Strategy Design And Optimization Fordynamic Spectrum Access System

Dynamic spectrum access(DSA) is a promising solution to address the “spectrum scarcity” and inefficiency of spectrum usage. The basic concept of dynamic spectrum access: Secondary users(SUs) temporarily occupy the spectrum owned by the primary system, as long as interference to the primary users(PUs) is insignificant. By using this technique, SUs can share the spectrum resources with PUs to increase the efficiency of spectrum utilization, as well as alleviate the pressure of spectrum scarcity.DSA strategy plays a crucial role in DSA system, which defines the interaction process between primary system and secondary system. Most existing works related to DSA are focusing on optimize the system performance with the determined DSA strategy. However, there still lacks of works on DSA strategy itself. In this thesis, we aim to design and optimize a flexible DSA strategy that effectively adapts to the dynamics of network environment, and thereby improve the spectrum efficiency. The main contents of the thesis include:First, we briefly introduce the background and significance of dynamic spectrum access technology. To summarize and analyze current development of domestic and foreign dynamic spectrum access technologies. Then, we make a brief introduction of continuous-time Markov chains(CTMC) and queuing theory which are necessary when modeling and analysis of dynamic spectrum access strategy; and three typical DSA strategies based on cognitive radio are elaborated: Random Preemptive DSA(RP-DSA), Collaborative Preemptive DSA(CP-DSA), and Limited Access DSA(LA-DSA), and we compare the advantages and disadvantages of these strategies. Secondly, A novel multi-strategy dynamic spectrum access scheme based on the three typical DSA strategies is proposed. To increase the flexibility of system, this scheme combines RP-DSA and CP-DSA according to the network scenarios. To achieve better system performance, this work shows how to allocate the channel and how to select the probability of false alarm in the sensing process. At last, we introduce the system model and protocol of the channel reorder mechanism with no sensing process or no spectral switching ability. By introducing quasi reserved channel, two optimization schemes are proposed to increase the flexibility of system. Unnecessary drop and block are avoided while the transmission quality of primary users can be guaranteed.