FPGA implementation of a neural network classifier for gas sensor array applications

Faycal Benrekia, Mokhtar Attari, Amine Bermak, Khaled Belhout

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

A primitive gas recognition system which can discriminate limited species of industrial gas was designed and simulated. The 'electronic nose' consists of an array of 8 microhotplate based SnO2thin film gas sensors with different selectivity patterns, signal collecting unit and a signal pattern recognition and decision part in programmable logic device chip. BP (Back Propagation) neural networks with Multilayer Perceptron structure was designed and implemented on FPGA (Field Programmable Gate Array), of twenty thousand gate level chip by VHDL language for processing the input signals from 8 kinds of gas sensors. The network contained eight input units, one hidden layer with 4 neurons and output with 5 regular neurons. The 'electronic nose' system successfully discriminated 5 kinds of industrial gases in computer simulation. A small application has been tested on the APS X208 FPGA test board.

Original languageEnglish
Title of host publication2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009
PublisherIEEE Computer Society
ISBN (Print)9781424443468
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009 - Djerba, Taiwan, Province of China
Duration: 23 Mar 200926 Mar 2009

Other

Other2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009
CountryTaiwan, Province of China
CityDjerba
Period23/3/0926/3/09

Fingerprint

Sensor arrays
Chemical sensors
Field programmable gate arrays (FPGA)
Classifiers
Neural networks
Neurons
Gases
Computer hardware description languages
Logic devices
Multilayer neural networks
Backpropagation
Pattern recognition
Computer simulation
Processing
Electronic nose

Keywords

  • E-nose
  • FPGA-implementation
  • Gas sensor
  • Neural network classifier
  • VHDL

ASJC Scopus subject areas

  • Hardware and Architecture
  • Signal Processing
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Benrekia, F., Attari, M., Bermak, A., & Belhout, K. (2009). FPGA implementation of a neural network classifier for gas sensor array applications. In 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009 [4956804] IEEE Computer Society. https://doi.org/10.1109/SSD.2009.4956804

FPGA implementation of a neural network classifier for gas sensor array applications. / Benrekia, Faycal; Attari, Mokhtar; Bermak, Amine; Belhout, Khaled.

2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009. IEEE Computer Society, 2009. 4956804.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Benrekia, F, Attari, M, Bermak, A & Belhout, K 2009, FPGA implementation of a neural network classifier for gas sensor array applications. in 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009., 4956804, IEEE Computer Society, 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009, Djerba, Taiwan, Province of China, 23/3/09. https://doi.org/10.1109/SSD.2009.4956804
Benrekia F, Attari M, Bermak A, Belhout K. FPGA implementation of a neural network classifier for gas sensor array applications. In 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009. IEEE Computer Society. 2009. 4956804 https://doi.org/10.1109/SSD.2009.4956804
Benrekia, Faycal ; Attari, Mokhtar ; Bermak, Amine ; Belhout, Khaled. / FPGA implementation of a neural network classifier for gas sensor array applications. 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009. IEEE Computer Society, 2009.
@inproceedings{947c674728b2439d8ec1a3d4e5c8c765,
title = "FPGA implementation of a neural network classifier for gas sensor array applications",
abstract = "A primitive gas recognition system which can discriminate limited species of industrial gas was designed and simulated. The 'electronic nose' consists of an array of 8 microhotplate based SnO2thin film gas sensors with different selectivity patterns, signal collecting unit and a signal pattern recognition and decision part in programmable logic device chip. BP (Back Propagation) neural networks with Multilayer Perceptron structure was designed and implemented on FPGA (Field Programmable Gate Array), of twenty thousand gate level chip by VHDL language for processing the input signals from 8 kinds of gas sensors. The network contained eight input units, one hidden layer with 4 neurons and output with 5 regular neurons. The 'electronic nose' system successfully discriminated 5 kinds of industrial gases in computer simulation. A small application has been tested on the APS X208 FPGA test board.",
keywords = "E-nose, FPGA-implementation, Gas sensor, Neural network classifier, VHDL",
author = "Faycal Benrekia and Mokhtar Attari and Amine Bermak and Khaled Belhout",
year = "2009",
doi = "10.1109/SSD.2009.4956804",
language = "English",
isbn = "9781424443468",
booktitle = "2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - FPGA implementation of a neural network classifier for gas sensor array applications

AU - Benrekia, Faycal

AU - Attari, Mokhtar

AU - Bermak, Amine

AU - Belhout, Khaled

PY - 2009

Y1 - 2009

N2 - A primitive gas recognition system which can discriminate limited species of industrial gas was designed and simulated. The 'electronic nose' consists of an array of 8 microhotplate based SnO2thin film gas sensors with different selectivity patterns, signal collecting unit and a signal pattern recognition and decision part in programmable logic device chip. BP (Back Propagation) neural networks with Multilayer Perceptron structure was designed and implemented on FPGA (Field Programmable Gate Array), of twenty thousand gate level chip by VHDL language for processing the input signals from 8 kinds of gas sensors. The network contained eight input units, one hidden layer with 4 neurons and output with 5 regular neurons. The 'electronic nose' system successfully discriminated 5 kinds of industrial gases in computer simulation. A small application has been tested on the APS X208 FPGA test board.

AB - A primitive gas recognition system which can discriminate limited species of industrial gas was designed and simulated. The 'electronic nose' consists of an array of 8 microhotplate based SnO2thin film gas sensors with different selectivity patterns, signal collecting unit and a signal pattern recognition and decision part in programmable logic device chip. BP (Back Propagation) neural networks with Multilayer Perceptron structure was designed and implemented on FPGA (Field Programmable Gate Array), of twenty thousand gate level chip by VHDL language for processing the input signals from 8 kinds of gas sensors. The network contained eight input units, one hidden layer with 4 neurons and output with 5 regular neurons. The 'electronic nose' system successfully discriminated 5 kinds of industrial gases in computer simulation. A small application has been tested on the APS X208 FPGA test board.

KW - E-nose

KW - FPGA-implementation

KW - Gas sensor

KW - Neural network classifier

KW - VHDL

UR - http://www.scopus.com/inward/record.url?scp=84988045141&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988045141&partnerID=8YFLogxK

U2 - 10.1109/SSD.2009.4956804

DO - 10.1109/SSD.2009.4956804

M3 - Conference contribution

SN - 9781424443468

BT - 2009 6th International Multi-Conference on Systems, Signals and Devices, SSD 2009

PB - IEEE Computer Society

ER -