Pattern recognition of dynamic chemical-sensor responses by using LVQ algorithm

Signal processing and computation of chemical sensor responses is an important step in recognizing the target gas because a chemical sensor is not usually specific to a particular kind of gas. The authors present a chemical-sensing system using the learning vector quantization (LVQ) algorithm to classify dynamic sensor array responses. The system consists of an array of piezoelectric quartz-crystal microbalance (QCM) sensors, each coated with a different organic film. They explain the mechanism of the sensor, providing the dynamic response model whose coefficients are used in recognizing the target gas. These coefficients are estimated iteratively using the Kalman filter algorithm, and are transformed to form a pattern vector, which is immediately classified by LVQ in real time. Identification tests of 14 vapors are performed using the system installed in a personal computer