Investigation of electronic nose sensor drift correction methods and their application to environmental samples

Sensor drift is one of the main problems connected to gas sensors, and is one of the main responsible of their limited diffusion and adoption for different real-life applications. For this reason, different algorithms have been proposed in literature for drift mitigation. This paper has the main objective of giving an overview of the methods that can be applied in order to correct sensor responses relevant to real environmental odours and investigating some of those methods that were identified as most suitable for the desired application. The two methods that were considered for the experimental work are the PCA component correction and the Orthogonal signal correction methods. For the application of the first method, different calibrants were investigated, finally selecting acetone and ammonia as the most suitable compounds for the electronic nose calibration. The preliminary results obtained prove both methods to be robust towards small calibration set sizes and show a good performance in improving improve data clusterization and discrimination.