摘要
针对计算机高速互联网中发送端速率调节的问题,在一般网络模型基础上,将BP(BackPropagation)神经网络运用到计算机网络的拥塞控制中,提出了一种基于BP神经网络的动态资源管理机制以解决网络的拥塞问题。对所提出的拥塞控制方案,进行了仿真分析,仿真结果显示,控制方案有较好的可扩展性、有效性,并使网络性能表现良好。
With regard to the flow regulation of the source rate in high-speed computer communication network, the neural network is used to congestion control of the computer network and a novel dynamic resource management scheme based on BP (back propagation) neural network is proposed to solve the congestion of network. To validate the proposed algorithm, it has fulfilled a variety of simulations under various traffic conditions and network environment. The results of simulation demonstrate the proposed control scheme is scalable and efficient, and has good network performance.
出处
《计算机工程》
CAS
CSCD
北大核心
2004年第24期35-36,127,共3页
Computer Engineering
基金
国家自然科学基金资助项目(60174043)
湖北省自然科学基金重点资助项目(2002AB025)
关键词
BP神经网络
缓冲占有量
拥塞控制
预测控制
BP neural network
Buffer occupancy
Congestion control
Predictive control
