An Investigation of Federated Learning for Air Quality Forecasting and Monitoring

On the Impact of Data Heterogeneity in Federated Learning: A Case Study on Air Quality Prediction

Federated learning has become a popular model to apply in privacy-preserving modeling in distributed settings, particularly when the models are applied to data that is distributed among various locations and is often sensitive, such as in air quality prediction. This paper looks into how effective federated learning is for predicting ozone (O₃) concentrations, under both independent and non-independent data distributions. In particular, two representative algorithms Federated Averaging (FedAvg)and Federated Averaging (FedAvg) are experimented on a real-world air quality dataset. An experimental framework was established that was relatively comprehensive and centralized training and local-only models were used as baselines. Mean absolute error (MAE) and root mean squared error (RMSE) are used to measure model performance. The findings suggest that federated learning performs much better than the isolated local models and the performance is similar to that of the centralized training. Having said that, data heterogeneity does present certain issues-it slows down convergence and decreases accuracy in prediction. In such non-IID conditions, FedProx is more stable and less erroneous than FedAvg implying that it is more resistant to client drift.Most importantly, this paper provides empirical data on the impact of data heterogeneity on federated learning, and proves that federated learning can be a viable alternative in privacy-sensitive environmental prediction problems.

Implementation and Validation of an Air Quality Prediction Model Based on Federated Learning

Ground ozone and other air pollutants pose a great threat to the health. Because the data of different monitoring stations are scattered, it is very complicated to accurately predict the ozone level. Federated Learning (FL) allows everyone to train models together without exchanging raw data, thus protecting privacy. This paper uses FL to predict air quality, and the key point is to establish a prediction model, not to change the algorithm. This paper used UCI air quality data set (with 8,762 records, each with four attributes) to simulate the situation of five users, and the data were separated by independent identically distributed (IID) and non-independent identically distributed (Non-IID). This paper trained 50 communication rounds with a multi-layer perceptron (MLP) and Federal Average (FedAvg) method. The model of centralized training and local training is also used to compare. The experimental results show that the average absolute error (MAE) of FedAvg is 48.9 under IID data and 58.3 under Non-IID data, which is much better than the local training models (MAE 78.5 and 85.2) and close to the effect of centralized training (MAE 42.3 and 51.6). This method of federated learning adapts well to different data. This study confirms that FL is a feasible method for distributed air quality prediction.

Air Quality Monitoring in Sparse Urban Networks: An Enhanced Fog Hierarchical Federated Learning Approach

Adaptive Multi-Task Learning for Urban Air Quality Assessment

Accurate air pollution forecasting is vital for protecting the environment and public health. However, predicting air pollution continues to be a challenge due to complicated relationships between meteorological variables, pollutant concentrations, and time-dependent characteristics. This paper proposes a Temporal-Aware Multi-Task Learning (TMTL-AQI) framework to assess urban air quality via structured data. The model simultaneously performs three tasks: Air Quality Index category classification, PM2.5 and PM10 particulate matter regression, and auxiliary AQI value prediction. The framework consists of data preprocessing, cyclical temporal encoding, a shared backbone based on hard parameter sharing, and three task-specific output heads optimized jointly. It uses cyclical transformations to encode temporal traits that capture periodic patterns of the environment and produces a common neural representation through hard parameter sharing. The TRAQID dataset was used for empirical testing. The model outperforms single-task and baseline multi-task models with an accuracy of 0.7605 and an F1-score of 0.7422 for AQI classification. Although based solely on structured input (no image-based features), the model performed competitively with a state-of-the-art approach due to the reduced prediction error in regression tasks (MAE values of 13.34 μg/m³ for PM2.5 and 22.06 μg/m³ for PM10).

Intelligent Air Quality Monitoring and Management System for Smart Cities

Urban air pollution has emerged as one of the foremost environmental and public health crises confronting rapidly growing cities worldwide. The convergence of industrial expansion, vehicular traffic, and urban sprawl continues to degrade ambient air quality at an alarming pace. Conventional monitoring setups, which rely on fixed sensor nodes and dedicated stations, incur substantial deployment costs and inherently limit spatial coverage. This paper introduces an Intelligent Air Quality Monitoring and Management System (IAQMMS) tailored for smart city ecosystems—a software-centric framework that fuses real-time pollutant data streams, machine learning inference, and predictive analytics to monitor, forecast, and govern urban air pollution. The system exploits publicly accessible air quality application programming interfaces (APIs) and historical pollution repositories to generate forward-looking Air Quality Index (AQI) estimates and pinpoint probable emission hotspots. A hybrid Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) architecture is adopted to simultaneously capture spatial pollutant distributions and temporal concentration dynamics. Supplementary capabilities encompass automated health risk stratification and context-aware precautionary guidance delivered through an interactive dashboard. Experimental results confirm competitive prediction accuracy, real-time responsiveness, and seamless scalability, affirming the system's suitability for next-generation smart city deployments.

An Intelligent Prediction Model for Air Quality Monitoring Using GA-ELM

Abstract— Air pollution poses one of the most serious environmental threats to public health and sustainable urban living. Accurate and timely forecasting of the Air Quality Index (AQI) can support policymakers, environmental agencies, and citizens in taking proactive measures to mitigate pollution effects. This project presents an optimized machine learning-based Air Quality Forecasting System that integrates Extreme Learning Machines (ELM) and Genetic Algorithms (GA) to predict short-term variations in air quality. The ELM component provides a high-speed, single-hidden layer feedforward neural network capable of mapping complex non-linear relationships between pollutants. To overcome the traditional challenge of random weight initialization in ELMs, a Genetic Algorithm (GA) is implemented to "evolve" and select the most optimal input weights and biases. This hybrid approach ensures that the model captures intricate correlations among multiple pollutants and meteorological variables with significantly higher computational efficiency than iterative deep learning models. A comprehensive dataset containing historical pollutant concentrations (PM2.5, PM10, NO₂, SO₂, CO, NH₃, O₃) and meteorological parameters (temperature, humidity, wind speed, etc.) is collected and preprocessed to remove noise, handle missing values, and normalize features using Min-Max scaling. The hybrid GA-ELM model is trained and evaluated using key performance metrics such as Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Experimental results demonstrate that the proposed architecture achieves superior prediction speed and stability compared to traditional regression models and standalone neural networks. In addition to the model, the system includes an interactive Flask-based web dashboard for real-time AQI visualization and forecasting. The dashboard provides 1–12 hour ahead predictions, dynamic charts, and color-coded AQI indicators based on national standards for easy interpretation.Overall, the developed system demonstrates a robust, scalable, and practical solution for short-term air quality prediction. .