Research On Cooperative Spectrum Sensing Algorithm Based On Full Duplex

As the existing wireless spectrum resources can not satisfy the growing demand for data transmission bandwidth, people have to develop new technologies to improve the frequency spectrum utilization. Full duplex communication has been proved to be an effective method to improve the utilization of frequency band. It can almost double the frequency band utilization of the existing spectrum, which is very significant.Compared to the traditional half duplex communication, the full duplex communication has characteristics as follows: to send and receive at the same time. The limit of the current utilization of frequency band is getting closer to that of Shannon's prediction,which forces people to find new ways to solve the problem. The spectrum sensing network is a solution to solve the problem of the frequency bands. Spectrum sensing is that secondary (unlicensed) users access licensed bands when the primary (licensed)users (PU) are not present, thus spectrum efficiency can be improved without interfering PU to enhance the network performance.This thesis firstly introduces the basic principle of the full duplex wireless communication system. While in the implementation of the full duplex wireless communication system the biggest difficulty is self-interference. Since full duplex is the method that the transceiver simultaneously transmits and receives, the power of the system is also the interference of the receiving antenna of the system itself compared to the half-duplex communication system. The greater the transmitting power is, the greater the self-interference is, which makes full duplex communication difficult to achieve. Many scholars have studied the self-interference cancellation technique.However, in their studies, they use self-interference channel simulation system of Rayleigh channel. In this thesis we use Nakagami-m channel since it can simulate self-interference channel better. This thesis derives the probability of false alarm, detection probability and a closed form solution of the throughput of the system, also verifies the correctness of the closed form solution through the simulation.Secondly, in order to further improve the accuracy of the system, this thesis introduces the cooperative spectrum sensing in the original non-slotted full duplex communication system. Cooperative spectrum sensing refers to multiple secondary users sense the primary user's working state, and then delivery their perception to the fusion center. Finally, the fusion center determines the primary user's working state according to the fusion rule, which is more accurate. In this system, this thesis also proposes a double threshold collaborative decision algorithm to reduce the interference between secondary users.Then, a novel cooperative sensing scheme based on phase difference is discussed.Compared to the conventional energy detection, the phase difference of perception algorithm does not need to know the transmit power of the primary user and the noise power in advance. More importantly, energy detection suffers great performance degradation in low SNR.Finally, based on the previous two thresholds this article introduces of multi threshold cooperative decision algorithm. Intuitively, each user has multiple threshold values, the decision contains more information to help make the correct judgment. With larger amount of information, this thesis proposes a weighted cooperative algorithm based on the Bayesian algorithm and the weighted cooperative algorithm relatively. In this thesis, we obtain a closed form solution of the probability of detection and false alarm. The simulation has proved that the cooperative algorithm proposed in this thesis outperforms the traditional single threshold collaborative decision algorithm. The more threshold number is, the better the performance is.