New lower and upper bounds for representing sequences

  • Authors:
  • Djamal Belazzougui;Gonzalo Navarro

  • Affiliations:
  • LIAFA, Univ. Paris Diderot - Paris 7, France;Department of Computer Science, University of Chile, Chile

  • Venue:
  • ESA'12 Proceedings of the 20th Annual European conference on Algorithms
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Sequence representations supporting queries access, select and rank are at the core of many data structures. There is a considerable gap between different upper bounds, and the few lower bounds, known for such representations, and how they interact with the space used. In this article we prove a strong lower bound for rank, which holds for rather permissive assumptions on the space used, and give matching upper bounds that require only a compressed representation of the sequence. Within this compressed space, operations access and select can be solved within almost-constant time.