Abstract: Air pollution has emerged as a major environmental and public health concern, particularly in urban areas where pollutant concentrations vary significantly across locations and time. Traditional air quality monitoring systems rely on centralized monitoring stations and lack real-time localized sensing and predictive capabilities. To address these limitations, this paper presents the design and implementation of Vento Aureo, an IoT-based air quality monitoring system integrated with machine learning for Air Quality Index (AQI) forecasting. The proposed system utilizes multiple environmental sensors interfaced with an ESP32 microcontroller to measure key air quality parameters such as particulate matter, gaseous pollutants, temperature, and humidity. Sensor data is transmitted to a cloud-hosted backend through REST APIs, where AQI values are computed using standardized pollutant-weighted formulas. Historical data is stored and processed to enable forecasting using time-series machine learning models. A Prophet-based model is employed for short-term AQI prediction, while a hybrid Prophet–LSTM model is implemented to support long-term forecasting.
A web-based dashboard is developed to visualize real-time air quality data, historical trends, and predicted AQI levels in an intuitive manner. Experimental evaluation demonstrates that the system effectively provides localized air quality monitoring along with meaningful AQI forecasts. The proposed solution offers a scalable and practical approach for environmental monitoring, public awareness, and smart city applications.

Keywords: Internet of Things (IoT), Air Quality Index (AQI), Air Pollution Monitoring, Machine Learning, Time-Series Forecasting, Prophet Model, LSTM, Cloud Computing, Real-Time Monitoring, Smart Cities.

Downloads: | DOI: 10.17148/IARJSET.2025.121232