Research On An Adaptive Vertical Handover Algorithm Based On The Traffic Type For Heterogeneous Networks

With the development of wireless network technology, there have emerged a large number of communication networks with different protocols, i.e. heterogeneous networks. These networks with various forms, have different characteristics, and will co-exist for a long time. The integration and interoperability of heterogeneous networks has become the inevitable trend of the future development of information networks. As the key technique to realize the seamless roaming among heterogeneous networks, the research of vertical handover algorithms has important significance.There are some problems in current vertical handover algorithms. Firstly, as most of the algorithms do not consider the impact of traffic type on handover decision, they cannot choose the most appropriate network depending on the current traffic type of users, resulting in that the selected network cannot meet the user's business needs, thus affecting the user experience. Secondly, current vertical handover algorithms mostly focus on the handover decision strategy, but ignore the trigger condition of vertical handover, which increases the handover delay and terminal power consumption. Thirdly, most algorithms do not use network filtering, which leads to a large amount of computation and a large handover delay. Fourthly, the majority of vertical algorithms cannot dynamic adjust according to the changes of the network environment, which affects the handover performance.In order to solve the above problem of current vertical handover algorithms, this thesis proposes an adaptive vertical handover algorithm based on the traffic type. In this thesis, the business is divided into four types: conversation class, steaming class, interactive class and background class. The proposed algorithm uses the multi-attribute decision making(MADM) method, which has the moderate calculation and better performance, to choose a best network. Then it adopts the forecast mechanism in the handover trigger stage, which can decrease some unnecessary handover, reduce the power consumption of mobile terminal, and timely access to the best network. Aiming to lessen the calculation burden of terminal and shorten the delay, the threshold selection method is used in the network discovery stage. In the handover decision stage, the analytic hierarchy process is used to adjust the parameter weights adaptively, so that the terminal can access the optimal network according to the change of traffic type and terminal capacity. It ensures the quality of service and improves the user experience.Finally, this thesis establishes a LTE-WLAN heterogeneous network model based on the OPNET simulation software. The proposed algorithm is simulated on the platform and its related performance is verified. The simulation data show that compared to the traditional algorithm, the proposed algorithm can flexibly adjust the weights of parameters according to the change of traffic type and terminal capacity. Whether in terms of reducing the network measurement times, handover times and packet loss rate, or in extension of the terminal communication ability, the proposed algorithm has better performance.