FAS-Enabled Anti-SSDF Cooperative Spectrum Sensing: A Novel Physical-Layer Security Framework
摘要
Cognitive radio (CR) has emerged as a promising solution to meet the spectrum demands of massive sensor networks; however, cooperative spectrum sensing (CSS) remains highly vulnerable to spectrum sensing data falsification (SSDF) attacks. In this paper, we investigate the physical-layer security of CSS from a fluid antenna system (FAS) perspective and propose a robust anti-SSDF framework by exploiting spatial adaptability. A cognitive wireless radio network (CWRN) with FAS-enabled secondary users is formulated, where each sensor dynamically selects antenna positions over a continuous aperture, inducing spatial decorrelation and channel randomness. Based on these properties, a spatial-consistency-based defense mechanism and a FAS-assisted secure fusion strategy are developed to effectively suppress coordinated and intelligent SSDF attacks. To further enhance security, we propose a novel FAS-Machine Learning hybrid scheme that integrates deep learning with FAS spatial diversity for adaptive attack detection. Simulation results demonstrate that the proposed scheme significantly enhances sensing accuracy and robustness under various attack scenarios and fusion rules, while maintaining low hardware complexity, highlighting the effectiveness of FAS as a physical-layer security enabler for CSS. The proposed framework achieves up to 25% higher detection accuracy compared to existing methods under cooperative SSDF attacks, with only 15% additional computational overhead.
类型
出版物
2026 6th International Conference on Computer Communication and Artificial
Intelligence (CCAI)