Research On Queue Scheduling Mechanism Of QoS Management Architecture In Cognitive Network

As society and technology develops, people's requests to the Internet technology are daily increasing. Besides, traditional Internet architecture has security flaws, constantly dynamic repairs, and other problems, and it needs to be improved. In such a situation, the research of novel architectures is emerging. As one of the research regions, the Cognitive Network is of more concern. Moreover, with the network applications increasing, though the processing capacity of network is improved a lot, it is still not nearly enough for the sharply growing user's need. Therefore, during the development of Internet in future, the study of the QoS related technologies, which can autonomicly adapt to dynamic network environments and meet the needs of users, has become one of the hot-spots of the researches on Internet architecture at present.Firstly the thesis outlined the research status of cognitive network at homeland and abroad and its related key technologies related, then reviewed the research state of QoS and analyzed its development trend.Secondly, a kind of QoS management architecture in cognitive network is introduced. It is based on hierarchical cognitive processes, and adopts hierarchical mapping to refine global goals to local ones, so that to implement the end-to-end QoS guarantee.Thirdly, on basis of the QoS management architecture mentioned above, a cross-layer based queue scheduling mechanism is proposed. The mechanism classifies the service flow, senses the quality of channel, and schedules the packet by analyzing the above information, then provide the service adapting the network condition. And then the performance of the mechanism is validated under NS-2. Finally, the last chapter concluded this thesis, analyzed the problems and our future work, and illuminated my work and achievements in my postgraduate study.