Electrochemical sensor for detection of multiple environmental contaminants through advanced signal processing

摘要

The possibility of large-scale attacks using chemical warfare agents (CWAs) has exposed the critical need for fundamental research enabling the reliable, unambiguous, and early detection of trace CWAs and toxic industrial chemicals. This paper presents an unique approach for identification and classification of environmental contaminants by perturbing an electrochemical (EC) sensor with an oscillating potential rather than static voltage levels for extraction of statistically rich information from the current response. The dynamic response, being a function of the degree and mechanism of contamination, is then processed with a symbolic dynamic filter for extraction of representative patterns, which are then classified using a trained neural network. The approach presented in this paper promises to extend sensing power and sensitivity of these EC sensors by augmenting and complementing the sensor technology with state-of-the-art embedded real time signal processing capabilities.