分组交换网络中拥塞计费策略及拥塞率的计算

随着网络视频和流媒体等应用的逐渐普及,用户对带宽的需求日益增加.基于多TCP流的应用大量挤占网络带宽,影响网页浏览和电子邮件等低数据量应用的体验.如何在不过度限制高带宽需求应用的前提下,有效改善低数据量应用的体验,成为目前分组交换网络亟待解决的问题.拥塞计费是该问题的一种可行的解决方法.其基本思想是在网络发生拥塞时按照用户所造成拥塞量的多少来对用户进行收费.本文介绍了基于拥塞量收费的网络拥塞计费策略,建立了多拥塞点网络的数学模型,给出了计算拥塞率的方法,并对其算法复杂度进行了分析.最后,本文通过网络仿真对所提出的模型进行了验证.

Congestion Pricing Strategy in Packet Switching Networks and Congestion Volume Rate Calculation

As bandwidth hungry applications, such as online video and streaming media, become more and more popular, the require- ment of bandwidth is increasing. Such bandwidth intensive applications will starve other light applications, like web browsing and email, by establishing a large number of TCP flows. Improving the users＇ experience of light application without affecting bandwidth intensive applications is an open question in current packet switching networks. Congestion pricing is considered to be a good candidate to solve this problem. The basic idea of congestion pricing is that users be charged for their amount of traffic that has actually caused congestion. This paper introduces the congestion pricing strategy and develops a mathematical model to calculate the congestion volume rate in nmhi - bottleneck networks. We analyze our results and discuss the computational complexity of the proposed model. We also verify the model with ns -2 simulation.