Research On Spectrum Sensing And Relay Selection In Full-Duplex Cognitive Radio Relay Networks

With the rapid development of wireless communication technology and the wide application of wireless communication system, the contradiction between the demand of frequency spectrum and resource scarcity has become the bottleneck of emerging wireless communication's development, such as the mobile Internet, Internet of things, D2 D communication, M2 M communication. On the one hand, the industry focuses on millimeter waves, visible light frequency band to find more spectrum resources. Due to the restriction of propagation characteristics of high frequency radio waves, it is difficult to meet the requirements of wide coverage. On the other hand, the industry is committed to improve spectrum efficiency. Full duplex(FD) communication can transmit and receive in the same area at the same time, and double the spectrum efficiency in theory. In cognitive radio system, primary users(PU) share spectrum resources with secondary users(SU), which also can improve spectrum efficiency to some extent.Taking full-duplex communication and cognitive radio technology into wireless relay networks can provide more access opportunities for wireless relay networks. Because of the full-duplex operation, secondary users can sense the activity of primary users while communicating in authorized spectrum. Meanwhile, SUs can terminate transmission when primary users become active such that the interference to PUs can be reduced. This paper investigates the spectrum sensing algorithm and the relay selection strategy in full-duplex cognitive wireless relay networks, the main research contents are as follows:1)Spectrum sensing algorithms in full-duplex cognitive wireless relay networks: At first, a system model of cognitive wireless relay network is given out, then the transmission model of cognitive wireless relay networks is established by combining the ON/OFF model of PU with the HD/FD frame structure. The classical energy detection is adopted for spectrum sensing. Detection and false alarm probabilities are analyzed in the presence of residual self-interference. The outage probability and average collision time of cognitive users in HD and FD modes are derived, respectively. Numerical results show the detection performance of the considered spectrum sensing algorithm as well as the outage probability and average collision time versus signal-to-noise ratio and transmission period.2)Relay selection strategy of full-duplex cognitive wireless relay networks: The system model of cognitive wireless relay networks is presented at first. Then, the relay selection strategies with equal power allocation are investigated, including best relay selection, max-min relay selection, best harmonic mean relay selection and hybrid duplex relay selection. Elements and outage probability of relay selection strategy with optimal power allocation are studied as well. Analysis and comparison of the proposed selection strategies are shown in numerical simulations.