分层异构网络信息交互效率优化仿真研究

摘要

In this paper,we proposed an efficiency optimization method of information interaction in hierarchical heterogeneous network based on the game learning.Firstly,the author introduced the game theory into information interaction process of hierarchical heterogeneous network and used the normalization function to characterize the heterogeneous requirement among different microcellular.Then,the network satisfaction utility was defined as the sum of all satisfaction utilities of microcellular and built the game model of heterogeneous information selection.Finally,the maximum network satisfaction utility and maximum potential energy function were defined and the information interaction efficiency integrated with the particle swarm method was optimized.The simulation results show that the model mentioned above can calculate the maximum network satisfaction utility accurately.It has higher modeling precision and is of guiding significance in improving the information interaction efficiency in hierarchical heterogeneous network.%对分层异构网络信息交互效率进行优化,可提高分层异构网络信息的处理效率.通过构建归一化函数建立分层异构网络信息选择博弈模型,定义最大化网络满意效用与最大化势能函数,才能对分层异构网络信息交互效率进行优化,但是传统方法通过先对不同的信息源进行定位,将空域和时域信息融合于分层异构网络信息交互中完成效率优化,但是不能准确构建归一化函数,无法通过建立信息选择博弈模型计算满意效用与最大化势能函数,降低了交互效率.提出基于博弈学习的分层异构网络信息交互效率优化方法.上述模型先将博弈理论引入到分层异构网络信息交互过程中,利用归一化函数表征不同微蜂窝之间的异构需求,将网络满意效用定义为全部的微蜂窝满意效用之和,构建异构信息选择博弈模型,定义最大化网络满意效用与最大化势能函数,并融合粒子群方法对分层异构网络信息交互效率进行优化.仿真结果表明,所提模型可准确计算网络最大化满意效用,建模精确度高,对分层异构网络中能效信息交互效率的提高具有指导意义.