Hexagonal Geometric Triangulations
Extracted source text ยท arXiv 2204.10791
\documentclass[11pt]{amsart}
\usepackage[top=1.4in, bottom=1.4in, left=1in, right=1in]{geometry}
%\usepackage{graphicx}
%\usepackage{amsmath,latexsym}
%\usepackage{amsfonts,amssymb}
%\usepackage{amscd, amsthm}
%\usepackage{stmaryrd}
%\usepackage{fancyhdr}
%\usepackage{commath,enumitem,chngcntr}
%\usepackage{indentfirst, hyperref}
%\usepackage[titletoc,toc,title]{appendix}
%\usepackage[]{algorithm2e}
%\usepackage{hyperref}
\usepackage{graphicx}
\usepackage{amsmath,latexsym}
\usepackage{amsfonts,amssymb}
\usepackage{amscd, amsthm}
\usepackage{stmaryrd}
\usepackage{fancyhdr}
\usepackage{commath,enumitem,chngcntr}
%\usepackage{indentfirst, hyperref}
\usepackage[titletoc,toc,title]{appendix}
\usepackage{bbm}
%\usepackage{refcheck}
\usepackage{pst-node}
\usepackage{tikz-cd}
\usepackage{pgf,tikz}
\usepackage{mathrsfs}
\usetikzlibrary{arrows,calc,positioning}
\usepackage{float}
%\usepackage{algorithm}
%\usepackage[noend]{algpseudocode}
\makeatletter
\def\BState{\State\hskip-\ALG@thistlm}
\makeatother
\setcounter{tocdepth}{3}
%\usepackage{showkeys}
%\usepackage{refcheck}
\usepackage{pst-node}
\usepackage{tikz-cd}
\usepackage{pgf,tikz}
\usepackage{mathrsfs}
\usetikzlibrary{arrows,calc,positioning}
\usepackage{float}
\DeclareGraphicsRule{.tif}{png}{.png}{`convert #1 `basename #1 .tif`.png}
\renewcommand{\theequation}{\thepart.\thesection.\arabic{equation}}
\newtheorem{theorem}{Theorem}[part]
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{proposition}[theorem]{Proposition}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem{definition}[theorem]{Definition}
\newtheorem{remark}[theorem]{Remark}
\newtheorem{example}[theorem]{Example}
\newtheorem{principle}[theorem]{Principle}
\newtheorem{conjecture}[theorem]{Conjecture}
%\counterwithin{figure}{section}
\numberwithin{equation}{subsection}
\numberwithin{theorem}{subsection}
\newcommand{\nf}{\normalfont}
%\pagestyle{fancy}
\fancyhf{}
\fancyhead[LE,RO]{\small Ian Tice (Scribe: Jingyang Shu)}
\fancyhead[RE,LO]{\small \leftmark}
\fancyfoot[C]{\thepage}
%\newcommand{\RN}[1]{\textup{\uppercase\expandafter{\romannumeral#1}}}
\newcommand{\A}{\mathcal{A}}
\newcommand{\C}{\mathbb{C}}
\newcommand{\D}{\mathbb{D}}
\newcommand{\F}{\mathcal{F}}
\newcommand{\N}{\mathbb{N}}
\newcommand{\p}{\mathbf{p}}
\newcommand{\q}{\mathbf{q}}
\newcommand{\R}{\mathbb{R}}
\newcommand{\T}{\mathbb{T}}
\newcommand{\Z}{\mathbb{Z}}
\newcommand{\dt}{\partial_t}
\newcommand{\diff}{\,\mathrm{d}}
\newcommand{\jump}[1]{\left\llbracket #1 \right\rrbracket}
\newcommand{\inner}[2]{\langle{#1}, {#2}\rangle}
\let\div\relax
\DeclareMathOperator{\sgn}{sgn}
\DeclareMathOperator{\div}{div}
\title{Hexagonal Geometric Triangulations}
\begin{document}
\author{Xiaoping Zhu}
\address{Department of Mathematics, Rutgers University, Piscataway, NJ, 08854}
\email{xz349@rutgers.edu}
%\subjclass[2000]{30C35, 53A55, 30C20, 30C62}
%\keywords{circle domains, convex hull, convex surfaces, hyperbolic and conformal geometries}
\begin{abstract}
It is well-known that the Euclidean plane has a standard 6-regular geodesic triangulation , and the unit sphere has a 5-regular geodesic triangulation, which is induced from the regular Dodecahedron, and the hyperbolic plane has an $n$-regular geodesic triangulation for any $n>6$.
