IoT-Enabled Real Time Air Quality Monitoring Using Hybrid CNN-RNN with Attention Mechanism

Monitoring the environment, particularly air quality, is vital for protecting public health, ensuring safety, and effectively reducing pollution levels. However, using physical models often leads to incorrect and imprecise outcomes. To address this, the study proposes an IoT-based system for real-time monitoring of air quality using sensors. Data undergoes through preprocessing, including cleaning, noise reduction, and missing value imputation. A hybrid CNN-RNN model with attention mechanism is used to capture temporal dependencies and predict pollution levels. To optimize performance, Harris Hawks Optimization (HHO) algorithm fine-tunes the model parameters. Proposed system implemented using Python and achieves MAE of 0.43 and RMSE of 0.53, demonstrating significant improvements in prediction accuracy and timely pollution alerts. This system enhances urban air quality monitoring and supports informed environmental management for sustainable urban living.