Fast Authentication and Connectivity Management Schemes in Heterogeneous Wireless Networks

• Main Authors: Huang, Kuei-Li, 黃貴笠
• Other Authors: Tseng, Chien-Chao
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/43641422388856841179

博士 === 國立交通大學 === 資訊科學與工程研究所 === 101 === A wireless broadband network such as WiMAX integrates various communication technologies to form a heterogeneous wireless network for multiple purposes. Integration of wireless access networks extends the access service to a wider range of users. To this end, wireless local area networks such as WiFi are integrated with a WiMAX network. A wireless local area network is considered complementary to a wireless broadband network, due to its cost-effectiveness on covering signal dead zones of WiMAX networks and its plentiful bandwidth for quality of service. However, vertical handovers in a heterogeneous wireless network suffers from the lengthy authentication problem. Therefore, we propose an fast authentication mechanism for mobility management in this network environment. In addition, a mobile user receiving a session call in a wireless network experiences various kinds of handovers. To evaluate the performance of mobility management of a wireless network, the ratio of one to another handovers is an important parameter, which can be used to compute average handover delay, average signaling cost, etc. Hence, we investigate the cluster-to-cell crossing ratio for session-based service in a general user mobility model, where cluster crossings and cell crossings trigger different types of handovers. On the other hand, to provide a wider range of application services, a wireless sensor network is integrated into a wireless broadband network through which the sensor network reports sensing data back to a server on Internet. Standardization groups, such as WiMAX Forum and 3GPP, have been working on such an integration. A wireless sensor network always faces the connectivity management issue due to battery-powered sensor nodes. Therefore, we propose three connectivity management schemes: a backbone-aware topology formation scheme allowing power-rich nodes forming a backbone, an asynchronous routing and route recovery scheme for mobile sensor nodes, and a load balancing scheme. We conduct NS-2 simulations to evaluate performance of the schemes.