东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2008, Vol. 29 ›› Issue (2): 157-160.DOI: -

• 论著 • 上一篇    下一篇

不确定TCP网络中的滑模主动队列管理算法

闫明;井元伟;沈孝钧;   

  1. 东北大学信息科学与工程学院;东北大学信息科学与工程学院;密苏里大学堪萨斯城分校计算与工程学院 辽宁沈阳110004;辽宁沈阳110004;美国堪萨斯城64110-2499
  • 收稿日期:2013-06-22 修回日期:2013-06-22 出版日期:2008-02-15 发布日期:2013-06-22
  • 通讯作者: Yan, M.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(60274009)

Sliding mode AQM algorithm for TCP networks with uncertainty

Yan, Ming (1); Jing, Yuan-Wei (1); Shen, Xiao-Jun (2)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China; (2) School of Computing and Engineering, University of Missouri-Kansas City, Kansas City 64110-2499, United States
  • Received:2013-06-22 Revised:2013-06-22 Online:2008-02-15 Published:2013-06-22
  • Contact: Yan, M.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

752

摘要： 为解决传输控制协议(transmission control protocol,TCP)网络中的拥塞问题,提出了一种基于滑模控制理论的主动队列管理(active queue management,AQM)算法.该算法基于线性TCP网络拥塞控制模型,为补偿网络中不确定因素的影响,采用线性矩阵不等式(linear matrix inequality,LMI)设计了一个渐近稳定的滑模面,从而使滑模面的设计问题转化为LMI的求解问题.通过在拥塞控制系统中应用一个改进的到达条件明显地降低了系统的抖振,满足该到达条件的控制器能够使路由器中队列长度的振荡得到有效的抑制.不同情况下的仿真结果表明该算法具有良好...

关键词: TCP网络拥塞控制, 主动队列管理, 滑模控制, 线性矩阵不等式, 到达条件

Abstract: An active queue management (AQM) algorithm based on sliding mode control (SMC) theory is proposed to solve the congestion problem in transmission control protocol (TCP) networks. The algorithm is put forward on the basis of linear TCP network congestion control model. An asymptotically stable sliding surface is designed by linear matrix inequality (LMI) to compensate for the effect of uncertainty in networks. In this way the design of the sliding surface is transformed into solving the LMI. Applying an improved reaching condition to the congestion control system, the chattering is obviously decreased. The controller satisfying the reaching condition can effectively constrain the oscillation of the queue length in router. The simulation results in different scenarios demonstrate the favorable stability and robustness of the algorithm.

中图分类号: