Tree topology inference from leaf-to-leaf path lengths using Prüfer sequence. (1st January 2014)

• Record Type: Journal Article
• Authors: Chellappan, Vanniarajan; Krithivasan, Kamala
• Is Part Of: International journal of communication networks and distributed systems. Volume 12:Number 2(2014)
• Page Start: 225
• Page End: 237
• Publication Date: 2014-01-01
• Subjects: graphs -- trees -- paths -- network tomography -- Prüfer codes -- topology inference -- distance matrix -- tree-realisation; Computer networks -- Periodicals; Telecommunication systems -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; 004.6
• Languages: English
• ISSNs: 1754-3916

Abstract

In this paper, we propose an algorithm to reconstruct the set of nodes and their topology relationships, particularly in case of a topological tree (where all edge weights are 1), given the set of terminal vertices or leaves and their pairwise distances or path lengths. Prüfer encoding and decoding techniques showed a one-to-one correspondence between n -labelled trees and ( n – 2)-tuples of vertex labels. We consider Prüfer encoding technique in discovering the intermediate vertices from the end-to-end path length between all pairs of terminal vertices. The results are useful in inferring graph characteristics, topology in particular, in the field of network tomography where the internal structure and link-level performance are inferred from end-to-end measurements.