电信科学 ›› 2011, Vol. 27 ›› Issue (5): 104-109.doi: 10.3969/j.issn.1000-0801.2011.05.026

• 研究与开发 • 上一篇    下一篇

基于GA-BP神经网络的VOD代理节目预取模型

蒋亚军1,杨震伦1,2   

  1. 1 广州番禺职业技术学院信息工程学院 广州511483
    2 华南理工大学计算机科学与工程学院 广州510640
  • 出版日期:2011-05-15 发布日期:2011-05-15
  • 基金资助:
    广州市高校科技计划资助项目

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

Yajun Jiang1,Zhenlun Yang1,2   

  1. 1 School of Information Engineering,Panyu Polytechnic College,Guangzhou 511483,China
    2 School of Computer Science and Engineering,South China University of Technology,Guangzhou 510640,China
  • Online:2011-05-15 Published:2011-05-15

在线阅读

8

PDF下载

315

摘要：

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。

关键词: 遗传算法, 反向传播神经网络, 节目预取, 代理服务器

Abstract:

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.

Key words: genetic algorithm, BP neural network, data prefetching,proxy

No Suggested Reading articles found!