Stochastic modeling of TCP over lossy links

Author

Abouzeid, Alhussein A. ; Roy, Sumit ; Azizoglu, Murat

Author_Institution

Dept. of Electr. Eng., Washington Univ., Seattle, WA, USA

Volume

3

fYear

2000

fDate

26-30 Mar 2000

Firstpage

1724

Abstract

An analytical framework for modeling the performance of a single TCP session in the presence of random packet loss is presented. A Markovian approach is developed that allows us to study both memoryless channels (IID packet loss) and channels with memory (correlated packet loss) modeled by a two-state continuous-time Gilbert model. The analytical results are validated against results using the ns simulator. It is shown that the model predicts throughput for LAN/WAN (low and high bandwidth-delay products) with good accuracy. Further, throughput for the IID loss model is found to be relatively insensitive to the probability density function (PDF) of the loss inter-arrival process. For channels with memory, we present an empirically validated rule of thumb to categorize the channel transition frequency

Keywords

Markov processes; continuous time systems; correlation theory; local area networks; memoryless systems; packet radio networks; performance evaluation; probability; queueing theory; telecommunication traffic; transport protocols; wide area networks; IID packet loss; LAN; Markovian approach; PDF; TCP; WAN; channel transition frequency; correlated packet loss; loss inter-arrival process; lossy links; memoryless channels; ns simulator; performance; probability density function; random packet loss; stochastic modeling; throughput; two-state continuous-time Gilbert model; Analytical models; Local area networks; Memoryless systems; Performance analysis; Performance loss; Predictive models; Probability density function; Stochastic processes; Throughput; Wide area networks;

fLanguage

English

Publisher

ieee

Conference_Titel

INFOCOM 2000. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies. Proceedings. IEEE

Conference_Location

Tel Aviv

ISSN

0743-166X

Print_ISBN

0-7803-5880-5

Type

conf

DOI

10.1109/INFCOM.2000.832572

Filename

832572