본문 바로가기
  • Home

A Prediction Model of Traffic Flooding in Limited Network Environments

  • Journal of Knowledge Information Technology and Systems
  • Abbr : JKITS
  • 2014, 9(5), pp.545-554
  • Publisher : Korea Knowledge Information Technology Society
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Published : October 31, 2014

이상우 1 백현철 2 홍석원 3 KimSangBok 4

1경남도립남해대학
2경남도립 남해대학
3경남도립 거창대학
4경상대학교

Accredited

ABSTRACT

The subjective of this study is to analyze a possibility of occurring traffic flooding in order to ensure stable services as bandwidth exhausting attacks are presented to users who use limited network bandwidths. Also, a model that can improve service availability and maintain service consistency through rapid responses is proposed. As abnormal traffics occur in a limited network environment, it makes possible to quickly response to traffic flooding attacks, which may occur in future, through estimating changes in subjective traffics. For achieving it, a traffic state model in normal users is extracted as samples of continuous and discrete values in a specific range and the error between these values and its transition process are also used as prediction and detection models. Then, the detection data extracted from normal users are compared and analyzed with the traffics in bandwidth exhausting attacks. Based on the results, the availability and consistency in Internet access services are improved through rapid responses for different unexpected traffics.

Citation status

* References for papers published after 2023 are currently being built.