Review on Prediction of Air Pollution in Specific City

Air pollution has become a serious environmental and public health issue in urban areas, driven by factors such as rapid urbanization, increased vehicular traffic, and industrial activities. Predicting air pollution in a specific city is challenging due to the dynamic and interrelated nature of influencing factors, including weather conditions, traffic patterns, seasonal variations, and geographical characteristics. In addition, issues such as incomplete or noisy data, limited real-time availability, and the difficulty of selecting suitable prediction models reduce forecasting accuracy. These challenges highlight the need for an efficient, localized, and real-time air pollution monitoring and prediction system. To address these issues, this system proposes a low-cost air pollution prediction system using an ESP32 microcontroller integrated with MQ-135, DHT11, and PM2.0 sensors. The system continuously monitors harmful gases, particulate matter, temperature, and humidity, and transmits the collected data wirelessly for further processing. Machine learning techniques are applied to analyze historical and real-time data to predict future air quality levels and classify them according to standard AQI categories. The predicted results are displayed on a web or mobile dashboard, with alerts generated when pollution exceeds safe limits. This system supports early warning, effective air quality management, and smart city development while improving public awareness and public health protection.