Federated Learning in the Sky: Aerial-Ground Air Quality Sensing Framework With UAV Swarms

Due to air quality significantly affects human health, it is becoming increasingly important accurately and timely predict the Air Quality Index (AQI). To this end, paper proposes a new federated learning-based aerial-ground sensing framework for fine-grained 3D monitoring forecasting. Specifically, in air, leverages light-weight Dense-MobileNet model achieve energy-efficient end-to-end learning from haze features of images taken by Unmanned Aerial Vehicles (UAVs) predicting AQI scale distribution. Furthermore, Federated Learning Framework not only allows various organizations or institutions collaboratively learn well-trained global monitor without compromising privacy, but also expands scope UAV swarms monitoring. For ground systems, we propose Graph Convolutional neural network-based Long Short-Term Memory (GC-LSTM) accurate, real-time future inference. The GC-LSTM utilizes topological structure station capture spatio-temporal correlation historical observation data, which helps system accurate Through extensive case studies on real-world dataset, numerical results show that proposed can privacy raw data.