Cooperative Spectrum Sensing In Cognitive Radio Networks

Cognitive radio (CR), an innovative technology to mitigate the spectrum scarcity problem, is being paid on more and more attention as a key technology in the next-generation wireless network. And spectrum sensing, one of the most important issues in CR networks, has become a hot research topic in recent years. Spectrum sensing is the procedure of the CR user detecting weak primary user signals in possibly unknown types. If unoccupied spectrum band is found, the CR user can use the corresponding band. Spectrum sensing should also monitor the activation of the primary users so that the CR user immediately vacates the occupied band. Cooperative spectrum sensing can improve the inference accuracy and sensitivity using the location diversity of CR users.The thesis summarizes existing cooperative spectrum sensing schemes, studies the reliability and security problems of the cooperative spectrum sensing, and proposes a reputation-based cooperative spectrum sensing scheme in CR networks. In the proposed scheme, the energy detection technology is adopted as the basic sensing block, and the reputation degree is used to measure the each CR user's performance during cooperation and it is also used to weight the sensing result from the corresponding CR user in the linear fusion process. With such a feedback process, the impact of unreliable and malicious CR users on the detection performance is effectively alleviated. With simulations, we analyzed and evaluated the performance of the reputation-based cooperative spectrum sensing scheme.The wideband spectrum sensing scheme is also investigated in the thesis, and a group-based cooperation wideband spectrum sensing model is constructed. In our proposed model, with the objective to maximize throughput and alleviate shadowing effect, a grouping algorithm combined with the user selection algorithm is proposed. Moreover, to reduce the communication overhead, we proposed an improved cooperative spectrum sensing scheme without degrading the detection performance. Finally, we validated the proposed scheme with various parameters, such as the number of CR users, the sub-band prior information, and the ratio of shadowed CR users, and so on.