Detecting contaminants in smart buildings by exploiting temporal and spatial correlation

Abstract

Monitoring the indoor air quality is one of the most critical activities within a smart building environment. The introduction of contaminant sources inside the building envelope can compromise the air quality and possibly endanger the lives of the inhabitants. In this paper, a new contaminant detection system is proposed for the prompt and effective detection (and isolation) of contaminant sources. Specifically, we address the challenging scenario where the contaminant of interest is also naturally present in the indoor building environment (e.g. CO2). A key feature of the proposed system is that it does not require a model of the contaminant propagation, but relies instead in its ability to exploit the temporal and spatial relationships present in the data streams acquired by the sensors deployed within the smart building. The effectiveness of the proposed system has been evaluated on a reference test bed.