TFCED: A novel smart primary user emulation attack detection for cognitive radio sensor networks

Abstract

The integration of dynamic spectrum access within wireless sensor networks is driving a fundamental evolution towards Cognitive Radio Sensor Network (CRSN) architectures, a pivotal component of next-generation communication infrastructures. Nevertheless, vulnerabilities at the physical layer, notably Spectrum Sensing Data Falsification (SSDF) and Primary User Emulation Attacks (PUEA), introduce significant challenges to reliable spectrum utilization in CRSN environments. This study introduces a novel methodology, termed Time–Frequency-Correlation-based Emulator Detection (TFCED), specifically engineered to counteract PUEA-related security breaches within CRSN deployments. TFCED operates by leveraging the inherent temporal and spectral signatures of both legitimate primary users and smart PUEA emitters, thereby achieving enhanced discriminatory capabilities. By refining the precision of PUEA signal identification, the proposed TFCED framework demonstrably extends network operational lifespan. Empirical evaluations reveal that, in comparison to contemporary CRSN security paradigms, TFCED achieves a notable augmentation in sensing accuracy, approximately 9% (95% CI: 0.946%–0.960%, p $<$ 0.0001), and a substantial reduction in error rates, around 23% (95% CI: 0.0929%–0.1071%, p $<$ 0.0001). To corroborate these findings, a practical validation employing a Software Defined Radio (SDR) platform was conducted, substantiating the efficacy of the proposed system architecture.