Adaptive Cross-layer QoS Mechanism for Cognitive Network Applications

• Main Authors: Shih-Wei Liu, 劉世偉
• Other Authors: Jiann-Liang Chen
• Format: Others
• Language: en_US
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/35500615839391852806

碩士 === 國立東華大學 === 資訊工程學系 === 97 === In the recent years, there has been a tremendous growth in mobile computing technologies and wireless network technologies. Wireless infrastructure has been deployed to be the complement and expansion of the wired infrastructure. There are many options available for users who want to access data and information wirelessly. With the popularization of notebook and handheld device, user emphasized that ubiquitous network environment and high request of network resource. In order to satisfy users’ high quality requirements, a QoS mechanism is needed to address in such heterogeneous wireless network environment. In a limited network resource, applications, especially for real time service, require a quality guaranteed service namely Quality of Service. Different applications have different kinds of QoS concerns and the QoS control parameters are not the same. For example, the VoIP phones concern about the jitter ratio and end to end delay, the online video stream pays much attention of packet loss ration which may influence the definition of video. In this thesis, referring to the QoS specification of three kinds of wireless access technologies, 3G, WiMAX and WiFi to design an adaptive QoS mechanism. Different applications concern about different quality of service parameters. Taking advantage on the concept of Cross-Layer, we integrate these parameters with spectrum and received signal strength sensing from cognitive network, and propose adaptive QoS algorithm to choose the best access network for user. The proposed QoS mechanism not only satisfies the requirements of different application but also achieves the quality of service guarantee. In the simulation results, the proposed adaptive QoS mechanism has performance improvement for real-time applications. It shows that the proposed mechanism decreases the delay time and jitter to 0.1566 seconds and 0.0855 milliseconds respectively for VoIP (Voice over IP) and decreases 3.42 % packet loss ratio for high definition video stream.