The recognition of Chinese spirits using electronic nose with dynamic method

Author

Yu, Peng ; Pan, Min ; Chen, Yuquan ; Li, Guang

Author_Institution

Biosensor Nat. Special Lab., Zhejiang Univ., Hangzhou, China

Volume

3

fYear

2001

fDate

2001

Firstpage

3118

Abstract

This paper describes a method to recognize Chinese spirits using a gas sensor array being dynamically heated. An FFT and an RBF neural network are employed for dynamic signal extraction and pattern recognition, respectively. Four kinds of SnO₂ gas sensors were chosen to build a sensor array for our experiments. Three Chinese spirits, which had similar odors were chosen to be recognized. The response signals were collected while the sensor array was periodically heated. After the FFT was applied on low frequency segment in a specific period, we used RBF neural network to analyze the results for pattern recognition. The recognition rate was 98%.

Keywords

chemistry computing; fast Fourier transforms; gas sensors; pattern recognition; radial basis function networks; signal processing; Chinese spirits recognition; RBF neural network; SnO₂; artificial olfactory system; dynamic method; dynamic signal extraction; electronic nose; gas sensor array; low frequency segment; odors; periodically heated sensor array; response signals; Electrical resistance measurement; Electronic noses; Gas detectors; Neural networks; Olfactory; Pattern recognition; Polymers; Sensor arrays; Sensor systems; Thermal sensors;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2001. Proceedings of the 23rd Annual International Conference of the IEEE

ISSN

1094-687X

Print_ISBN

0-7803-7211-5

Type

conf

DOI

10.1109/IEMBS.2001.1017462

Filename

1017462