Covariance and eigenvalue based spectrum sensing using USRP in real environment
C I Muhammed Althaf, S. Chris Prema · 2018
Cognitive radio are useful for secondary users to sense spectral environment and opportunistically access licensed spectrum band when unutilized. In this paper, we present an experimental study in real environment using USRP (Universal Software defined Radio Peripheral) to evaluate the detection performance of covariance and Eigenvalue based spectrum sensing techniques for detection of primary users. The detection performance was analyzed with minimum Eigenvalue detection (EME), maximum to minimum Eigenvalue detection (MME), absolute covariance detection (CAV) and energy detection (ED). The detection performance was compared in terms of detection probability (Pd), probability of false alarm (Pfa) and SNR (Signal to Noise Ratio). The maximum to minimum Eigenvalue detection (MME) method outperformed the other methods as shown in theoretical results. The detection performance were analyzed and presented for low SNR value upto -25 dB.