Spectrum occupancy modeling on the 450 MHz band for cognitive radio application

Currently the spectrum occupancy has been the major concern for regulatory bodies because there are repressed demands of telecommunications users, especially in large cities due to lack of free frequencies. The cognitive radio technology appears on the scene as a solution to improve the spectrum efficiency by increasing the density of users on radio frequencies. However, there are still open issues to optimize techniques for radio spectrum dynamic access, in order to achieve proficiently the spectral vacancy, called white spaces. In this study measurements were performed on various primary users on the 450 MHz band, which is vastly used, in order to understand the spectrum occupancy models. In the research, the captured signals were demodulated and the audio communication was evaluated in order to identify each sequence considered as a conversation, requiring human interpretation. Within each conversation, bursts of audio (speech of interlocutors) were listed. Besides it the spaces between these bursts and the intervals among conversations were recorded as modeling parameters. The results were statistically analyzed and the study proposes to use this data as a basis for developing an algorithm that aims to improve techniques for access the radio spectrum to enable the powerfully utilization of cognitive radios.