Here we constructed a 6-regular geodesic triangulation of the hyperbolic plane.
\end{abstract}
\maketitle
\tableofcontents
\section{Introduction}
Here we consider the problem on the existence of special classes of geometric triangulations on the Euclidean plane and the hyperbolic plane.
A triangulation $T$ on a surface $S$ is \textit{$k$-regular} for some integer $k>0$ if the degrees of all its vertices are $k$, and it is \textit{geometric} if all the edges are embedded as geodesic arcs with respect to the metric. The most familiar geometric $6$-regular triangulation is the standard hexagonal triangulation on the plane, and the $7$-regular equilateral triangulation of the hyperbolic plane. The triangular groups \cite{martelli2016introduction}
$$\Delta(3, k) = \{ a, b, c | a^3 = b^k = c^2 = 1 \}$$
produce $k$-regular equilateral triangulations for any $k>6$.
\begin{figure}[!ht]
\includegraphics[width=0.3\linewidth]{order6.png}
\includegraphics[width=0.3\linewidth]{order7.png}
\includegraphics[width=0.3\linewidth]{order8.png}
\caption{Degree-regular geometric triangulations. }
\end{figure}
We ask whether we can construct $6$-regular geometric triangulations on the hyperbolic plane. Moreover, we can consider the general question as follows.\\
\textbf{Question}: Does there exist $k$-regular geometric triangulations on the Euclidean plane $\mathbb{E}^2$, the round 2-sphere $\mathbb{S}^2$ and the hyperbolic plane $\mathbb{H}^2$ for any $k\geq 3$? Furthermore, can we construct such a geometric triangulation $T$ with a uniform bound on the edge lengths of $T$?
If $k = 6$, the number of vertices grows linearly with respect to the combinatorial distance from one point. If $k>6$, the number of vertices grows exponentially. Intuitively, they correspond to the Euclidean and hyperbolic geometry. We show that we can construct $6$-regular geometric triangulation on $\mathbb{H}^2$.
%This answers a question proposed by Luo \cite{luo}.
The semi-regular tiling of the Euclidean and hyperbolic plane by polygons has been studied by Datta and Gupta \cite{datta2019semi}.
\section{Preliminary case: closed surfaces}
The basic combinatorial formulas below of Euler characteristics of closed surfaces show that there is no $k$-regular triangulation on $\mathbb{E}^2$ and $\mathbb{H}^2$ if $k = 3, 4, 5$, and there is no $k$-regular triangulation on $\mathbb{S}^2$ if $k\geq 6$.
From the combinatorial formula
$$12(1 - g) = 6\chi(S) = \sum_{i\in V} (6-d_i) = |V|(6-k),$$
we have a necessary condition for a closed orientable surface $S$ to admit a $k$-regular triangulation $T$, namely $$(k-6)|12(g-1).$$
It implies immediately that there is no $k$-regular triangulation on $\mathbb{S}^2$ if $k \geq 6$, and the only $k$-regular triangulation on tori is when $k = 6$. This condition turns out to be also sufficient for the existence of $k$-regular triangulation on $S$ with genus $g$.
\begin{lemma}
Let $S$ be a surface with genus $g$. Then there exists a $k$-regular triangulation on $S$ if and only if $$(k-6)|12(g-1).$$
\end{lemma}
\begin{proof}
The necessity is proved above. The sufficiency follows from the result by Jucovic and Trenkler \cite{jucovivc1973theorem}. First notice that if a $k$-regular triangulation $T$ exists on $S$, then set $V_0$, $E_0$, and $F_0$ be the number of vertices, edges, and faces of $T$. Then $V_0 = 6\chi(S)/(6-k)$, $E_0 = 3k\chi(S)/(6-k)$, and $F_0 = 2k\chi(S)/(6-k)$.
