Dynamic load-balancing spectrum decision for cognitive radio networks with multi-class services

In this paper, we study dynamic load-balancing spectrum decision for cognitive radio networks that dynamically distributes packets from a secondary user (SU) to different available primary channels. Different from the existing works in the literature, we consider two different classes of services at the SU, delay sensitive (DS) and best effort (BE) services, while assigning a higher priority to the DS services. We propose a new queueing analytical model to address this priority issue and analyze the delay performance for the two services separately. Based on the analytical results, two Markov decision processes (MDPs) are formulated with objectives to minimize the average delay of both services, while guaranteeing the priority of the DS services. Simulation results show that the proposed MDP scheme greatly improved the delay performance for both DS and BE services, compared to the shortest queue scheme.