Combinaton of nonlinear equalization and simple detection

Hangyu Cho, Choongchae Woo, Daesik Hong

Research output: Contribution to journalConference article

Abstract

The linear equalizer in PRML was replaced with partial response neural equalizer (PRNEML) to combat nonlinear distortions. The advantage of discrete matched filter structure was that the decision feedback could be incorporated into neural networks. MDFE with neural networks yielded poor performance due to conditional training of inner-level eye-pattern.

Original languageEnglish
JournalDigests of the Intermag Conference
Publication statusPublished - 2002 Dec 1

Fingerprint

Equalizers
Neural networks
Nonlinear distortion
Matched filters
Feedback

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

@article{9bc4e1f2158a44a3a6d977ec4bf3cb21,
title = "Combinaton of nonlinear equalization and simple detection",
abstract = "The linear equalizer in PRML was replaced with partial response neural equalizer (PRNEML) to combat nonlinear distortions. The advantage of discrete matched filter structure was that the decision feedback could be incorporated into neural networks. MDFE with neural networks yielded poor performance due to conditional training of inner-level eye-pattern.",
author = "Hangyu Cho and Choongchae Woo and Daesik Hong",
year = "2002",
month = "12",
day = "1",
language = "English",
journal = "Digests of the Intermag Conference",
issn = "0074-6843",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Combinaton of nonlinear equalization and simple detection. / Cho, Hangyu; Woo, Choongchae; Hong, Daesik.

In: Digests of the Intermag Conference, 01.12.2002.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Combinaton of nonlinear equalization and simple detection

AU - Cho, Hangyu

AU - Woo, Choongchae

AU - Hong, Daesik

PY - 2002/12/1

Y1 - 2002/12/1

N2 - The linear equalizer in PRML was replaced with partial response neural equalizer (PRNEML) to combat nonlinear distortions. The advantage of discrete matched filter structure was that the decision feedback could be incorporated into neural networks. MDFE with neural networks yielded poor performance due to conditional training of inner-level eye-pattern.

AB - The linear equalizer in PRML was replaced with partial response neural equalizer (PRNEML) to combat nonlinear distortions. The advantage of discrete matched filter structure was that the decision feedback could be incorporated into neural networks. MDFE with neural networks yielded poor performance due to conditional training of inner-level eye-pattern.

UR - http://www.scopus.com/inward/record.url?scp=17944395176&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17944395176&partnerID=8YFLogxK

M3 - Conference article

JO - Digests of the Intermag Conference

JF - Digests of the Intermag Conference

SN - 0074-6843

ER -