DIGITAL SIGNAL PROCESSING ALGORITHMS FOR GAS CONCENTRATION MEASUREMENT

B. Baraboi, M. Cretu, C. Donciu, C. Fosalau

Abstract

The paper presents two signal processing algorithms for continuous measurement of gas concentration using low cost metal oxide semiconductor sensors. A major limitation of these sensors in quantitative analysis of gases is their non-linearity. A good solution of this problem is to combine these sensors with appropriate signal processing techniques. The first algorithm that we developed is based on artificial neural networks. As the second option, we constructed a model based on the characteristic formula of metal oxide semiconductor sensors. A detailed analysis of the developed algorithms and a comparison between them is presented in the paper.

Keywords

quantitative gas analysis, metal oxide semiconductor sensors, digital signal processing, virtual instrumentation

Download

IMEKO-TC4-2001-133.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings