PrivFly: A Privacy-Preserving Self-Supervised Framework for Rare Attack Detection in IoFT

TL;DR

PrivFly tackles rare-attack detection in IoFT by unifying self-supervised representation learning with differential privacy. It employs VIME-based pretraining to learn robust tabular representations, and DP-SGD to protect training data, augmented by CTGAN/SMOTE oversampling to address severe class imbalance. On the ECU-IoFT dataset, PrivFly achieves up to 99% F1-score under moderate privacy, with strong performance across majority and minority classes, and SHAP analysis provides insight into feature importance and privacy-induced attribution shifts. The framework offers a practical balance between privacy and detection performance for secure IoFT systems, with future directions toward distributed learning and adversarial robustness.

Abstract

The Internet of Flying Things (IoFT) plays a vital role in modern applications such as aerial surveillance and smart mobility. However, it remains highly vulnerable to cyberattacks that threaten the confidentiality, integrity, and availability of sensitive data. Developing effective intrusion detection systems (IDS) for IoFT networks faces key challenges, including data imbalance, privacy concerns, and the limited capability of traditional models to detect rare but potentially damaging cyber threats. In this work, we propose PrivFly, a privacy-preserving IDS framework that integrates self-supervised representation learning and differential privacy (DP) to enhance detection performance in imbalanced IoFT network traffic. We propose a masked feature reconstruction module for self-supervised pretraining, improving feature representations and boosting rare-class detection. Differential privacy is applied during training to protect sensitive information without significantly compromising model performance. In addition, we conduct a SHapley additive explanations (SHAP)-based analysis to evaluate the impact of DP on feature importance and model behavior. Experimental results on the ECU-IoFT dataset show that PrivFly achieves up to 98% accuracy and 99% F1-score, effectively balancing privacy and detection performance for secure IoFT systems.

Figures (6)

• Figure 1: System model.
• Figure 2: The architecture of the proposed PrivFly intrusion detection framework along with the VIME self-supervised representation learning module.
• Figure 3: Performance comparison of classifiers based on Accuracy and F1-Score.
• Figure 4: Performance evaluation and SHAP-based explainability analysis of the DNN model with and without differential privacy (DP).
• Figure 5: Trade-Off between privacy budget and model performance.