Rank extraction in tin-oxide sensor arrays

Abstract

It is shown that data pre-processing by rank-order filtering can significantly improve the odor discrimination capability of an array of chemical sensors, while simultaneously reducing the amount of data to be processed. This work is a first example in feature extraction from tin-oxide sensors that both reduces the size of the data set and simultaneously improves the discrimination performance of the array. This work is aimed toward the design of remote sensor modules where bandwidth reduction and improved accuracy are both essential to system performance. The effectiveness of extracting rank from a 30-element array of tin-oxide sensors is presented. Results are extrapolated to other arrays of chemical sensors whose specificities and response characteristics overlap. Methods for processing data and extracting rank-related features from arrays of tin-oxide sensors are comparatively analyzed. Processing parameters studied include those related to temporal filtering and window-averaging, pre-scaling (to remove baseline), sample acquisition time, and the number of ranks used in rank-order filtering of the data during the transient and steady state response. Cluster analysis, including principal component analysis (PCA) and a novel method described herein, is used to determine which of these processing techniques are most effective. Artificial neural networks, specifically multi-layer perceptrons and radial basis function networks, are used to further investigate the ability to discriminate odors on the basis of the extracted features. The analysis is performed for an array of 30 tin-oxide sensors applied to detecting a sampling of breath alcohol mixtures (beer, wine, vodka) and common interferents (acetone, formaldehyde, isopropyl). Ammonia is included as a contrast substance. For the set of seven odorants studied, it is found that using rank-order filtering with 10 or more ranks improves odor recognition rate by a multi-layer perceptron neural network from 92% to 95%. If one odor (vodka) is removed from the study set, the recognition rate for the remaining odors improves from 95% (with no rank-order filtering) to 99%. Simultaneously, the dimensions of the data set for each odor are reduced from 30 sensors×18,000 time steps (12 bit samples) to N integer values, where N is the number of ranks used in the rank-order filtering.