Research On Cross-Layer Optimization In Cognitive Wireless Networks

With the rapid development of wireless communication network in recent years, cognitive wireless network leads the development direction of wireless networks featuring the heterogeneous network collaboration capability, adapting to the changing environment capability, guaranteeing high quality wireless communication services, dynamic management of wireless resources and protocol reconfigurability. Cross layer optimization is one of the important methods to guarantee end-to-end service quality and enhance the efficiency of utilizing the wireless resources in cognitive wireless networks. The content of this thesis includes following aspects:Firstly, this thesis analyzes the tendency of network development, the character of wirelsee network and the challenge and chance from it. Then the cognitive wireless network technology is introduced to prove that it is an efficient method in the heterogeneous wieless network. It can realize the interconnection of networks and optimize the usage of resource in different netwoks. By summarizeing and analyzing the state of the art of the cross layer opitimization technology, this thesis state the role and status of it in cognitive wireless network.Secondly, this thesis introduced the cognitive wireless network and its related technology and research work. Novel cognitive wireless network architecture is proposed to make the cognitive process go in gear and efficiently. In the proposed architecture, the protocol stack level, cognitive elements level and system level of architecture and related functionality is designed. We developed the cognitive plane and cognitive stream concept based on the designed architecture which make heterogeneous network converging through a unified and efficient approach.Thirdly, this thesis proposes a cross-layer optimization scheme for QoS guarantee in multi-user and multi-service OFDMA system. The compound objectives of QoS (requirements for data rate, delay, packet loss, etc.) are comprehensively considered. A model has been established to map these objectives into the lower layer parameters. Multi-user and multi-service diversities are utilized by jointly adjusting the certain physical and link layer parameters along with allocating the OFDMA resources. The allocation problem is solved by Hungarian algorithm and dynamic programming. Numerous simulation results reveal that, on the basis of QoS guarantee, the proposed cross-layer optimization scheme improves the system throughput and the spectrum efficiency.In the end of this thesis, a summarization is made and the future direction of the study is also pointed out.