An IoT-Based Air Quality Monitoring System: Evaluating and Improving Low-Cost Particulate Matter Sensors

Air quality monitoring in densely populated areas is attracting increasing public interest, leading to a rise in demand for reliable, low-cost IoT monitoring systems capable of supplementing expensive reference stations to achieve greater spatial coverage of an area. This research work presents an in-depth comparative evaluation of a custom-built, low-cost IoT node for air quality monitoring, which incorporates two widely used optical particle sensors: the Plantower PMS5003 and the Sensirion SPS30, also include a temperature and humidity sensors. Unlike commercially available devices that rely on proprietary “black-box” calibration algorithms, this study conducts an extensive analysis aimed at recording and analyzing the raw output data from the sensors (PM2.5 and PM10) in order to evaluate their basic performance in comparison with reference measurements. In addition, a transparent correction model was developed that takes into account the ambient temperature and relative humidity—measured by the temperature and humidity sensors, respectively—with the aim of minimizing environmental cross-sensitivities. The experimental results demonstrate that the SPS30 exhibits higher measurement responsiveness and temporal stability compared to the PMS5003. By providing an open calibration methodology and a direct comparative evaluation of the performance of both sensors, this study offers practical insights for the development of accurate, scalable, and low-cost air quality monitoring networks via the IoT.