Location-based access discovery based on pattern classification in B3G mobile systems

With various radio access networks being deployed over the world, B3G mobile system is conceived to integrate all these radio access networks in order to provide ubiquitous multiservice network to end users. The discovery of available Radio Access Networks is essential in B3G mobile systems. In this study, a location-based access discovery mechanism based on pattern classification is proposed. It is argued that location-based access discovery problem can be handled by an origin-derivate pattern classification method. Accordingly, a system based on multiple probability neural networks is proposed. The experiment results show that the system can effectively discover available Radio access in B3G mobile systems.