Feature extraction for neural network equalizers trained with multi-gradient

Chulhee Lee, Jinwook Go, Byungjoon Baek

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

In this paper, we view equalization as a multi-class classification problem and use neural networks for classification. In particular, we use a recently published training algorithm, multi-gradient, to train neural networks. Then, we apply a feature extraction method to obtain more efficient neural networks. Experiments show that the neural network equalizers which view equalization as multi-class problems provide significantly improved performances compared to neural network equalizers trained by the conventional LMS algorithm while the feature extraction method significantly reduces the complexity of the neural network equalizers.

Original languageEnglish
Pages (from-to)1575-1578
Number of pages4
JournalIEEE International Conference on Neural Networks - Conference Proceedings
Volume2
Publication statusPublished - 2004 Dec 1

Fingerprint

Equalizers
Feature extraction
Neural networks
Experiments

All Science Journal Classification (ASJC) codes

  • Software

Cite this

@article{3a0eaa97e0754aeb960cfff90f74b096,
title = "Feature extraction for neural network equalizers trained with multi-gradient",
abstract = "In this paper, we view equalization as a multi-class classification problem and use neural networks for classification. In particular, we use a recently published training algorithm, multi-gradient, to train neural networks. Then, we apply a feature extraction method to obtain more efficient neural networks. Experiments show that the neural network equalizers which view equalization as multi-class problems provide significantly improved performances compared to neural network equalizers trained by the conventional LMS algorithm while the feature extraction method significantly reduces the complexity of the neural network equalizers.",
author = "Chulhee Lee and Jinwook Go and Byungjoon Baek",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "2",
pages = "1575--1578",
journal = "IEEE International Conference on Neural Networks - Conference Proceedings",
issn = "1098-7576",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Feature extraction for neural network equalizers trained with multi-gradient. / Lee, Chulhee; Go, Jinwook; Baek, Byungjoon.

In: IEEE International Conference on Neural Networks - Conference Proceedings, Vol. 2, 01.12.2004, p. 1575-1578.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Feature extraction for neural network equalizers trained with multi-gradient

AU - Lee, Chulhee

AU - Go, Jinwook

AU - Baek, Byungjoon

PY - 2004/12/1

Y1 - 2004/12/1

N2 - In this paper, we view equalization as a multi-class classification problem and use neural networks for classification. In particular, we use a recently published training algorithm, multi-gradient, to train neural networks. Then, we apply a feature extraction method to obtain more efficient neural networks. Experiments show that the neural network equalizers which view equalization as multi-class problems provide significantly improved performances compared to neural network equalizers trained by the conventional LMS algorithm while the feature extraction method significantly reduces the complexity of the neural network equalizers.

AB - In this paper, we view equalization as a multi-class classification problem and use neural networks for classification. In particular, we use a recently published training algorithm, multi-gradient, to train neural networks. Then, we apply a feature extraction method to obtain more efficient neural networks. Experiments show that the neural network equalizers which view equalization as multi-class problems provide significantly improved performances compared to neural network equalizers trained by the conventional LMS algorithm while the feature extraction method significantly reduces the complexity of the neural network equalizers.

UR - http://www.scopus.com/inward/record.url?scp=10944250514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=10944250514&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:10944250514

VL - 2

SP - 1575

EP - 1578

JO - IEEE International Conference on Neural Networks - Conference Proceedings

JF - IEEE International Conference on Neural Networks - Conference Proceedings

SN - 1098-7576

ER -