Study On Cooperative Spectrum Sensing Algorithm In Cognitive Radio

With the rapid development of wireless communication technologies, the shortage of wireless spectrum resource mainly caused by fixed spectrum allocation policy is growing sharply. In order to improve the spectrum utilization rate and to solve the spectrum scarcity problem essentially, Joseph Mitola proposed the concept of cognitive radio, whose core idea is that vacant licensed bands can be reused without interfering licensed users’communication. Cognitive radio technique can dynamically use wireless licensed bands not fully utilized in time and space, and provide available spectrum resource for new radio communication services. Spectrum sensing is one of the key technologies in cognitive radio which is deemed to the basis and premise because its primary task is to detect vacant bands in wireless environment periodically. Therefore, the research on spectrum sensing has important theoretical significance and practical value. This paper applied the generalized likelihood ratio criterion based on power method to the problem of cooperative spectrum sensing, and proposed the generalized likelihood ratio based on power method spectrum sensing algorithm. The main duties of this paper are as follows:(1) The research background and significance of spectrum sensing are elaborated. Research status at home and abroad is summarized. Single node spectrum sensing methods and cooperative spectrum sensing rules are discussed.(2) Energy detection (ED) algorithm is susceptible to noise uncertainty, frequent burst interference and and other factors in the complex wireless communication environment. Aiming at this problem, frequency domain GLRT (short for F-D-GLRT) spectrum sensing algorithm is studied. According to likelihood ratio criterion, this algorithm produces a test statistic irrelevant to noise power to accomplish spectrum sensing process. Simulation results indicated that this algorithm is superior to ED and effectively overcomes noise uncertainty.(3) Aiming at the problem of past spectrum sensing algorithms with high computational complexity, a GLRT spectrum sensing algorithm based on power method (short for P-GLRT) is proposed. This algorithm firtly utilizes maximum likelihood criterion to estimate unknown parameters solved by power method and matrix trace, and then according to likelihood criterion, produces a test statistic irrelevant to noise power to accomplish spectrum sensing process. This algorithm only utilizes several iterations to achieve the principal eigenvalue and corresponding eigenvector. The presented algorithm has low computational complexity due to avoiding the eigenvalue decomposition process, can effectively resist noise uncertainty and possesses better sensing performance.