Proactive channel access in cognitive radio networks based on users´ statistics

The issue of channel selection in a decentralized cognitive radio network having a spectrum broker is discussed in this paper. In the proposed method we take into account both the Primary user (PU) channel usage statistics and, Secondary user (SU) channel usage statistics as perceived by an SU of interest. In the marked Markov process based method proposed, each SU calculates the reward of using each channel and selects the channel set with the highest reward, which will in turn be used for proactive spectrum access and hand-off. We compare our scheme with two other schemes, where the channel set is randomly selected in one and selected based only on PU statistics in the other. By comparison of the simulation results, we show that our scheme causes less interference to the PUs and has comparable throughput performance which compensates for the increased computational complexity.