An Integrated FL--DRL Framework for Adaptive Eavesdropping Mitigation in Internet of Drones Networks
Abstract
This dataset supports the study “An Integrated FL–DRL Framework for Adaptive Eavesdropping Mitigation in Internet of Drones Networks.” It contains simulated wireless-channel and mobility observations for UAV swarms operating in mixed urban, suburban, and rural environments under normal, eavesdropping, and jamming conditions. Each sample includes channel-related features such as CSI amplitude and phase, signal-to-noise ratio, Doppler shift, and coherence time, together with scenario labels and parameters required to reproduce the anomaly-detection and adaptive-control experiments. The dataset was designed to evaluate privacy-preserving federated anomaly detection based on a Conv1D–LSTM autoencoder and a hybrid deep reinforcement learning controller that combines Double DQN for transmit-power selection and PPO for trajectory adaptation. It also supports analysis of non-IID data distributions across UAVs, secrecy performance, latency, energy consumption, privacy–utility trade-offs, and robustness against adaptive adversaries.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.