基于GA-BP神经网络的VOD代理节目预取模型

Data Prefetching Model for VOD Proxy Based on GA-BP Neural Network

VOD代理服务器的节目预取方法决定了园区网VOD系统的整体运行效率。提出一种节目预取模型,采用BP神经网络构建分类器并对VOD节目进行分类,再根据分类结果采用基于分组的方法实现代理服务器的节目预取。模型中引入遗传算法对已建立的分类模型进行改进,以克服局部极小值问题。仿真实验表明,该预取模型具有较高的命中率,能有效提高代理服务器的利用率。

The prefetch algorithm of proxy servers is the key element to determine the efficiency of VOD system in campus networks.This paper presents a data prefetching model,BP neural network is used to build the classifier,and the classifier is applied to classify the programs in VOD system,and the classification results were used to commit data prefetch based on group structure and demand rate consideration.Genetic algorithm was introduced to improve the BP neural network,so the problem of incidental trap in local minima with BP neural network was overcome.The simulation results show that model can achieve a good hit ratio and effectively improve the utilization of the VOD proxy servers.