Wavelet-Assisted Phase Space Analysis for Improved VOCs Discrimination Using SAW Sensor Transients
- 17 Downloads
SAW sensor is a well-known chemical sensor used for the identification of different chemical vapors. When a vapor is subjected to the SAW sensor, it results a unique dynamic trace as a transient signal which encodes its identity information. By analyzing this transient signal, the identity of the vapors can be obtained. This work presents an improved wavelet-assisted phase space-based vapor identification methodology for SAW sensor analysis. The phase space is created by computing the transient signal and its first time derivative. The performance of the proposed approach is compared with commonly used wavelet-assisted transient signal analysis in the measurement space. The discrimination analysis is performed qualitatively by principal component analysis and quantitatively by class separability measure. In this work, seven organic vapors are subjected to a PIB-coated SAW sensor in virtual environment and their transient responses are simulated. For making the transient responses near to the actual experimental conditions, random noise is added into the responses. It is observed that the wavelet-assisted phase space analysis shows better discrimination in principal component space and produces more than four times separability in comparison to the wavelet-assisted transient signal analysis.
KeywordsSAW sensor Wavelet Phase space Class separability measure
Authors gratefully acknowledges DST-SERB, Government of India (N-PDF reference no. PDF/2016/000512) and National Centre for Flexible Electronics, IIT Kanpur, India.
- 2.Sberveglieri, G.: Gas Sensors- Principles, Operations and Applications, pp. 281–306. Springer Science + Business Media, Dordrecht (1992)Google Scholar
- 3.Pearce, T.C., Schiffman, S.S., Nagle, H.T., Gardner, J.W.: Handbook of Machine Olfaction, 2nd edn, pp. 33–98. Wiley-VCH, Weinheim (2003)Google Scholar
- 4.Corine, D., Dominique, R., Jacques, P., Colette, T., Roger, P.: A surface acoustic wave Gas sensor: detection of organophosphorus compounds. Sens. Actuators B 24(1–3), 58–61 (1995)Google Scholar
- 13.Singh, P., Yadava, R.D.S.: Transient feature extraction based on phase space fusion by partial-least-square regression analysis of sensor array signals. In: IEEE International Conference on Emerging Trends in Electrical and Computer Technology, pp. 676–680, Nagercoil, India (2011)Google Scholar
- 14.Rao, R.M., Boparkidar, A.S.: Wavelet transform: introduction to theory and applications, 1st edn. Addision-Wiesley, USA (1998)Google Scholar
- 18.Singh, P., Yadava, R.D.S.: A fusion approach to feature extraction by wavelet decomposition and principal component analysis in transient signal processing of SAW odor sensor array. Sens. Transducers J. 126(3), 64–73 (2011)Google Scholar