P1.9.22 An Electronic Nose Recognition Algorithm Based on PCA-ICA Preprocessing and Fuzzy Neural Network
- Event
- 14th International Meeting on Chemical Sensors - IMCS 2012
2012-05-20 - 2012-05-23
Nürnberg/Nuremberg, Germany - Chapter
- P1.9 Technology and Application
- Author(s)
- W. Yan, Z. Tang, J. Yang - School of Electronic Science and Technology, Dalian University of Technology (P.R. China), G. Wei - School of Information and Electronics, Shandong Institute of Business and Technology (P.R. China)
- Pages
- 1227 - 1230
- DOI
- 10.5162/IMCS2012/P1.9.22
- ISBN
- 978-3-9813484-2-2
- Price
- free
Abstract
To improve the recognition performance of electronic noses detecting gas mixtures, a PCA-ICA signal preprocessing and fuzzy neural network based recognition algorithm is proposed. In this approach, signals of electronic noses are firstly preprocessed effectively by combination of Principal Component Analysis (PCA) and Independent Component Analysis (ICA), and then processed with a fuzzy Takagi- Sugeno system integrated with multi neural networks for the purpose of quantification of gas concentrations. Experiment results show that the alcohol concentration recognition performance is highly improved in alcohol and gasoline mixtures even interfered by smokes.