Opportunistic Spectrum Access with Spectrum Heterogeneity in Cognitive Networks

We study the problem of dynamic spectrum sensing and access in cognitive radio networks. A distributed C-MAC (cognitive media access control) protocol is proposed. And based on the theory of the partially observable Markov decision process, a greedy opportunistic spectrum access scheme that maximizes the current reward of secondary users is proposed considering spectrum heterogeneity of secondary users. Numerical results are provided to show the performance characters and good profit of the proposed scheme in cognitive networks. It is show that the performance is influenced by the heterogeneity greatly when the available channels are scarce. Along with the number of common channels increasing, the proposed scheme gets a good performance.