Quantization errors in committee machine for gas sensor applications

Minghua Shi, Sofiane Brahim-Belhouari, Amine Bermak

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In a digital implementation of a gas identification system, the mapping of continuous real parameters value into a finite set of discrete values introduces an error into the system. This paper presents the results of an investigation into the effects of parameters quantization on different classifiers (KNN, MLP and GMM). We propose a committee machine to decrease the classification performance degradation due to the quantization errors. The simulation results show that the committee machine always outperforms single classifier and the gain in classification performance is greater for reduced number of bits.

Original languageEnglish
Article number1464986
Pages (from-to)1911-1914
Number of pages4
JournalProceedings - IEEE International Symposium on Circuits and Systems
DOIs
Publication statusPublished - 2005
Externally publishedYes

Fingerprint

Chemical sensors
Classifiers
Identification (control systems)
Degradation
Gases

Keywords

  • Committee machine
  • Gas sensors
  • Quantization error

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Quantization errors in committee machine for gas sensor applications. / Shi, Minghua; Brahim-Belhouari, Sofiane; Bermak, Amine.

In: Proceedings - IEEE International Symposium on Circuits and Systems, 2005, p. 1911-1914.

Research output: Contribution to journalArticle

@article{7066dc71893a48aa9ad0c112eec33f50,
title = "Quantization errors in committee machine for gas sensor applications",
abstract = "In a digital implementation of a gas identification system, the mapping of continuous real parameters value into a finite set of discrete values introduces an error into the system. This paper presents the results of an investigation into the effects of parameters quantization on different classifiers (KNN, MLP and GMM). We propose a committee machine to decrease the classification performance degradation due to the quantization errors. The simulation results show that the committee machine always outperforms single classifier and the gain in classification performance is greater for reduced number of bits.",
keywords = "Committee machine, Gas sensors, Quantization error",
author = "Minghua Shi and Sofiane Brahim-Belhouari and Amine Bermak",
year = "2005",
doi = "10.1109/ISCAS.2005.1464986",
language = "English",
pages = "1911--1914",
journal = "Proceedings - IEEE International Symposium on Circuits and Systems",
issn = "0271-4310",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Quantization errors in committee machine for gas sensor applications

AU - Shi, Minghua

AU - Brahim-Belhouari, Sofiane

AU - Bermak, Amine

PY - 2005

Y1 - 2005

N2 - In a digital implementation of a gas identification system, the mapping of continuous real parameters value into a finite set of discrete values introduces an error into the system. This paper presents the results of an investigation into the effects of parameters quantization on different classifiers (KNN, MLP and GMM). We propose a committee machine to decrease the classification performance degradation due to the quantization errors. The simulation results show that the committee machine always outperforms single classifier and the gain in classification performance is greater for reduced number of bits.

AB - In a digital implementation of a gas identification system, the mapping of continuous real parameters value into a finite set of discrete values introduces an error into the system. This paper presents the results of an investigation into the effects of parameters quantization on different classifiers (KNN, MLP and GMM). We propose a committee machine to decrease the classification performance degradation due to the quantization errors. The simulation results show that the committee machine always outperforms single classifier and the gain in classification performance is greater for reduced number of bits.

KW - Committee machine

KW - Gas sensors

KW - Quantization error

UR - http://www.scopus.com/inward/record.url?scp=67649124845&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649124845&partnerID=8YFLogxK

U2 - 10.1109/ISCAS.2005.1464986

DO - 10.1109/ISCAS.2005.1464986

M3 - Article

SP - 1911

EP - 1914

JO - Proceedings - IEEE International Symposium on Circuits and Systems

JF - Proceedings - IEEE International Symposium on Circuits and Systems

SN - 0271-4310

M1 - 1464986

ER -