An End-to-End Support for Short-Lived TCP Flows in Heterogeneous Wired-cum-Wireless Networks: An Ana

An End-to-End Support for Short-Lived TCP
Flows in Heterogeneous Wired-cum-Wireless
Networks: An Analytical Study

Haya Sammaneh1, Jamal Al-Karaki2, and Sameer Bataineh1
1Computer Engineering Department, Jordan University of Science and Technology, Jordan
2Computer Engineering Department, The Hashemite University, Jordan

Abstract: Transmission control protocol has been adapted to various types of wireless networks (e.g., IEEE 802.11 WLAN, 3G cellular and ad hoc networks). However, wireless transmission control protocol flows are much shorter than wired flows due to the time varying characteristics of wireless networks. Hence, transmission control protocol performance in these networks is determined by the start up effects such as connection establishment. Several analytical models were proposed to describe the steady-state behaviour of short-lived transmission control protocol flows in wired networks, while few similar studies targeted wireless networks. Moreover, almost all of the previous models did not investigate the effect of heterogeneity (e.g., link speed and segment loss probability) of wireless networks on the end-to-end performance of transmission control protocol flows. In this paper, a recursive and analytical model is developed and used to determine the performance of short- lived transmission control protocol flows in heterogeneous wired-cum-wireless networks in terms of average completion time. Two different schemes are proposed to solve the wireless loss problems by the end hosts, namely, end-to-end scheme and connection division scheme. The proposed analytical model has been validated by means of simulations using NS-2 simulator. Performance results show that the proposed model is in close analogy with values obtained from the analytical model. As such, the proposed model can be used to accurately tune many parameters that affect the behaviour of transmission control protocol in wired-cum-wireless networks.

Keywords: Software engineering education, and reflective learning and teaching.

Received June 29, 2009; accepted November 5, 2009

Full Text
Read 5491 times Last modified on Wednesday, 15 December 2010 03:01
Share
Top
We use cookies to improve our website. By continuing to use this website, you are giving consent to cookies being used. More details…