論文

査読有り
2014年

Traffic Separation Using Dual Wi-Fi Interface to Enhance WLAN System Throughput

2014 8TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING AND COMMUNICATION SYSTEMS (ICSPCS)
  • Takayuki Nishio
  • ,
  • Ryo Nishioka
  • ,
  • Masahiro Morikura
  • ,
  • Koji Yamamoto

記述言語
英語
掲載種別
研究論文(国際会議プロシーディングス)
出版者・発行元
IEEE

Frame collisions in a wireless local area network (WLAN) reduce system throughput and remain an open problem in WLANs. A number of studies have proposed methods to solve this problem and improve WLAN system throughput successfully. However, some of the methods and the conventional optional functions of WLAN such as the Request to Send/Clear to Send (RTS/CTS) mechanism and channel bonding work well only in specific cases, and their efficiency relies heavily on traffic in the WLAN.
In this paper, we propose a cross-layer traffic separation scheme that enables WLAN systems to improve their performance by using WLAN options more effectively than a conventional WLAN. We focus on the characteristics of traffic in the media access control (MAC) layer. In the MAC mechanism, stations (STAs) employing User Datagram Protocol (UDP) uplink flows tend to cause frame collisions, whereas STAs using Transmission Control Protocol (TCP) flows tend not to cause collisions because the TCP congestion control manages the density of traffic. In the proposed scheme, STAs use dual WLAN interfaces; traffic is separated into two categories, and each category is sent through each interface. WLAN systems select options on each interface in accordance with the characteristics of the traffic at hand. We evaluate the performance of our proposed scheme through a computer simulation. The results show that the proposed scheme increases system throughput by up to 250% compared to conventional WLAN and effects a greater degree of fairness between uplinks and downlinks.

リンク情報
Web of Science
https://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=JSTA_CEL&SrcApp=J_Gate_JST&DestLinkType=FullRecord&KeyUT=WOS:000380403200035&DestApp=WOS_CPL
ID情報
  • Web of Science ID : WOS:000380403200035

エクスポート
BibTeX RIS