• DocumentCode
    668717
  • Title

    An adaptive active queue management algorithm

  • Author

    Hongcheng Huang ; Gaofei Xue ; Yuanming Wang ; Hongmei Zhang

  • Author_Institution
    Dept. of Software Center, Chongqing Univ. of Posts & Telecommun., Chongqing, China
  • fYear
    2013
  • fDate
    20-22 Nov. 2013
  • Firstpage
    72
  • Lastpage
    75
  • Abstract
    When we face the request of the massive users and the dynamic change of network, the existing active queue management mechanism can not satisfy the effectiveness, stability and robustness at the same time, thus we can´t better control the congestion state of the network. This paper based on PI (Proportional Integral) algorithm, it proposed an adaptive PI active queue management algorithm API(Adaptive Proportional Integral). API gets the current queue length by real-time monitoring router buffer queue. Based on the current queue length with the target queue length deviation, it dynamically adjusts the relevant parameters in PI algorithm and the packet loss probability. Through the ns-2 simulations, it shows that, API has a faster response speed and a smaller queue jitter than PI algorithm, in the dynamic change of network, it shows a better stability and robustness, and more effective to alleviate the network congestion.
  • Keywords
    PI control; queueing theory; telecommunication congestion control; telecommunication network management; adaptive active queue management algorithm; adaptive proportional integral; congestion state control; network congestion; proportional integral algorithm; real-time monitoring router buffer queue; Algorithm design and analysis; Heuristic algorithms; Packet loss; Queueing analysis; Robustness; Stability analysis; active queue management; adaptive; congestion control; proportional integral;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Consumer Electronics, Communications and Networks (CECNet), 2013 3rd International Conference on
  • Conference_Location
    Xianning
  • Print_ISBN
    978-1-4799-2859-0
  • Type

    conf

  • DOI
    10.1109/CECNet.2013.6703275
  • Filename
    6703275