The MDHC Detection Scheme for Full-Duplex Enabled Spectrum Sharing System

Qifan Fu, Danyang Wang, Peihan Qi, Zan Li, Jian You

Abstract: Spectrum sharing, as a key concept in cognitive radio (CR), is considered to be a promising technique in mitigating the under-use of spectrum resources. Compared with the traditional half-duplex cognitive nodes (HD-CNs), full-duplex cognitive nodes (FD-CNs) can achieve continuous sensing and realize much higher efficient spectrum utilization, which greatly reduces the possibility of adverse impact on the primary user (PU). However, the spectrum sharing with FD-CNs is challenged by self-interference and the practical non-Gaussian transmitted signals (e.g. the digital modulated signals), which makes the identification of PU's working states more difficult. In this paper, we propose a spectrum sensing scheme namely multidimensional high-order cumulant (MDHC) detection method. Two cases are studied based on whether the secondary user (SU) transmits or not. Interestingly, the test statistics derived by posterior probability ratio are the same at both cases, while the decision thresholds of these two cases are distinct from each other. Moreover, the closed-form solution of the decision threshold for each case is obtained by the Neyman-Pearson (NP) criterion. Rich spectrum information can be got via calculating the high order cumulants of the received signals. The proposed method can also extract a non-Gaussian signal from Gaussian noise even when the noise is colored and eliminate the negative effect of the noise power uncertainty. Finally, simulation results are provided to evaluate the proposed schemes.