Channel reservation for dynamic spectrum access of cognitive radio networks with prioritized traffic

In this paper, we propose a strategy to coordinate the dynamic spectrum access (DSA) of different types of traffic. It is assumed that the DSA assigns spectrum bands to three kinds of prioritized traffic, the traffic of the primary network, the Class 1 traffic and Class 2 traffic of the secondary network. Possessing the licensed spectrum, the primary traffic has the highest access priority and can access the spectrum bands at anytime. The secondary Class 1 traffic has higher priority compared to secondary Class 2 traffic. In this system, a channel reservation scheme is deployed to control spectrum access of the traffic. Specifically, the optimal number of reservation channels is applied to minimize the forced termination probability of the secondary traffic while satisfying a predefined blocking probability of the primary network. To investigate the system performance, we model state transitions of the DSA as a multi-dimensional Markov chain with three-state variables representing the number of primary, Class 1, and Class 2 packets in the system. Based on this chain, important performance measures, i.e., blocking probability and forced termination probability are derived for the Class 1 and Class 2 secondary traffic.