Impact of multiple home agents placement in mobile IPv6 environment

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

Mobile IPv6 is an IETF (Internet Engineering Task Force) standard which permits node mobility in IPv6. To manage mobility, it establishes a centralized mediator, Home Agent (HA), which inevitably introduces several penalties like triangular routing, single point of failure and limited scalability. Some later extensions such as Global HAHA, which employed multiple HAs, made to alleviate above shortcomings by introducing Distributed Mobility Management (DMM) approach. However, Multiple HA model will not be beneficial, unless the HAs are located finely. But, no major research paper has focused on locating HAs. This paper examines impact of single and multiple HA placements in data plane, by using an Autonomous System (AS) level topology consisting of 30,000 nodes with several evaluation criteria. All possible placements of HA(s) are analysed on a fair, random set of 30,000 node pairs of Mobile Nodes (MN) and Correspondent Nodes (CN). Ultimate result provides a concise account of different HA placements: i.e. cost centrality interprets performance variation better than degree centrality or betweenness. 30,000 ASs are classified into three groups in terms of Freeman's closeness index and betweenness centrality: 1) high range group, 2) mid range group, and 3) low range group. Considering dual HA placement, if one HA is placed in an AS in the high range group, then any subsequent HA placement gives worse results, thus single HA placement is adequate. With the mid range group, similar results are demonstrated by the upper portion of the group, but the rest yields better results when combined with another HA. Finally, from the perspective of low range group, if the subsequent HA is placed in the high range group, it gives better result. On the other hand, betweenness based grouping yields varying results. Consequently, this study reveals that the Freeman's closeness index is most appropriate in determining impacts of HA placements among considered indices.

Original languageEnglish
Pages (from-to)967-980
Number of pages14
JournalIEICE Transactions on Communications
VolumeE97-B
Issue number5
DOIs
Publication statusPublished - 2014 May

Keywords

  • Data plane
  • Distributed mobility management
  • Freeman's closeness index
  • Global HAHA
  • Home agent placement
  • Mobile IPv6

ASJC Scopus subject areas

  • Software
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Impact of multiple home agents placement in mobile IPv6 environment'. Together they form a unique fingerprint.

Cite this