A Tree-Based Approach for Computing Double-Base Chains

  • Authors:
  • Christophe Doche;Laurent Habsieger

  • Affiliations:
  • Department of Computing, Macquarie University, Australia;Institut Camille Jordan, CNRS UMR 5208, Villeurbanne Cedex, France 69622

  • Venue:
  • ACISP '08 Proceedings of the 13th Australasian conference on Information Security and Privacy
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

We introduce a tree-based method to find short Double-Base chains. As compared to the classical greedy approach, this new method is not only simpler to implement and faster, experimentally it also returns shorter chains on average. The complexity analysis shows that the average length of a chain returned by this tree-based approach is $\frac{\log_2 n }{4.6419}\cdotp$ This tends to suggest that the average length of DB-chains generated by the greedy approach is not O(logn/loglogn). We also discuss generalizations of this method, namely to compute Step Multi-Base Representation chains involving more than 2 bases and extended DB-chains having nontrivial coefficients.