A Robust andLow-Complexity Gas Recognition Technique for On-Chip Tin-Oxide Gas Sensor Array

[摘要] Gas recognition is a new emerging research area with many civil, military, and industrial applications. The successof any gas recognition system depends on its computationalcomplexity and its robustness. In this work, we propose anew low-complexity recognition method which is tested andsuccessfully validated for tin-oxide gas sensor array chip. Therecognition system is based on a vector angle similarity measurebetween the query gas and the representatives of the different gasclasses. The latter are obtained using a clustering algorithm basedon the same measure within the training data set. Experimented results on our in-house gas sensors array show more than98%of correct recognition. The robustness of the proposed methodis tested by recognizing gas measurements with simulated drift.Less than1%of performance degradation is noted at the worst casescenario which represents a significant improvement whencompared to the current state-of-the-art.

[学科分类] 自动化工程