Review on data-driven approaches for improving the selectivity of MOX-sensors

Metal Oxide sensors, thanks to their low cost, small size and wide recovery, are increasingly used in various industrial applications for the detection of gases and gas mixtures. However, due to their operating principle, these sensors are not selective enough, thereby preventing the expansion of their fields of application. The scientific community has tackled this issue by proposing methods and algorithms to improve the selectivity of these sensors. This paper provides an overview of existing approaches for detection and identification in the general case, as well as those used in the context of gas discrimination. A discussion of the methods used and the results announced is proposed to highlight their performance and to identify promising research directions and perspectives to allow a significant advance in research in this field.