English

Reconstruction in one dimension from unlabeled Euclidean lengths

Metric Geometry 2024-06-27 v2 Combinatorics

Abstract

Let GG be a 33-connected ordered graph with nn vertices and mm edges. Let p\mathbf{p} be a randomly chosen mapping of these nn vertices to the integer range {1,2,3,,2b}\{1, 2,3, \ldots, 2^b\} for bm2b\ge m^2. Let \ell be the vector of mm Euclidean lengths of GG's edges under p\mathbf{p}. In this paper, we show that, with high probability over p\mathbf{p}, we can efficiently reconstruct both GG and p\mathbf{p} from \ell. This reconstruction problem is NP-HARD in the worst case, even if both GG and \ell are given. We also show that our results stand in the presence of small amounts of error in \ell, and in the real setting, with sufficiently accurate length measurements. Our method combines lattice reduction, which has previously been used to solve random subset sum problems, with an algorithm of Seymour that can efficiently reconstruct an ordered graph given an independence oracle for its matroid.

Keywords

Cite

@article{arxiv.2007.06550,
  title  = {Reconstruction in one dimension from unlabeled Euclidean lengths},
  author = {Robert Connelly and Steven J. Gortler and Louis Theran},
  journal= {arXiv preprint arXiv:2007.06550},
  year   = {2024}
}

Comments

25 pages, 2 figures. Final version, to appear

R2 v1 2026-06-23T17:05:06.861Z