ÖZDEĞER TABANLI SPEKTRUM ALGILAMA YÖNTEMİNDE KÜMELEME YAKLAŞIMI

With the development of wireless communication recently, the problem of bandwidth shortage has become more prominent. Researches have shown that most of the spectrum is empty most of the time and this part is licensed spectrum. Therefore, to solve the problem of underutilization of the spectrum, Cognitive Radio emerged as a system that allows secondary users to use the licensed band when it is not used by primary users. Cognitive radio (BR) has evolved as a technology that can intelligently detect an empty spectrum and allocate it to secondary users without interfering with licensed users. There are many spectrum detection techniques to detect the presence of primary users. In this paper, eigenvalue-based detection method is preferred as spectrum detection technique. In the sample covariance matrix obtained in this method, MME (maximum minimum eigenvalue-based detection) is used for the ratio of the largest and smallest eigenvalue data. The presence of the signal is investigated by comparing this ratio with a threshold value. It is aimed to detect the presence of the signal by grouping the secondary users with the k-means method, which is the most used clustering method. In addition, the effect on spectrum detection by increasing the number of users is shown in the graphics.