An Effective Approach to Handling Noise and Drift in Electronic Noses
Sensor drift and noise handling in electronic noses (E-noses) are two different challenging problems. Sensor noise is caused by many factors such as temperature, pressure, humidity and cross interference. Noise can occur at any time, producing irrelevant or meaningless data. On the other hand, drift is appeared as a long term signal variation caused by unknown dynamic physical and chemical complex processes. Because sensor drift is not purely deterministic, it is very hard, if not impossible, to distinguish it from noise and vice versa. With respect to this property of E-nose, we propose a new approach to handle noise and sensor drift simultaneously. Our approach is based on kernel Fuzzy C-Mean clustering and fuzzy SVM (K-FSVM). The proposed method is compared to other currently used approaches, SVM, F-FSVM and KNN. Experiments are conducted on publicly available datasets. As the experimental results demonstrate, the performance of our proposed K-FSVM is superior than all other baseline methods in handling sensor drift and noise.
KeywordsE-nose noise drift classification Kernel Fuzzy C-Mean
Unable to display preview. Download preview PDF.
- 1.AL-Maskari, S., Saini, D.O.W.: Cyber infrastructure and data quality for environmental pollution control in Oman. In: Proceedings of the 2010 DAMD International Conference on Data Analysis, Data Quality and Metada, p. 71 (2010)Google Scholar
- 3.Pan, L., Yang, S.X.: A new electronic nose for downwind livestock farm odour measurement. In: Proceedings of the 2006 IEEE International Conference on Networking, Sensing and Control, ICNSC 2006, pp. 410–415 (2006)Google Scholar
- 4.Vergara, A., Vembu, S., Ayhan, T., Ryan, M.A., Homer, M.L., Huerta, R.: Chemical gas sensor drift compensation using classifier ensembles. Sensors and Actuators B: Chemical 166-167, 320–329 (2012)Google Scholar
- 6.Sharma, R., Chan, P., Tang, Z., Yan, G., Hsing, I., Sin, J.: Investigation of stability and reliability of tin oxide thin-film for integrated micro-machined gas sensor devices. Sensors and Actuators B: Chemical 81(1), 9–16 (2001)Google Scholar
- 9.Romain, A., Nicolas, J.: Long term stability of metal oxide-based gas sensors for e-nose environmental applications: An overview. Sensors and Actuators B: Chemical 146(2), 502–506 (2010); Selected papers from the 13th International Symposium on Olfaction and Electronic Nose ISOEN 2009Google Scholar
- 12.Marco, S., Pardo, A., Ortega, A., Samitier, J.: Gas identification with tin oxide sensor array and self organizing maps: Adaptive correction of sensor drifts. In: Instrumentation and Measurement Technology Conference, IMTC 1997, Proceedings of the Sensing, Processing, Networking, vol. 2, pp. 904–907. IEEE (May 1997)Google Scholar
- 13.Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Inc., New York (1995)Google Scholar
- 14.Zhang, D.Q., Chen, S.C.: A novel kernelized fuzzy c-means algorithm with application in medical image segmentation. Artificial Intelligence in Medicine 32(1), 37–50 (2004) (in China)Google Scholar