A novel olfactory neural network for classification of Chinese liquors using electronic nose

Abstract

We proposed a novel bio-inspired olfactory neural network to perform data processing in our designed electronic nose (e-nose). This olfactory neural network mimics the main structures of the mammal animals' olfactory system. It contains olfactory sensing neurons, mitral cells and granule cells. The proposed method can directly use the raw data collected from sensors in the e-nose without any data preprocessing or feature extraction. The output neurons of the network can change the sensors' responses into two new time series, of which we only use the variances to perform classification. This significantly simplifies the data processing procedure in e-noses. In order to test the performance of the e-nose and the proposed olfactory neural network, five kinds of Chinese liquors of the same aroma style were used. The classification rates of the proposed method and traditional data processing method combined with linear discrimination analysis (LDA) are 94% and 84%, respectively.