Improved chemical identification from sensor arrays using intelligent algorithms

Thaddeus A. Roppel; Denise M. Wilson

doi:10.1117/12.417459

26 February 2001 Improved chemical identification from sensor arrays using intelligent algorithms

Thaddeus A. Roppel, Denise M. Wilson

Proceedings Volume 4205, Advanced Environmental and Chemical Sensing Technology; (2001) https://doi.org/10.1117/12.417459
Event: Environmental and Industrial Sensing, 2000, Boston, MA, United States

Abstract

Intelligent signal processing algorithms are shown to improve identification rates significantly in chemical sensor arrays. This paper focuses on the use of independently derived sensor status information to modify the processing of sensor array data by using a fast, easily-implemented "best-match" approach to filling in missing sensor data. Most fault conditions of interest (e.g., stuck high, stuck low, sudden jumps, excess noise, etc.) can be detected relatively simply by adjunct data processing, or by on-board circuitry. The objective then is to devise, implement, and test methods for using this information to improve the identification rates in the presence of faulted sensors. In one typical example studied, utilizing separately derived, a-priori knowledge about the health of the sensors in the array improved the chemical identification rate by an artificial neural network from below 10 percent correct to over 99 percent correct. While this study focuses experimentally on chemical sensor arrays, the results are readily extensible to other types of sensor platforms.

Citation Download Citation

Thaddeus A. Roppel and Denise M. Wilson "Improved chemical identification from sensor arrays using intelligent algorithms", Proc. SPIE 4205, Advanced Environmental and Chemical Sensing Technology, (26 February 2001); https://doi.org/10.1117/12.417459

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available