Journal of Northeastern University Natural Science ›› 2017, Vol. 38 ›› Issue (4): 492-496.DOI: 10.12068/j.issn.1005-3026.2017.04.008

• Information & Control • Previous Articles     Next Articles

Characteristics Mining and Analysis for Internet Prefix Announcements

DENG Wen-ping, LI Zhu-cun, WANG Hong, GAO Xian-ming   

  1. School of Computer, National University of Defense Technology, Changsha 410073, China.
  • Received:2015-04-29 Revised:2015-04-29 Online:2017-04-15 Published:2017-04-11
  • Contact: DENG Wen-ping
  • About author:-
  • Supported by:
    -

Abstract: The BGP routing information was dogged deeply on the basis of a large number of the history of BGP routing table snapshot. A method to measure stability of prefix announcements was designed, it was verified that vast majority of routing announcement was stable, and the historical routing hijacking was short lived (without stability). A similarity measuring algorithm of prefix announcement was presented, and a large number of the history BGP routing announcements were analyzed. The results showed that the announced prefixes of most large ASes are in line with the property of self-similarity, i.e., the same AS declaring multiple routing prefixes with certain continuity. A trustworthy set of prefix-AS mapping was extracted on the basis of these two characteristics, and a trustworthy knowledge base of BGP routing announcement was designed to provide the basis for prefix hijacking detection and routing security monitoring.

Key words: prefix announcement, autonomous system, prefix hijacking, RouteViews, characteristics mining

CLC Number: