Low-cost sensor networks, data science and forecasting
Many cheap sensors, carefully corrected, plus models that predict tomorrow’s air
How the work is done
Co-location calibration against reference monitors. Humidity and temperature correction. Machine learning and deep learning time-series models, tested on held-out time periods with several random seeds, with the uncertainty reported. Interpretation of model drivers, so each prediction can be explained.
Publications 9
-
Interpretable machine learning tools to analyze PM2.5 sensor network data so as to quantify local source impacts and long-range transport
Atmospheric Research, 311, 107656
Low-cost sensor networks, data science and forecasting Source apportionment and emission sources Urban and indoor air quality in Dhaka
-
Characterizing Indoor-Outdoor PM2.5 Concentrations Using Low-Cost Sensor Measurements in Residential Homes in Dhaka, Bangladesh
American Chemical Society (ACS)
Low-cost sensor networks, data science and forecasting Urban and indoor air quality in Dhaka
10.26434/chemrxiv-2024-7wrlw
Open access (green) -
Particulate Matter Emissions at Different Microenvironments Using Low-Cost Sensors in Megacity Dhaka, Bangladesh
Atmosphere, 15(8), 897
Low-cost sensor networks, data science and forecasting Source apportionment and emission sources Urban and indoor air quality in Dhaka
10.3390/atmos15080897
Open access (gold) -
Research Priorities of Applying Low-Cost PM2.5 Sensors in Southeast Asian Countries
International Journal of Environmental Research and Public Health, 19(3), 1522
Low-cost sensor networks, data science and forecasting Urban and indoor air quality in Dhaka
10.3390/ijerph19031522
Open access (gold) -
Distinguishing Air Pollution Due to Stagnation, Local Emissions, and Long-Range Transport Using a Generalized Additive Model to Analyze Hourly Monitoring Data
ACS Earth and Space Chemistry, 5(9), 2329-2340
Low-cost sensor networks, data science and forecasting Source apportionment and emission sources Urban and indoor air quality in Dhaka
-
Global Sources of Fine Particulate Matter: Interpretation of PM<sub>2.5</sub> Chemical Composition Observed by SPARTAN using a Global Chemical Transport Model
Environmental Science & Technology
Aerosol optical properties and remote sensing Low-cost sensor networks, data science and forecasting Urban and indoor air quality in Dhaka
Theses 2
-
Retrospective Analysis of PM2.5 Concentrations in Dhaka Using 24 Deep Learning Models and Predictive Deployment
Supervisor: Prof. Dr. Abdus Salam
Co-supervisor: Prof. Dr. Md. Safiqul IslamLow-cost sensor networks, data science and forecasting Urban and indoor air quality in Dhaka
-
Spatial and Temporal Variations of PM2.5 Concentrations across Dhaka Using TSI BlueSkyTM Sensors
Supervisor: Prof. Dr. Abdus Salam
Co-supervisor: Shatabdi RoyLow-cost sensor networks, data science and forecasting Urban and indoor air quality in Dhaka