[论文] FreqSpaNet: Frequency and Spatial Learning of SFPF for Physical Layer ...

论文概要

研究领域: ML
作者: Xiaoxuan Huang, Jinlong Xu, YiZhe Wang, Meng Zhang, Xian Li, Yuying Bian
发布时间: 2026-09-15
arXiv: 2609.17491

中文摘要

未经授权的硬件更换可以在保留无线设备逻辑身份的同时改变其物理实现，这对硬件完整性验证构成了挑战。时空频率极化指纹（SFPF）捕获了设备依赖的跨频率和方向响应，但其频率和空间维度表现出不同的结构依赖性。我们提出 FreqSpaNet，一种用于开放集硬件异常检测的 SFPF 表示学习网络。频率分支捕获相邻频率之间的局部变化，几何感知的空间分支利用角度信息建模方向关系。两种表示通过自适应融合结合，互补的预训练进一步捕获共享信息，同时保留频率和空间表示的独特特征。实验表明，FreqSpaNet 的平均 AUROC 达到 96.31%，比基线高出 9.05 个百分点。在七种硬件更换场景下的结果进一步验证了 FreqSpaNet 的有效性。

原文摘要

Unauthorized hardware replacement can preserve a wireless device's logical identity while altering its physical implementation, posing a challenge to hardware integrity verification. Spatio-frequency polarization fingerprints (SFPFs) capture device-dependent responses across multiple frequencies and directions, but their frequency and spatial dimensions exhibit different structural dependencies. We propose FreqSpaNet, an SFPF representation learning network for open set hardware anomaly detection. A frequency branch captures local variations among neighboring frequencies, while a geometry-aware spatial branch models directional relationships using angular information. The two representations are combined through adaptive fusion, and complementary pretraining further captures shared informa...

自动采集于 2026-09-17

#论文 #arXiv #ML #小凯