| Summary: | 博士 === 國立交通大學 === 資訊科學與工程研究所 === 95 === In recent years, there are many forces urging operators to provide the converged services. First, customers hope to enjoy unified services on different type of devices. Second, users want to have uniform services and contents regardless of the kind of underlying access network being used, such as WiFi, 3G, ethernet, WiMax, ADSL, or cable. Third, fixed network providers’ profit is reducing, so they must begin to look for the opportunities in wireless domain. And wireless operators’ speed of growth slows down, makes them begin to explore how to expand their markets. Finally, the competition between the fixed and wireless carriers will drive the network service toward convergence, and some successful cases of Internet telecommunications services such as Yahoo! Messenger and Skype, etc., make Internet Protocol (IP) applications take very important role during the process toward convergence.
As consumers become increasingly mobile, they will demand wireless Internet access from everywhere. In keeping with these requirements of end users, IP Multimedia Subsystem (IMS) standards based on the 3GPP UMTS become more and more important. In the IMS system, we propose an integrated call agent of the converged VoIP network. We presente a simple, flexible framework for the interworking functions of VoIP protocols such as Media Gateway Control Protocol (MGCP), Session Initiation Protocol (SIP) and H.323 base on Intelligent Network (IN).
In UMTS two-pass authentication, many steps in the General Packet Radio Service (GPRS) authentication and IMS authentication are duplicated. Therefore, we propose an one-pass authentication procedure, in which only the GPRS authentication procedure is performed. In the IMS network, the authentication is implicitly executed in the IMS registration. We formally prove that the IMS user is correctly authenticated, and the one-pass authentication saves up to 50% of the IMS registration/authentication traffic.
In the service platform of IP multimedia services, we describe the design and implementation of a SIP-based VoIP call center with waiting time prediction. The SIP-based plug-in modular call center architecture and detailed message flows are elaborated. We propose two output measures and develop a discrete event simulation model to investigate the performance of the waiting time prediction algorithm for the call center.
These research results presented in this dissertation can be viewed as a useful foundation for further study in call control for IP multimedia services and authentication.
|
|---|
