Abstract

This study proposes a real-time air quality forecasting system for smart cities utilizing an edge–cloud Internet of Things architecture combined with filtered sensor data. The system processes environmental data collected from distributed IoT sensors at the edge layer to reduce latency, while leveraging cloud computing resources for advanced predictive modeling. Machine learning algorithms are applied to forecast air quality indices with high temporal resolution. The experimental results demonstrate that the proposed architecture achieves accurate and timely air quality predictions, supporting urban environmental management and public health decision-making.