The proposed filter can be obtained by solving the LMI, which can be easily facilitated using standard numerical packages. Two numerical examples are given to illustrate the effectiveness of the proposed filter.

KW - Exponential filter

KW - Hopfield neural networks

KW - Linear matrix inequality (LMI)

KW - Lyapunov-Krasovskii stability theory

KW - Passive filter

KW - Takagi-Sugeno fuzzy

UR - http://www.scopus.com/inward/record.url?scp=84876830627&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876830627&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2013.03.004

DO - 10.1016/j.ins.2013.03.004

M3 - Article

AN - SCOPUS:84876830627

VL - 238

SP - 126

EP - 137

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

ER -