An efficient mechanism of TCP-vegas on mobile IP networks

Cheng Yuan Ho, Yi-Cheng Chan, Yaw Chung Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Mobile IP network provides hosts the connectivity to the Internet while changing locations. However, when using TCP Vegas over a mobile network, it may respond to a handoff by invoking a congestion control algorithm, thereby resulting in performance degradation, because TCP Vegas is sensitive to the change of RTT (Round-Trip Time) and it may recognize the increased RTT as a result of network congestion. Since TCP Vegas could not differentiate whether the increased RTT is due to route change or network congestion. This work investigates how to improve the performance of Vegas after a Mobile IP handoff and proposes a variation of TCP Vegas, so-called DemoVegas, which is able to detect the movement of two end-hosts of a connection, and re-measure the BaseRTT (Minimum RoundTrip Time) if necessary. The proposed mechanism maintains end-to-end semantics, and operates under the existing network infrastructure. Demo-Vegas presents a simple modification in the two end sides of a connection and uses one reserved bit in TCP header. Simulation results demonstrate that Demo-Vegas features higher performance than Vegas in Mobile IP networks.

Original languageEnglish
Title of host publicationProceedings - INFOCOM 2006
Subtitle of host publication25th IEEE International Conference on Computer Communications
Publication statusPublished - 2006 Dec 1
EventINFOCOM 2006: 25th IEEE International Conference on Computer Communications - Barcelona, Spain
Duration: 2006 Apr 232006 Apr 29

Publication series

NameProceedings - IEEE INFOCOM
ISSN (Print)0743-166X


OtherINFOCOM 2006: 25th IEEE International Conference on Computer Communications

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'An efficient mechanism of TCP-vegas on mobile IP networks'. Together they form a unique fingerprint.

Cite this