<DIV>I have seen a few papers (such as TFRMCC draft) used a simplified TCP throughput equation of Padye where</DIV>
<DIV> (1) estimate RTO = 4 * RTT</DIV>
<DIV> (2) the number of packes acknowledged by each ACK is estimated to 1.</DIV>
<DIV> (3) Min(1, 3*sqrt(3*L/8)) = 1 ; when L = packet loss rate</DIV>
<DIV> </DIV>
<DIV>Then, we can estimateTCP Rate from packet size, RTT and loss rate.</DIV>
<DIV> </DIV>
<DIV>However, I have never seen any papers justifying this simplification. </DIV>
<DIV> </DIV>
<DIV>Why these three simplification making sense?</DIV>
<DIV>Would it be any cases that these simplifications are not good?</DIV>
<DIV> </DIV>
<DIV>thank you,</DIV>
<DIV>Atsuo</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV><p><hr SIZE=1>
Do you Yahoo!?<br>
<a href="http://us.rd.yahoo.com/evt=10469/*http://sitebuilder.yahoo.com">Yahoo! SiteBuilder</a> - Free, easy-to-use web site design software