Reference Hub1
Improving Throughput of Starved TCP Flow by Sidestepping Bottleneck Nodes Using Concurrent Transmission

Improving Throughput of Starved TCP Flow by Sidestepping Bottleneck Nodes Using Concurrent Transmission

Rajesh Verma, Arun Prakash, Rajeev Tripathi, Neeraj Tyagi
Copyright: © 2010 |Volume: 2 |Issue: 1 |Pages: 16
ISSN: 1937-9412|EISSN: 1937-9404|ISSN: 1937-9412|EISBN13: 9781616929527|EISSN: 1937-9404|DOI: 10.4018/jmcmc.2010103004
Cite Article Cite Article

MLA

Verma, Rajesh, et al. "Improving Throughput of Starved TCP Flow by Sidestepping Bottleneck Nodes Using Concurrent Transmission." IJMCMC vol.2, no.1 2010: pp.68-83. http://doi.org/10.4018/jmcmc.2010103004

APA

Verma, R., Prakash, A., Tripathi, R., & Tyagi, N. (2010). Improving Throughput of Starved TCP Flow by Sidestepping Bottleneck Nodes Using Concurrent Transmission. International Journal of Mobile Computing and Multimedia Communications (IJMCMC), 2(1), 68-83. http://doi.org/10.4018/jmcmc.2010103004

Chicago

Verma, Rajesh, et al. "Improving Throughput of Starved TCP Flow by Sidestepping Bottleneck Nodes Using Concurrent Transmission," International Journal of Mobile Computing and Multimedia Communications (IJMCMC) 2, no.1: 68-83. http://doi.org/10.4018/jmcmc.2010103004

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The TCP congestion control mechanism along with unfairness problem poses poor performance when IEEE 802.11 MAC protocol is used in multi-hop ad hoc networks because the traditional TCP has poor interaction with the IEEE 802.11 MAC protocol. Because of the greedy nature of TCP, starvation problem of TCP flows with longer paths is severe. In this paper, we first illustrate that the fairness, congestion control and medium contention are closely coupled issues and the spatial reuse of the channel can improve the performance of wireless ad hoc network. By using concurrent transmission protocol at the MAC layer, like CTMAC, in multi-hop networks we can achieve simultaneous transmissions within the interference regions. Further, we illustrate with extensive simulations in ns-2 that by scheduling multiple concurrent transmissions along the path links, the starvation problem due to greedy nature of TCP can be eliminated and ensuing higher flow throughput and lower end-to-end delay.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.