English

A Deterministic Partition Tree and Applications

Computational Geometry 2025-07-03 v1 Data Structures and Algorithms

Abstract

In this paper, we present a deterministic variant of Chan's randomized partition tree [Discret. Comput. Geom., 2012]. This result leads to numerous applications. In particular, for dd-dimensional simplex range counting (for any constant d2d \ge 2), we construct a data structure using O(n)O(n) space and O(n1+ϵ)O(n^{1+\epsilon}) preprocessing time, such that each query can be answered in o(n11/d)o(n^{1-1/d}) time (specifically, O(n11/d/logΩ(1)n)O(n^{1-1/d} / \log^{\Omega(1)} n) time), thereby breaking an Ω(n11/d)\Omega(n^{1-1/d}) lower bound known for the semigroup setting. Notably, our approach does not rely on any bit-packing techniques. We also obtain deterministic improvements for several other classical problems, including simplex range stabbing counting and reporting, segment intersection detection, counting and reporting, ray-shooting among segments, and more. Similar to Chan's original randomized partition tree, we expect that additional applications will emerge in the future, especially in situations where deterministic results are preferred.

Keywords

Cite

@article{arxiv.2507.01775,
  title  = {A Deterministic Partition Tree and Applications},
  author = {Haitao Wang},
  journal= {arXiv preprint arXiv:2507.01775},
  year   = {2025}
}

Comments

To appear in ESA 2025

R2 v1 2026-07-01T03:43:22.330Z