T-DLRP: Detection of Fast Retransmission Losses Using TCP Timestamp for Improving the End-to-End Performance of TCP over Wireless Networks

Author:

Sreekumari Prasanthi1,Chung Sang-Hwa2,Lee Meejeong1,Kim Won-Suk2

Affiliation:

1. School of Computer Science and Engineering, Ewha Womans University, Seoul 120-750, Republic of Korea

2. School of Computer Engineering, Pusan National University, Busan 609-735, Republic of Korea

Abstract

Improving the end-to-end performance of TCP over wireless networks has been an active research area. When TCP operates in wireless networks, the performance of TCP degrades significantly. It is well known that frequent retransmission timeouts are the leading source of TCP performance degradation over wireless networks. Retransmission timeouts are unavoidable, when the fast retransmission of a lost packet fails to reach the destination. There have been many schemes developed for reducing the retransmission timeouts of TCP. However, these schemes have no mechanism to detect the loss of fast retransmitted packets over wireless networks. In this paper, we introduce an efficient TCP mechanism, called T-DLRP, which is able to detect the loss of fast retransmitted packets and react accordingly. To justify our contributions, we compared T-DLRP against TCP New Reno, Vegas, New Jersey, Veno, Cerl, and DAC via extensive simulations using Qualnet 4.5 simulator. The results demonstrate that our mechanism achieves significant improvement over existing TCP variants especially in a coexisted network with packet loss and fast retransmission losses. Compared to DAC, T-DLRP gains more than 70% improvement in terms of accuracy over wireless networks. Also, our experiments of multiple connections of the same TCP flows reveal that T-DLRP has better fairness compared to TCP NewReno.

Publisher

SAGE Publications

Subject

Computer Networks and Communications,General Engineering

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3