Segment based model for TCP protocol optimization: Enhancing the bandwidth and congestion

Mojtaba Jafari, Abeer Alsadoon, Chandana Withana, Salih Ali, A. Elchouemic

Research output: Book chapter/Published conference paperConference paper

3 Citations (Scopus)

Abstract

This paper aims to provide an efficient segment-based model for the Transmission Control Protocol (TCP) protocol to improve the transmission mechanism by enhancing both bandwidth utilization and congestion. The study puts forward a segment-based model that encompassed three phases, namely: network environment evaluation to calculate bandwidth, which is the Round Tip Time (RTT) independence, organizing packets into groups of segments, and transmission of segments within a fraction of a millisecond delay with the predefined speed. Network evaluation is done by considering the capacity of the network and the time taken for some insignificant packets to be transmitted to the server. Packet segmentation organizes several packets into a group at predefined speed, which then will be transmitted to the destination at a fraction of a millisecond. Experimental results using the segment-based model for the TCP protocol show that the proposed solution significantly improves the transmission mechanism of TCP. Further, the results of the experiment show that bandwidth is utilized to near the same value of available bandwidth. Furthermore, the transmission time is improved compared to previous models of TCP. The proposed segment-based model allows for the TCP protocol to utilize better bandwidth and avoid congestion in a network while transmitting data. The transmission time will decrease significantly, allowing more data to be transferred, resulting in the robust reliability of the TCP protocol.
Original languageEnglish
Title of host publicationProceedings of the 2018 IEEE 8th Annual Computing and Communication Workshop and Conference
Place of PublicationUSA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages918-924
Number of pages7
ISBN (Electronic)9781538646496
DOIs
Publication statusPublished - 27 Feb 2018
Event8th IEEE Annual Computing and Communication Workshop and Conference: IEEE-CCWC 2018 - University of Nevada, Las Vegas, United States
Duration: 08 Jan 201810 Jan 2018
https://web.archive.org/web/20171117133749/http://ieee-ccwc.org/ (Archived page)
http://ieee-ccwc.org/ (Conference website)

Conference

Conference8th IEEE Annual Computing and Communication Workshop and Conference
CountryUnited States
CityLas Vegas
Period08/01/1810/01/18
OtherWe are proud to present IEEE CCWC 2018 which will provide an opportunity for researchers, educators and students to discuss and exchange ideas on issues, trends, and developments in Computing and Communication. The conference aims to bring together scholars from different disciplinary backgrounds to emphasize dissemination of ongoing research in the fields of in Computing and Communication. Contributed papers are solicited describing original works in the above mentioned fields and related technologies. The conference will include a peer-reviewed program of technical sessions, special sessions, business application sessions, tutorials, and demonstration sessions. All accepted papers will be presented during the parallel sessions of the Conference and papers will be submitted for publication at IEEE Xplore Digital Library.
This conference will also promote an intense dialogue between academia and industry to bridge the gap between academic research, industry initiatives, and governmental policies. This is fostered through panel discussions, keynotes, invited talks and industry exhibits where academia is exposed to state-of-practice and results from trials and interoperability experiments. The industry in turn benefits by exposure to leading-edge research in networking as well as the opportunity to communicate with academic researchers regarding practical problems that require further research.
Internet address

Fingerprint Dive into the research topics of 'Segment based model for TCP protocol optimization: Enhancing the bandwidth and congestion'. Together they form a unique fingerprint.

Cite this