Abstract:The throughput of AIMD protocols in general and of TCP in particular, has been computed in many existing works by modeling the round-trip time as a constant and thus replacing it by its expectation. There are however many scenarios in which the delays of packets vary, causing a variation of the round-trip time. Many typical scenarios occur in wireless and mobile networks. We propose in this paper an analytical model that accounts for the variability of delay, while computing the throughput of an AIMD protocol. We derive a closed-form expression for the throughput, that illustrates the impact of delay variability. We show by analysis and simulation, that an increase in the variability of delay improves the performance of an AIMD protocol. Thus, an analytical model that only considers the average delay could underestimate the performance of an AIMD protocol in scenarios where delay is variable.
Key-words: TCP, delay variability, stochastic recursive equations.This is an extension of a paper that appeared at INFOCOM'04 which was restricted to a Poisson loss process with fixed parameter.
£University of Nice-Sophia Antipolis.
¤Also with Universidad Autónoma Metropolitana, Mexico.
Analyse de protocoles AIMD sur des trajectoires à délai variableRésumé : Les modèles traditionnels des protocoles AIMD, et en particulier ceux du protocole TCP, calculent le débit en modélisant le délai d'aller-retour comme étant une constante, ainsi celui-ci est remplacé par son espérance. Pourtant, il existe plusieurs scénarios pour lesquels le délai des paquets varie, ce qui cause une variation du temps d'aller-retour. Les réseaux mobiles et les réseaux sans fil sont les meilleurs exemples de ce type de scénario. Nous proposons dans ce papier, un modèle analytique prenant en compte la variabilité du délai pour calculer le débit d'un protocole AIMD. Nous obtenons une expression close pour le débit, et montrons par analyse et simulation qu'une augmentation de la variabilité du délai améliore les performances d'un protocole AIMD. En conséquence, un modèle analytique qui ne considère que le délai moyen pourrait sous-estimer les performances de ce type de protocole dans des scénarios où le délai est variable.
Mots-clés :TCP, variabilité du délai, équations récursives stochastiques.
AIMD protocols over paths with variable delay3
In interactive voice applications, FEC allows to recover from losses. FEC schemes need to be simple in order to allow interactivity. We study a simple scheme implemented in FreePhone audio tool [High Speed Networking Group, INRIA, Sophia Antipolis, http://www-sop.inria.fr/rodeo/fphone] and RAT: Robust Audio Tool [Multimedia Integrated Conferencing for European Researchers, University College, London], where for every packet n, redundant information is added in packet n þ /. The quality of a reconstructed copy depends on the amount of redundancy. Using queuing analysis we obtain simple expressions for the audio quality as a function of the amount of redundancy. We show that the expected amount of useful information deteriorates for any amount of FEC and any /. We find conditions under which FEC leads to quality improvement. Ó
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.