Search papers, labs, and topics across Lattice.
This paper introduces SecureDrive-FL, a federated learning framework that integrates Gradient-Aware Selective Homomorphic Encryption (GASHE) with differentially private stochastic gradient descent (DP-SGD) to enhance privacy in driver monitoring applications. By selectively encrypting only the gradient components that exceed a sensitivity threshold, the framework effectively mitigates risks from Man-in-the-Middle attacks and model poisoning while maintaining comparable accuracy to DP-SGD alone. The evaluation shows that SecureDrive-FL achieves robust performance against adversarial threats with only a modest increase in runtime overhead, demonstrating a significant advancement in privacy-preserving machine learning techniques.
SecureDrive-FL withstands Man-in-the-Middle attacks without sacrificing accuracy, achieving a remarkable balance of privacy and performance in federated learning.
Federated Learning (FL) enables privacy-aware distributed training, yet gradient updates remain exploitable: Man-in-the-Middle (MitM) interception exposes updates in transit, while model poisoning corrupts global convergence. We first introduce GASHE (Gradient-Aware Selective Homomorphic Encryption), a novel selective encryption strategy that dynamically identifies and encrypts only the gradient components exceeding a DP-calibrated sensitivity threshold, rather than encrypting all parameters uniformly as in static layer-based or full-parameter CKKS schemes. Building on GASHE, we introduce SecureDrive-FL, a federated driver monitoring framework that couples DP-SGD with GASHE to create the first closed-loop DP+HE privacy pipeline: DP-SGD calibration parameters directly derive the GASHE encryption mask, unifying training-time privacy and communication-time confidentiality. Evaluated on a ten-class distracted driver classification task under non-IID federated splits, SecureDrive-FL matches DP-SGD alone's poisoning resistance (73.6% vs. 74.0% accuracy, 3.9% Attack Success Rate for both) while additionally withstanding MitM interception, where DP-SGD alone collapses to near-random accuracy (78.2% vs. 10.4%), all under only approx. 8--10% additional runtime overhead relative to DP-SGD alone---under DP-SGD noise injection with per-round privacy parameter epsilon_0=4.
