A novel principal component-based virtual sensor approach for efficient classification of gases/odors

Abstract High-performance detection and estimation of gases/odors are challenging, especially in real-time gas sensing applications. Recently, efficient electronic noses (e-noses) being developed using convolutional neural networks (CNNs). Further, CNNs perform better when they operate on a minimal size vector response. In this paper, dimensions the operational vectors have been augmented by virtual sensor responses. These responses obtained from principal components physical Accordingly, two sets data upscaled as one-dimensional one. Another level upscaling is further mirror mosaicking technique. Hence, with our proposed novel approach, final for CNN operations achieves new dimension. With hybrid dataset, consisting responses, simpler has achieved 100 percent correct classification different experimental settings. To best authors information, it first time that an e-nose designed component-based hybrid, dataset considered gases/odors.