The Theory Research And Realiazation Of Traffic Off-Loading Algorithm And Schemes In Heterogeneous Networks

The continuous development and breakthroughs of wireless communication theory and practical techniques has created a heterogeneous network scenario of2G/3G/4G/WLAN. At the same time, the rapid development of mobile internet traffic is a great challenge for resource management for network operators.The techniques of multi-mode terminals provide a creative solution for resource management schemes for multiple traffic requirements especially in the situation of limited available bandwidth, which is traffic splitting scheme. After network operators receive the requirements of users, they will assess the network conditions of heterogeneous networks to judge the radio of traffic to different networks. This provides a concurrent multi-path traffic transmission, which can utilize the ability of network cooperation and improve the user Qos.We pay a great attention to the traffic splitting schemes and algorithms. Besides the combination of theory study and practical use, we also provide a common method based on utility functions for the more complex scenario.Firstly, we provide a Queuing Theory based traffic splitting scheme. Through modifying the serve speed of queuing system, we make it more suitable to dynamic wireless scenario.Secondly, for multi-user traffic splitting situation, we introduce utility functions. Utility function is a way to solve complex traffic splitting problems. We verify the performance gain of our provide schemes over two another classic resource management schemes.We also built a video traffic splitting system in LTE-WLAN heterogeneous network situation, in which we realize the basic modules and process of traffic splitting system. After that, we make some algorithms validation and achievements. The results show that our schemes can truly improve network efficiency and assure user experience.Finally, we briefly summarize the work done during the graduation design and prospect the future study of the streaming in heterogeneous multi-gateway networks.