sat-ll

From: ketan bajaj ([email protected])
Date: Wed May 24 2000 - 04:08:02 EDT

  • Next message: Craig Partridge: "Re: sat-ll"

    Hi Everyone,
    Attached is a paper which was accepted for publication in Globecomm'99, but
    couldn't be published due to some unavoidable conditions.
    I do not intend to publish it, and it is available for use via this mailing
    list.

    In brief, it presents a protocol which improves TCP performance over
    satellite links by suppressing all the duplicate acknowledgements which
    arise due to losses over the satellite link. As these duplicate
    acknowledgments do not reach the sender TCP, the satellite channel losses
    become transparent to the sender TCP.

    This paper presents a design which works when the satellite link could be
    any intermediate hop in the network. It remains completely transparent to
    the end hosts and the rest of the network, as only the satellite channel's
    end points require the implementation of the design. It preserves all TCP
    semantics.

    The email address shown in the paper no longer exists. If you have some
    questions regarding the design, you can ask me over the mailing list or at
    [email protected].
    Regards,
    Ketan
    ________________________________________________________________________
    Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com





    This archive was generated by hypermail 2b29 : Wed May 24 2000 - 05:40:32 EDT