Tracking the State of Primary Users Based on Purely Discontinuous Markov Process

This paper focuses on tracking the state of primary users in spectrum detection in cognitive radio. During the process of spectrum detecting, there exist only two states of the primary user: occupying or leave. The cumulative occupying time of the primary users and their distribution in time domain along with the variation of the state represented by the transition matrix can be deduced by Fokker-Plank equation and the property of the residence time in any time for purely discontinuous Markov process following exponential distribution. Analysis of computer simulation results suggest that the probability distribution of the acquired primary user occupying time is the main basis for the tracking of the state of the primary users.