Machine-learnt versus analytical models of TCP throughput
نویسندگان
چکیده
منابع مشابه
Machine-learnt versus analytical models of TCP throughput
We first study the accuracy of two well-known analytical models of the average throughput of long-term TCP flows, namely the so-called SQRT and PFTK models, and show that these models are far from being accurate in general. Our simulations, based on a large set of long-term TCP sessions, show that 70% of their predictions exceed the boundaries of TCP-Friendliness, thus questioning their use in ...
متن کاملOn the Accuracy of Analytical Models of TCP Throughput
Based on a large set of TCP sessions we first study the accuracy of two well-known analytical models (SQRT and PFTK) of the TCP average rate. This study shows that these models are far from being accurate on average. Actually, our simulations show that 70% of their predictions exceed the boundaries of TCP-Friendliness, thus questioning their use in the design of new TCP-Friendly transport proto...
متن کاملAnalytical TCP Throughput Model for HSDPA∗
In this paper, an approximate, Padhye model based TCP throughput calculation method is presented for mobile data services over HSDPA. The Padhye model is defining the TCP throughput based on two input parameters: the packet loss probability and the TCP Round Trip Time. In order to provide the input parameters for the TCP throughput calculation, an equivalent queuing network model of the HSDPA s...
متن کاملNew Analytical Model for the TCP Throughput in Wireless Environment
In this paper, we propose a new analytical model of different versions of the TCP, viz., Tahoe, Reno and New Reno, where TCP mechanisms such as slow start, fast retransmit, fast recovery and timeout are modeled as a Markov chain. In our proposed model, we consider the exponential increase of the congestion window and the exponential increase of the timeout back-off. Finally, we focus on the bul...
متن کاملImproving TCP in Wireless Networks with an Adaptive Machine-Learnt Classifier of Packet Loss Causes
TCP understands all packet losses as buffer overflows and reacts to such congestions by reducing its rate. In hybrid wired/wireless networks where a non negligible number of packet losses are due to link errors, TCP is unable to sustain a reasonable rate. In this paper, we propose to extend TCP Newreno with a packet loss classifier built by a supervised learning algorithm called ’decision tree ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computer Networks
سال: 2007
ISSN: 1389-1286
DOI: 10.1016/j.comnet.2006.11.017