Game Theoretic Pattern Analysis for Identification of Odors/Gases Using Response of a Poorly Selective Sensor Array

This paper presents a novel approach toward the identification of odors/gases using game theory for feature selection. A previously reported tin-oxide-based sensor array capable of sensing at room temperature is chosen to extract the data. The sensor array is exposed to four different odors/gases and the response pattern obtained reveals its poor selectivity. The task of classifying the sampled data into four classes is modeled as a coalitional game in which each sensor of the array acted like a player forming coalitions with other players. A pay-off function is associated with each possible coalition of players with higher pay-offs being given to coalitions that maximize class separability of the data. Shapley value is used to quantify the contribution of each player yielding a standard pattern for each odor class. Each sample of the test vector is then matched with the standard pattern and the odor class yielding minimum Euclidean distance is assigned to the test sample. A weighting scheme for relative scaling of the test samples is also proposed. It is observed that more than 89% of the samples were identified correctly using the proposed technique thereby proving its efficacy.

I. Introduction

Identification of individual odors using an array of solid state chemical sensors is a highly popular research topic in the current technological scenario. Low cost of fabrication, design ruggedness and portability are some of the advantages associated with solid state sensing devices. Tin oxide based chemical sensors fabricated with thick film technology deserve special mention since they entail all the above mentioned advantages of a popular solid state sensor. Their cost of fabrication is arguably the lowest. They have a robust design and are highly versatile since they possess appreciable sensitivity to a plethora of odors and gases. Researchers have obtained encouraging results by employing thick film tin oxide sensors in the analysis of wine odor [1]–[3], in food quality monitoring [4]–[6] and in discrimination of volatile organic compounds [7]. However, there are some limitations associated with thick film tin oxide based sensors most prominent among them are enlisted as follows. 1)

Poor selectivity due to cross sensitivity of a sensor to different odors.

2)

Inability to sense at room temperature and thus, requirement of an integrated heater fabricated along with the device.