Multilayer perceptron-based DFE with lattice structure

The severely distorting channels limit the use of linear equalizers and the use of the nonlinear equalizers then becomes justifiable. Neural-network-based equalizers, especially the multilayer perceptron (MLP)-based equalizers, are computationally efficient alternative to currently used nonlinear fi...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Zerguine, A. (author)
مؤلفون آخرون: Shafi, A. (author), Bettayeb, M. (author), unknown (author)
التنسيق: article
منشور في: 2001
الموضوعات:
الوصول للمادة أونلاين:https://eprints.kfupm.edu.sa/id/eprint/14047/1/14047_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14047/2/14047_2.doc
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513393329700864
author Zerguine, A.
author2 Shafi, A.
Bettayeb, M.
unknown
author2_role author
author
author
author_facet Zerguine, A.
Shafi, A.
Bettayeb, M.
unknown
author_role author
dc.creator.none.fl_str_mv Zerguine, A.
Shafi, A.
Bettayeb, M.
unknown
dc.date.none.fl_str_mv 2001-05
2020
dc.format.none.fl_str_mv application/pdf
application/msword
dc.identifier.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14047/1/14047_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14047/2/14047_2.doc
(2001) Multilayer perceptron-based DFE with lattice structure. Neural Networks, IEEE Transactions on, 12.
dc.language.none.fl_str_mv en
en
dc.publisher.none.fl_str_mv IEEE
dc.relation.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14047/
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Computer
dc.title.none.fl_str_mv Multilayer perceptron-based DFE with lattice structure
dc.type.none.fl_str_mv Article
PeerReviewed
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description The severely distorting channels limit the use of linear equalizers and the use of the nonlinear equalizers then becomes justifiable. Neural-network-based equalizers, especially the multilayer perceptron (MLP)-based equalizers, are computationally efficient alternative to currently used nonlinear filter realizations, e.g., the Volterra type. The drawback of the MLP-based equalizers is, however, their slow rate of convergence, which limit their use in practical systems. In this work, the effect of whitening the input data in a multilayer perceptron-based decision feedback equalizer (DFE) is evaluated. It is shown from computer simulations that whitening the received data employing adaptive lattice channel equalization algorithms improves the convergence rate and bit error rate performances of multilayer perceptron-based DFE. The adaptive lattice algorithm is a modification to the one developed by Ling and Proakis (1985). The consistency in performance is observed in both time-invariant and time-varying channels. Finally, it is found in this work that, for time-invariant channels, the MLP DFE outperforms the least mean squares (LMS)-based DFE. However, for time-varying channels comparable performance is obtained for the two configurations
eu_rights_str_mv openAccess
format article
id KFUPM_7627bbca31cf053ab0165cc4e22f6f79
identifier_str_mv (2001) Multilayer perceptron-based DFE with lattice structure. Neural Networks, IEEE Transactions on, 12.
language_invalid_str_mv en
network_acronym_str KFUPM
network_name_str King Fahd University of Petroleum and Minerals
oai_identifier_str oai::14047
publishDate 2001
publisher.none.fl_str_mv IEEE
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Multilayer perceptron-based DFE with lattice structureZerguine, A.Shafi, A.Bettayeb, M.unknownComputerThe severely distorting channels limit the use of linear equalizers and the use of the nonlinear equalizers then becomes justifiable. Neural-network-based equalizers, especially the multilayer perceptron (MLP)-based equalizers, are computationally efficient alternative to currently used nonlinear filter realizations, e.g., the Volterra type. The drawback of the MLP-based equalizers is, however, their slow rate of convergence, which limit their use in practical systems. In this work, the effect of whitening the input data in a multilayer perceptron-based decision feedback equalizer (DFE) is evaluated. It is shown from computer simulations that whitening the received data employing adaptive lattice channel equalization algorithms improves the convergence rate and bit error rate performances of multilayer perceptron-based DFE. The adaptive lattice algorithm is a modification to the one developed by Ling and Proakis (1985). The consistency in performance is observed in both time-invariant and time-varying channels. Finally, it is found in this work that, for time-invariant channels, the MLP DFE outperforms the least mean squares (LMS)-based DFE. However, for time-varying channels comparable performance is obtained for the two configurationsIEEE2001-052020ArticlePeerReviewedinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfapplication/mswordhttps://eprints.kfupm.edu.sa/id/eprint/14047/1/14047_1.pdfhttps://eprints.kfupm.edu.sa/id/eprint/14047/2/14047_2.doc (2001) Multilayer perceptron-based DFE with lattice structure. Neural Networks, IEEE Transactions on, 12. enenhttps://eprints.kfupm.edu.sa/id/eprint/14047/info:eu-repo/semantics/openAccessoai::140472019-11-01T14:03:53Z
spellingShingle Multilayer perceptron-based DFE with lattice structure
Zerguine, A.
Computer
status_str publishedVersion
title Multilayer perceptron-based DFE with lattice structure
title_full Multilayer perceptron-based DFE with lattice structure
title_fullStr Multilayer perceptron-based DFE with lattice structure
title_full_unstemmed Multilayer perceptron-based DFE with lattice structure
title_short Multilayer perceptron-based DFE with lattice structure
title_sort Multilayer perceptron-based DFE with lattice structure
topic Computer
url https://eprints.kfupm.edu.sa/id/eprint/14047/1/14047_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14047/2/14047_2.doc