English

Space Efficient Multi-Dimensional Range Reporting

Data Structures and Algorithms 2009-04-24 v2 Computational Geometry

Abstract

We present a data structure that supports three-dimensional range reporting queries in O(loglogU+(loglogn)3+k)O(\log \log U + (\log \log n)^3+k) time and uses O(nlog1+\epsn)O(n\log^{1+\eps} n) space, where UU is the size of the universe, kk is the number of points in the answer,and \eps\eps is an arbitrary constant. This result improves over the data structure of Alstrup, Brodal, and Rauhe (FOCS 2000) that uses O(nlog1+\epsn)O(n\log^{1+\eps} n) space and supports queries in O(logn+k)O(\log n+k) time,the data structure of Nekrich (SoCG'07) that uses O(nlog3n)O(n\log^{3} n) space and supports queries in O(loglogU+(loglogn)2+k)O(\log \log U + (\log \log n)^2 + k) time, and the data structure of Afshani (ESA'08) that uses O(nlog3n)O(n\log^{3} n) space and also supports queries in O(loglogU+(loglogn)2+k)O(\log \log U + (\log \log n)^2 + k) time but relies on randomization during the preprocessing stage. Our result allows us to significantly reduce the space usage of the fastest previously known static and incremental dd-dimensional data structures, d3d\geq 3, at a cost of increasing the query time by a negligible O(loglogn)O(\log \log n) factor.

Keywords

Cite

@article{arxiv.0806.4361,
  title  = {Space Efficient Multi-Dimensional Range Reporting},
  author = {Marek Karpinski and Yakov Nekrich},
  journal= {arXiv preprint arXiv:0806.4361},
  year   = {2009}
}
R2 v1 2026-06-21T10:54:44.949Z