Multimedia over cognitive radio networks: Towards a cross-layer scheduling under Bayesian traffic learning

Mobility and spectrum change of the cognitive radio networks (CRNs), make the traffic information exchange among the secondary users (SUs) a high-overhead task. In order to quickly estimate the queuing delay in the multimedia over CRN applications without exchanging traffic information among SUs, we are the first group to propose the use of Dirichlet-prior based fully Bayesian model in each individual SU to automatically update its statistical distribution on other SUs' Non-Contiguous (NC)-OFDM subcarrier selection strategy. Such a statistical distribution is used to estimate the probability of queuing delay being less than a threshold. In addition, we introduce a new concept called the Time Window, to accurately determine how many packets can be transmitted simultaneously over multiple subcarriers. Then, we propose a comprehensive, intelligent cross-layer scheduling scheme that can generate the optimal subcarrier selection, power and modulation allocation for each multimedia packet. Our experiments on real video transmission validate our intelligent cross-layer scheduling schemes. The simulation results match with theoretical analysis very well, and the reconstructed video quality using our proposed scheduling scheme is superior to four other popularly used schemes.