Sen presented a poster at Ubicomp 2012 this year:
uSmell: A Gas Sensor System to Classify Odors in Natural, Uncontrolled Environments
Sen Hirano, Khai Truong, Gillian Hayes
Smell can be used to infer context about environments. Previous research primarily has shown that gas sensors can be used to discriminate accurately between odors when used in testing chambers. However, potential real-world applications require these sensors to perform an analysis in uncontrolled environments, which can be challenging. In this paper, we discuss our design and evaluation of uSmell—a gas sensor system for sensing smell in ubiquitous computing environments—to address these challenges. Our system samples an odor fingerprint from eight metal oxide semiconductor (MOS) gas sensors every second. It then processes the time series data to extract three features that highlight how time and distance affect the eight MOS gas sensors’ ability to react to the gas molecules released by an odor every five seconds; this generates 24 features in total that are then used to train a decision tree classifier. Using this approach, our system can successfully discriminate a set of odors when placed, both, in a small container with the samples as well as in open air 0.5-2 m from the odor samples. We also demonstrate its ability to classify odors in natural and uncontrolled environments by deploying it in a private bathroom for a week. These results show the potential for applying this sensing towards the development of context-aware systems, such as lifelogging applications or those geared towards enhancing the sustainability of natural resources (e.g., an automatic dual flush toilet that always uses an appropriate amount of water based on the user’s toileting activities).
