Finite Difference approach on RBF networks for on-line system identification with lost packet
Radial Basis Function networks (RBF) is one form of feed forward neural network architecture which is popular besides multi layer preceptor (MLP). It is widely used especially in identifying a black box system. In many cases, identifying of the system process normally has lack of data or may lose so...
Saved in:
Main Authors: | , , |
---|---|
Format: | Conference or Workshop Item |
Published: |
2009
|
Subjects: | |
Online Access: | http://eprints.utp.edu.my/4643/1/ieee2009Afny.pdf http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5254687 http://eprints.utp.edu.my/4643/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.utp.eprints.4643 |
---|---|
record_format |
eprints |
spelling |
my.utp.eprints.46432017-01-19T08:25:16Z Finite Difference approach on RBF networks for on-line system identification with lost packet Andryani, N.A.C. Asirvadam, V.S. Hamid, N.H. TK Electrical engineering. Electronics Nuclear engineering QA75 Electronic computers. Computer science Radial Basis Function networks (RBF) is one form of feed forward neural network architecture which is popular besides multi layer preceptor (MLP). It is widely used especially in identifying a black box system. In many cases, identifying of the system process normally has lack of data or may lose some packets data needed in the identifying process. Finite Difference approach with its enhancement, Richardson Extrapolation, is used to improve the learning performance especially in the non linear learning parameter update for identifying system with lost packet data case in online manner. Since initializing of non linear learning's parameters is crucial in RBF networks' learning, random initialization is placed with some clustering method. Some unsupervised learning methods such as, K means clustering and Fuzzy K means clustering are used to replace it. All the possible combination methods in the initialization and update process try to improve the whole performance of the learning process regarding to the system identification with lost packet data case. It can be showed that Finite difference approach with dynamic step size on recursive prediction error for the non linear parameter update with appropriate initialization method succeed to perform better performance compared to extreme learning machine (ELM) as the previous learning method. 2009-08 Conference or Workshop Item NonPeerReviewed application/pdf http://eprints.utp.edu.my/4643/1/ieee2009Afny.pdf http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5254687 Andryani, N.A.C. and Asirvadam, V.S. and Hamid, N.H. (2009) Finite Difference approach on RBF networks for on-line system identification with lost packet. In: Electrical Engineering and Informatics, 2009. ICEEI '09. International Conference on, 5-7 Aug 2009, Bangi, Selangor, Malaysia. http://eprints.utp.edu.my/4643/ |
institution |
Universiti Teknologi Petronas |
building |
UTP Resource Centre |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Teknologi Petronas |
content_source |
UTP Institutional Repository |
url_provider |
http://eprints.utp.edu.my/ |
topic |
TK Electrical engineering. Electronics Nuclear engineering QA75 Electronic computers. Computer science |
spellingShingle |
TK Electrical engineering. Electronics Nuclear engineering QA75 Electronic computers. Computer science Andryani, N.A.C. Asirvadam, V.S. Hamid, N.H. Finite Difference approach on RBF networks for on-line system identification with lost packet |
description |
Radial Basis Function networks (RBF) is one form of feed forward neural network architecture which is popular besides multi layer preceptor (MLP). It is widely used especially in identifying a black box system. In many cases, identifying of the system process normally has lack of data or may lose some packets data needed in the identifying process. Finite Difference approach with its enhancement, Richardson Extrapolation, is used to improve the learning performance especially in the non linear learning parameter update for identifying system with lost packet data case in online manner. Since initializing of non linear learning's parameters is crucial in RBF networks' learning, random initialization is placed with some clustering method. Some unsupervised learning methods such as, K means clustering and Fuzzy K means clustering are used to replace it. All the possible combination methods in the initialization and update process try to improve the whole performance of the learning process regarding to the system identification with lost packet data case. It can be showed that Finite difference approach with dynamic step size on recursive prediction error for the non linear parameter update with appropriate initialization method succeed to perform better performance compared to extreme learning machine (ELM) as the previous learning method. |
format |
Conference or Workshop Item |
author |
Andryani, N.A.C. Asirvadam, V.S. Hamid, N.H. |
author_facet |
Andryani, N.A.C. Asirvadam, V.S. Hamid, N.H. |
author_sort |
Andryani, N.A.C. |
title |
Finite Difference approach on RBF networks for on-line system identification with lost packet |
title_short |
Finite Difference approach on RBF networks for on-line system identification with lost packet |
title_full |
Finite Difference approach on RBF networks for on-line system identification with lost packet |
title_fullStr |
Finite Difference approach on RBF networks for on-line system identification with lost packet |
title_full_unstemmed |
Finite Difference approach on RBF networks for on-line system identification with lost packet |
title_sort |
finite difference approach on rbf networks for on-line system identification with lost packet |
publishDate |
2009 |
url |
http://eprints.utp.edu.my/4643/1/ieee2009Afny.pdf http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5254687 http://eprints.utp.edu.my/4643/ |
_version_ |
1738655358962892800 |
score |
13.211869 |