Calibration of NO, SO2, and PM using Airify: A low-cost sensor cluster for air quality monitoring

During the past decade, substantial efforts have been dedicated to advancing cost-effective sensor platforms for air quality monitoring. Calibration is a crucial step in ensuring that the data quality of low-cost air monitoring systems meets established standards. Recent research has extensively evaluated low-cost air monitoring platforms against the data quality objectives set by the European Directive. This paper introduces a novel calibration model for low-cost air quality sensors, significantly improving the accuracy of the measurement of nitric oxide (NO), sulfur dioxide (SO 2 ), and particulate matter (PM 1 , PM 2.5 , and PM 10 ), while promoting accessibility and adaptability in environmental monitoring technologies. This study extends the evaluation of a developed platform capable of integrating a diverse array of sensors to measure up to 12 parameters. Our proposed models demonstrate a significant improvement, achieving a 60% better accuracy for SO 2 . Additionally, these models deliver similar results for PM x and NO or exceed those of state of the art research. The calibration methodology meets the requirements of the Data Quality Objectives (DQO) for all monitored parameters and also achieves indicative levels for PM parameters. • The proposed calibration models enhance accuracy for NO, SO 2 , and PM parameters. • Both MLR and RF models achieve indicative levels for PM. • Only RF calibration model is effective for NO and SO 2 . • RF model reached the DQO estimation level faster than MLR for all parameters.