By the Main Theorem in \cite{jucovivc1973theorem}, if three positive numbers $V$, $E$, and $F$ on $S$ satisfies two conditions
\begin{enumerate}
\item $F - (4-k)V = 8(1-g)$;
\item $kV = 3F = 2E$,
\end{enumerate}
then a $k$-regular triangulation with $V$ vertices, $E$ edges and $F$ faces exists on $S$. It is straightforward to check that $V_0$, $E_0$, and $F_0$ satisfies the two conditions.
\end{proof}
\begin{lemma}
There is no $k$-regular triangulation on $\mathbb{E}^2$ and $\mathbb{H}^2$ if $k = 3, 4, 5$.
\end{lemma}
\begin{proof}
If such a triangulation $T$ exists, take an arbitrary subcomplex $T'$ which is homeomorphic disk. A combinatorial formula for a triangulated disk is given by
$$\sum_{i\in V_I} (6 - d_i) = 6 + \sum_{i \in V_B} (d_i - 4) $$
where $V_I$ and $V_B$ are the index sets for interior vertices and boundary vertices.
Since $T$ is $k$-regular for $k<6$, if $v_i$ is a boundary vertex of $T'$, then $d_i<5$ since $T'$ is part of $T$. Then the left side is no larger than $6$. Hence the disk $T'$ can only contains at most $6$ vertices. This leads to a contradiction because we can take $T'$ as large as we want.
\end{proof}
For $k = 3, 4, 5$, we have $k$-regular geometric triangulations on $\mathbb{S}^2$, given by projecting regular tetrahedron, octahedron, and icosahedron to the unit sphere from the center. It is straightforward to construct a $6$-regular triangulation on a torus using the standard hexagonal triangulation on $\mathbb{E}^2$.
\begin{figure}[H]
\includegraphics[width=0.25\linewidth]{order3}
\includegraphics[width=0.25\linewidth]{order4}
\includegraphics[width=0.25\linewidth]{order5}
\caption{Degree-regular geometric triangulations on the unit 2-sphere.}
\end{figure}
\section{General facts about regular geometric triangulations}
The goal is to construct $k$-regular geometric triangulations on $\mathbb{E}^2$ with $k>6$, and $6$-regular geometric triangulations on $\mathbb{H}^2$. We first point out that such triangulations contain skinny triangles.
\begin{lemma}
Let $T$ be a $k$-regular geometric triangulations on $\mathbb{E}^2$ with $k>6$, or a $6$-regular geometric triangulations on $\mathbb{H}^2$. Then the infimum of the angles in $T$ is zero.
\end{lemma}
\begin{proof}
If not, in the first case for $\mathbb{E}^2$ we can construct a quasiconformal homeomorphism from $\mathbb{E}^2$ to $\mathbb{H}^2$ by sending a triangle in $T$ to the corresponding triangle in the standard geometric triangulation of $\mathbb{H}^2$ generated by reflecting equilateral triangles with inner angle $2\pi/k$. Similarly, nn the second case we can construct a quasiconformal homeomorphism from $\mathbb{H}^2$ to $\mathbb{E}^2$ sending $T$ to the standard equilateral hexagonal triangulation. This contradicts to the fact that there is no quasiconformal map from $\mathbb{E}^2$ to $\mathbb{H}^2$.
\end{proof}
In this sense, the ``best" $k$-regular geometric triangulation we can expect is a geometric triangulation with a uniform low bound $c>0$ and a uniform upper $C$ on the lengths of edges.
\begin{conjecture}
There is no $k$-regular geometric triangulation on $\mathbb{E}^2$ with a uniform low bound $c>0$ and a uniform upper $C$ on the lengths of edges for $k \geq 7$. Similarly, there is no $6$-regular geometric triangulation on $\mathbb{H}^2$ with a uniform low bound $c>0$ and a uniform upper $C$ on the lengths of edges.
