Research On The Technology Of Spectrum Sensing Based On Random Matrix Theory

As cognitive radio develops, the technology of spectrum sensing has gained widely attention. As the basis of cognitive radio system, it detects frequently to determine whether there is available idle spectrum. Random matrix theory is an effective way to deal with large dimensional data. It develops rapidly in recent years, and has gradually become one of the important tools in the field of wireless communication. Random matrix theory has been widely used in spectrum sensing, due to its unique advantages compared with the traditional spectrum sensing. More and more researchers study the theory of random matrix. This paper aims to construct the spectrum sensing algorithms with excellent performance based on the new results of random matrix theory.Firstly, this paper introduces the background of the research, and summarizes the research status of spectrum sensing technology based on random matrix theory. Then the basis of the random matrix theory is studied. After that the principle of the maximum-minimum eigenvalue(MME) detection algorithm based on the random matrix theory is specifically analyzed, as well as its improved process. And the general methods and steps of applying random matrix theory to spectrum sensing are summarized afterwards.Secondly, the application of Single Ring Theorem in spectrum sensing is studied, and the spectrum sensing based on Single Ring Theorem is proposed. Single Ring Theorem is a new result of the random matrix theory. Researches show that the eigenvalues of the received signal matrix are generally distributed in an annulus when the signal of the primary user does not exist, while it ’s no longer suitable for the signal-present case. The difference in two cases can be utilized to detect the presence of the primary signal. In order to express this difference better, we define "mean eigenvalue radius"(MER) as the test statistic of the algorithm. The decision threshold of the algorithm is also given in this paper. Simulation results show that the algorithm is not affected by noise uncertainty, and has a good detection probability even if the received signals among sensing nodes are uncorrelated.Finally, the spectrum sensing based on the eigenvalue and eigenvector is studied, and a dual-feature spectrum sensing algorithm based on the maximum-minimum eigenvalue and the principal eigenvector is proposed. Eigenvalues and eigenvectors of the signal covariance matrix both contain the signal’s characteristics. Existing detection algorithms, however, only utilize the properties of eigenvalues or eigenvectors, and do not combine them together for spectrum sensing. The proposed algorithm in this paper considers the characteristics of both of them, and the relationship between the false-alarm probability and the threshold value is derived according to random matrix theory. Theoretical research and simulation results show that this algorithm has some advantages compared with existing algorithms using only the eigenvalues or the eigenvectors.