Primary signal detection algorithms for spectrum sensing at low SNR over fading channels in cognitive radio

摘要

The generalized detector (GD) can be implemented at the low signal-to-noise ratio (SNR) in cognitive radio (CR) systems to improve the spectrum sensing performance under correlated antenna array elements. The weighted GD (WGD) and the generalized likelihood ratio test for GD (GLRT-GD) are proposed to be used for coarse spectrum sensing when the noise power is known and unknown, respectively. The GD optimal detection threshold is defined based on the minimum probability of error criterion for various fading channels, namely, the additive white Gaussian noise (AWGN), Nakagami-m, and Rayleigh fading channels. The performance of the proposed algorithms are compared with the spectrum sensing performance of the energy detector (ED), weighted ED (WED), maximum-minimum eigenvalue (MME) detector, generalized likelihood ratio test for ED (GLRT-ED), matched filter (MF), arithmetic to geometric mean (AGM) detector, scaled largest eigenvalue (SLE) detector, moment based detector (MBD), covariance based detector (CBD), and others. The simulation results demonstrate superiority in the spectrum sensing performance of the proposed algorithms in comparison with the above-mentioned detectors. For example, the GLRT-GD achieves the SNR gain equal to 1.2 dB, 4.0 dB, and 4.5 dB in comparison with GLRT-ED, MME, and GM detectors, respectively, at the probability of false alarm P-FA = 0.1. The WGD and GLRT-GD implementation allows us to achieve a considerable spectrum sensing performance improvement at small number of samples under the low SNR and the correlated antenna array elements. (C) 2019 Elsevier Inc. All rights reserved.