Spatiotemporal Distribution of Indoor Particulate Matter Concentration with a Low-Cost Sensor Network
Abstract
Real-time measurement of particulate matter (PM) is important for the maintenance of acceptable air quality. The high cost of conventional instruments typically limits the number of monitoring sites, which in turn undermines the accuracy of real-time mapping of sources and hotspots of air pollutants with sufficient spatial resolution. In this study, a wireless network of low-cost particle sensors that can be deployed indoors was developed. To overcome the well-known limitations of low sensitivity and poor signal quality associated with low-cost sensors, a sliding window and a low pass filter were developed to enhance the signal quality. Utility of the networked system with improved sensitivity was demonstrated by deploying it in a woodworking shop. Data collected by the networked system was utilized to construct spatiotemporal PM concentration distributions using an ordinary Kriging method and an Artificial Neural Network model to elucidate particle generation and ventilation processes.
Recommended Citation
J. Li et al., "Spatiotemporal Distribution of Indoor Particulate Matter Concentration with a Low-Cost Sensor Network," Building and Environment, vol. 127, pp. 138 - 147, Elsevier, Jan 2018.
The definitive version is available at https://doi.org/10.1016/j.buildenv.2017.11.001
Department(s)
Civil, Architectural and Environmental Engineering
Keywords and Phrases
Air quality; Interpolation; Learning systems; Low pass filters; Neural networks; Radio; Sensor networks; Artificial neural network modeling; Conventional instruments; Kriging; Low-cost sensors; Ordinary kriging methods; Real time measurements; Spatial-temporal distribution; Spatiotemporal distributions; Costs; Air quality; Artificial neural network; Concentration (composition); Indoor air; Kriging; Machine learning; Particulate matter; Sensor; Spatial distribution; Temporal distribution; Spatial temporal distribution; Wireless
International Standard Serial Number (ISSN)
0360-1323
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2018 Elsevier, All rights reserved.
Publication Date
01 Jan 2018
Comments
This work was partially supported by Fullgraf Foundation; and McDonnell Academy Global Energy and Environmental Partnership (MAGEEP) at Washington University in St. Louis.