• DocumentCode
    437575
  • Title

    Fuzzy based dual active queue management scheme for high performance networks

  • Author

    Ravichandran, K.S.

  • fYear
    2005
  • fDate
    19-22 March 2005
  • Firstpage
    461
  • Lastpage
    466
  • Abstract
    In recent years, active queue management schemes (AQM) with explicit congestion notification (ECN), which support the end-to-end congestion control mechanism of TCP/IP by performing congestion control, have been actively studied by many researchers. It is used to improve the performance of TCP/IP networks. Fuzzy double slope random early detection (FDSRED) and fuzzy enhanced random exponential marking (FEREM) are an efficient active queue management schemes to achieve high link utilization with low queuing delay. However, in our study, we reveal that, FDSRED has better throughput and low delay for low load of networks and FEREM has better throughput and low delay for heavy load of networks. In this paper, we have combined both FDSRED and FEREM with fuzzy aggregation operations called dual AQM. Result shows that our proposed scheme "fuzzy based dual active queue management scheme for high performance networks" achieves better throughput and low delay for both low load and heavy load of networks.
  • Keywords
    IP networks; fuzzy set theory; queueing theory; telecommunication congestion control; transport protocols; IP networks; TCP/IP; end-to-end congestion control mechanism; explicit congestion notification; fuzzy aggregation operations; fuzzy based dual active queue management scheme; fuzzy double slope random early detection; fuzzy enhanced random exponential marking; high performance networks; transport protocols; Art; Delay; Engineering management; IP networks; Internet; Scalability; Synthetic aperture sonar; TCPIP; Throughput;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Networking, Sensing and Control, 2005. Proceedings. 2005 IEEE
  • Print_ISBN
    0-7803-8812-7
  • Type

    conf

  • DOI
    10.1109/ICNSC.2005.1461234
  • Filename
    1461234