Study On The Key Technology Of Network Selection In Multi-Connection Environment

The merging of multiple access technology has become an important trend in the development of the network. The network resources of heterogeneous network give full play to their advantages of access technologies and access technologies can work together with each other. It is an important direction for multi-connection technology development. Meanwhile, this is the fundamental way to achieve the most optimal allocation of network resources and provide users the best network service. In the multi-connection networks, the terminal can freely choose any access network service. Different network technology can provide a variety of network service, allowing users to get a better experience. How to collect relevant information about the network and make the best network choice based on the information become a hot issue in wireless network fields.This thesis discusses an overview of the development of wireless network. Meanwhile, it analyzes the current network environment. The main work of this thesis is the research of the key technology and network selection algorithm in multi-connection environment. This thesis studies the architecture of multi-connection, network selection technology, application scenario analysis, the challenges faced by multi-connection networks in the future and the development of multi-connection networks. This thesis studies the network selection algorithm and analyzes the existing classical network selection algorithm. And this thesis analyzes the disadvantages of existing classical network selection algorithm. This thesis proposes a new network selection algorithm. It is a cost-sensitive network selection algorithm based on business type. For the problem of cost, the users can set the cost threshold, which accurately grasp the requirements of the users. Networks, terminals and user-related parameters are normalized. Normalization ensures the rationality of network selection. In this thesis, the algorithm has been implemented. The algorithm achieves the expected results.