Electronic noses and their applications in environmental monitoring

Compact, portable systems capable of quickly identifying contaminants in the field are of great importance when monitoring the environment. In this paper, we examine the effectiveness of using artificial neural networks for real-time data analysis of a sensor array. Analyzing the sensor data in parallel may allow for rapid identification of contaminants in the field without requiring highly selective component sensors. A sensor array combined with a data analysis module is referred to as an electronic nose. In this paper, we investigate the trade off between sensor sensitivity and selectivity relating to the applications of neural network based-electronic noses in environmental monitoring. We use a prototype electronic nose which consists of nine tin-oxide Taguchi-type sensors, a temperature sensor, and a humidity sensor. We illustrate that by using neural network based analysis of a sensor data, the selectivity of a sensor array may be significantly improved, especially when some (or all) sensors are not highly selective.