Year: 2025
Type: article
Abstract: Air pollution is a major environmental issue affecting public health and urban sustainability. Traditional air quality monitoring systems are costly, have limited coverage, and lack predictive capabilities. This paper proposes an IoT-enabled Air Quality Monitoring and Predicting System consisting of MQ-series gas sensors, Raspberry Pi, Random Forest machine learning algorithm, and cloud computing for real-time monitoring and AQI prediction.The system uses MQ-2, MQ-4, MQ-7, MQ-9, and MQ-135 sensors in conjunction with a DHT-11 temperature and humidity sensor, connected to a Raspberry Pi via an MCP3208 ADC. The sensor data is processed using the Random Forest algorithm to provide predictions for AQI. The data is streamed over the ThingSpeak IoT cloud, graphed, and made accessible via a mobile application. Users receive real-time alerts when air quality is poor.Outcomes show that the system gives a sensor accuracy of 92-95 percent and AQI forecast accuracy of 85-90 percent. The cloud-to-mobile latency is approximately three seconds, which means almost instantaneous updates. Compared to conventional monitoring mediums, the system is cost-effective, scalable, efficient, and suitable for use in smart cities and industries.The study concludes that the convergence of IoT, machine learning, and cloud computing makes real-time air quality monitoring and forecasting possible. Some future enhancements include deep learning, edge AI, and increased sensor coverage.
Authors: Kadagala Venkatesh, Konijeti Jaswanth, Inavoli Preethi, Jujjavarapu Ganesh, Mandapati Thirupathamma
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4409195304