基于预判决和改进灰关联的异构网络切换算法

传统异构网络切换算法容易引起乒乓效应和切换准确度不高等问题; 同时，现有异构网络切换算法存在网络属性权重确定不合理问题，容易增加不必要的切换次数。因此，针对异构网络切换算法判决不准确带来的乒乓效应问题进行了研究，提出了一种基于差值门限比率和改进灰色关联分析相结合的异构网络切换算法。首先采用差值门限比率进行预判决，初步确定网络是否切换；再利用改进灰色关联分析算法选取目标切换网络，从而决定是否切换网络。搭建了WLAN和UMTS融合的异构网络环境，采用了MATLAB、OPNET等仿真工具。仿真结果表明：该算法有效提高了网络切换判决的准确性，减少了网络切换次数，降低了网络切换失败率。

Heterogeneous network switching algorithm based on pre-decision and improved gay relation

Traditional heterogeneous network switching algorithm easily resulted in the problems table tennis effect and switching accuracy was not high. In the meanwhile, Current heterogeneous network switching algorithm made network attribute weight unreasonable so that it increased unnecessary switching times. Therefore, in view of table tennis effect problem inaccurate judgment of heterogeneous network switching algorithm brought, this paper proposed a heterogeneous switching algorithm which combined difference threshold with improved gay relation analysis. Firstly, this paper used difference threshold rate to make pre-decision whether the network is switched or not. Secondly, this paper used improved gay relation analysis algorithm to select target switching network so as to realize the network switching. This paper established the heterogeneous network fusion environment of WLAN and UMTS, and used MATLAB, OPNET as simulation tools. Simulation results show this algorithm effectively improves the accuracy of network switching, and reduces the number of network switching, and decreases the failure of network switching.