RUS  ENG
Full version
JOURNALS // Matematicheskie Zametki // Archive

Mat. Zametki, 2025 Volume 117, Issue 5, Pages 672–679 (Mi mzm14366)

Efficient search for a minimum tree in a space with the $l_1$-norm

K. V. Kaymakovab, D. S. Malyshevcd

a National Research University Higher School of Economics, Moscow
b Coleman Group, Moscow
c National Research University – Higher School of Economics in Nizhny Novgorod
d Moscow Institute of Physics and Technology (National Research University), Dolgoprudny, Moscow Region

Abstract: The problem of finding a minimum spanning tree (MST) on an arbitrary set of $n$ points in $d$-space with the $l_1$-norm is considered. It is known that, for each fixed $d\geqslant 2$, there exists an algorithm solving this problem in time $O(n(\log n+\log^{r_d}n\log\log n))$, where $r_d\in\{0,1, 2,4\}$ if $d\in \{2,3,4,5\}$ and $r_d=d$ if $d\geqslant 6$. For $d=3$, the complexity can be improved to $O(n\log n)$. In the paper, for any fixed $d\geqslant 2$, an algorithm of complexity $O(n\log^{d-1} n)$ solving the MST problem under consideration is proposed, which improves the existing algorithms for $d\geqslant 6$.

Keywords: computational geometry, minimum spanning tree problem, efficient algorithm.

UDC: 519.163

MSC: 05C85

Received: 12.05.2024

DOI: 10.4213/mzm14366


 English version:
Mathematical Notes, 2025, 117:5, 762–768

Bibliographic databases:


© Steklov Math. Inst. of RAS, 2026