A Practical IoT-Based PM2.5 Air Contaminant Tracking Mechanism with Suitable Machine Learning Support

The air quality of any environment is of crucial concern. With developments in (CPS) cyber-physical systems and data analytics, air contaminants can be tracked employing mechanisms of Internet of Things (IoT) and machine learning (ML). Therefore, this paper proposes a practical IoT-based PM 2.5 air contaminant tracking mechanism with suitable machine learning support. With the suggested mechanism, readings on particulate matter concentration can be tracked and revealed in real times. The proposed mechanism applies benefit of IoT-enabled gas sensors and a decision tree ML-model to track and analyze fine particle (PM2.5) levels with time trends. The IoT nodes incorporates MQ-135 and DHT-11 sensors with ESP32 chip. The nodes exploit power-saving modes of the chip to assure optimal duty cycling and sustained operation. The tracked data amassed at the devices are sent to a gateway that finally forward the data to a server for recovery at the user end. Ultimately, the analysis result recommends that the presented IoT-based tracking mechanism can support practical air quality tracking and forecasting in various pollutant concentrated atmospheres while guaranteeing minimal power consummation at the devices. The proposed scheme can be adopted in diverse pollution concentrated zones to provide sustained real-time information on PM2.5 concentrations.