The Stable Channel State Analysis for Multimedia Packets Allocation over Cognitive Radio Networks

In cognitive radio networks (CRNs), the spectrum utilization can be dramatically improved with the secondary users (SUs) accessing the unoccupied licensed channels opportunistically. However, how to fully utilize the spectrum holes to meet the quality of service requirements of SUs is still an open issue. In this paper, we focus on multi-user, multi-channel case, and analyze the stable channel state after allocating packets over a licensed channel. We first assume that the SUs' packet arrival rate obeys Poisson distribution, and the lost packets will be retransmitted with exponential backoff delay. Then, we analyze the stable channel state when a SU selects and allocates certain percentage of its packets over one channel. The theoretic analysis shows that such stable channel state can be solved by the steady-state equations. Based on such stable channel state analysis, we propose a novel greedy packet allocation scheme in multi-user and multi-channel S-ALOHA system. The proposed packet allocation scheme will obtain a maximal spectrum utilization, which will support more SUs to share the spectrum holes or cause less access conflict with PUs. In the simulation part, we assume the PUs' channel access pattern follows Markov model, and study the spectrum efficiency after packet loading over licensed channels and make comparisons with different packet allocation schemes. The simulation results show that the proposed packet allocation scheme outperforms other related works in terms of successful packet delivery ratio, the number of collision packets, and spectrum efficiency.