A Method of Feature Extraction on Recovery Curves for Fast Recognition Application With Metal Oxide Gas Sensor Array

In the fast recognition applications of electronic nose, not only the recognition time is important, another parameter response-recovery time also needs to be considered. The response-recovery time could be defined as the time from the beginning of measuring one sample to the state of being ready for new sample measurement. An electronic nose with nine metal oxide (MOX) gas sensors and a method of feature extraction on sensor recovery curves were presented in this paper. The electronic nose was designed to reduce the recognition time and the response-recovery time synchronously. In the sampling module of the electronic nose, there were two pumps, which could let the sensor quickly recovered. The feature extraction method could rapidly extract features from sensor recovery curves with robust information. Nine volatile organic compounds (VOCs) gas samples were measured with the electronic nose. The correct recognition ratios under 10 and 15 s recognition time are 91.0% and 95.8%, respectively. The mean response-recovery time of these sensors in the measurements was 33.5 s, which was about 42.7% of the response-recovery time in typical traditional gas sample measurements. The results show that the proposed feature extraction method could extract robust information with short recognition time and response-recovery time.