Improving the quality of service for users in cognitive radio network using priority queueing analysis

Through cognitive radio (CR), unlicensed users can occupy idle spectrum bands without interference with the primary user (PU). Generally, when the arrival rate of licensed users is high, secondary users (SUs) may starve. In this study, the authors propose two models to improve the average total waiting time for the SU. They are the M/D/1 model of a PU delay system with non-pre-emptive priority and pre-emptive priority. They are compared to an M/D/1 model with PU delay. Numerical analysis and Monte Carlo simulations are performed for all results. Improvements in the average total waiting time of 13-18% for PU and 24-32% for SU in the non-pre-emptive priority scheme are obtained. In the pre-emptive priority scheme, 19-22 and 5-7% improvements are obtained for the PU and SU respectively. Furthermore, same models, but with finite buffer capacity for both users are investigated to model a real-time system. In these schemes, the system improves the average total waiting time by 20-42% for the PU and 34-42% for the CR with the non-pre-emptive priority. Similarly, the pre-emptive model shows 25-44 and 6-24% reduction in the PU and SU waiting times respectively.