\end{conjecture}
\section{k-Regular geometric triangulations on The Euclidean plane}
It is known that one can construct $k$-regular geometric triangulation on $\mathbb{E}^2$ for $k>6$.
The idea is to construct chains of circles with radius $n$ and distribute the points on the circle evenly. The similar computation is given in \cite{datta2019semi} . We can show that these triangulations have the desired uniform bounds.
\begin{lemma}
For any $k\geq 6$, there exist $k$-regular geometric triangulations on $\mathbb{E}^2$ with uniform upper bounds on the lengths of edges.
\end{lemma}
\begin{proof}
The construction of a $k$-regular geometric triangulation is given above. Let $C_n$ be circles with radius $r_n = n$ centered at the origin of $\mathbb{E}^2$. Let $a_n$ be the number of vertices on $C_n$ on this construction. These vertices have combinatorial distance $n$ from the vertex at the origin.
We have the following recursive relation:
$$a_{n+1} = (k-4)a_n - a_{n-1}$$
with $a_0 = 0$ and $a_1 = k$.
Solve this series and we have the formula for $a_n$ with $n\geq 1$:
$$a_n = k\frac{\alpha^n - \beta^n}{\alpha - \beta}$$
where $\alpha = \cosh^{-1}((k-4)/2)$ and $\beta = 1/\alpha$. Thus, asymptotically $a_n$ grows as $\alpha^{n-1}$ to infinity. But the length of $C_n$ is given by $l_n = 2\pi n$. So the arclength of each small arc goes to zero as $n$ goes to infinity, hence the length between two consecutive vertices on $C_n$ goes to zero.
By the construction above, if $v$ is a vertex in $C_n$, then the $k-4$ or $k-3$ vertices $v_i$ in $C_{n-1}$ connecting to $v$ forms an arc $\gamma$ in $C_{n+1}$, and the ray starting from the origin passing through $v$ intersects with $\gamma$. As the length of $\gamma$ goes to zero, all the lengths of the edges connecting $v$ to $v_i$ goes to $r_{n+1} - r_n = 1$.
So as $n\to \infty$, the lengths of edges approaches either $1$ or $0$. Hence we can take the maximal length of a compact part of $T$ bounded by $C_n$ with $n$ large enough so that its complement contains edges no longer than $2$. Then we can find a uniform bound.
\end{proof}
\begin{figure}[!ht]
\includegraphics[width=0.4\linewidth]{Eorder7}
\caption{A 7-regular geometric triangulations on the Euclidean plane.}
\end{figure}
Using the same idea, we can construct a $k$-regular geometric triangulation of $\mathbb{E}^2$ whose edges has a uniform lower bound. We can pick $r_n = 2^n$ and follow the construction above.
%\section{Natural Approach to construct 6-Regular triangulations on the hyperbolic plane}
%In this section, we first introduce two natural approaches to construct $6$-regular geometric triangulations on $\mathbb{H}^2$ by mimicking the previous construction. We show that this construction does not generate geometric triangulation.
%
%Specifically, we first position one vertex at the origin in the Poincare disk model. Then we choose a chain of circles $C_n$ centering at the origin with radius $r_n$ (in terms of hyperbolic length). Next, we distribute vertices which have combinatorial distance $n$ from the origin on $C_n$ evenly and connect vertices in $C_n$ with $C_{n-1}$ so as to satisfy the degree requirement.
%
%Notice that instead of the exponential growth of $a_n$ for the $k$-regular triangulation with $k>6$, the number of vertices on $C_n$, denoted by $b_n$, grows linearly. By a simple computation we have for $n >0$,
%$$b_n = 6n.$$
%
%\subsubsection{Case 1: $r_n = n$}
%We first assume that $r_n = n$. The perimeter $l_n$ of $C_n$ is given by $l_n = 2\pi \sinh n$. Hence the $l_n$ grows exponentially while $b_n = 6n$, so the arclength between two consecutive vertices on $C_n$ goes to infinity. By the hyperbolic cosine law, the corresponding edge length $x$ between these two vertices is given by
%$$\cosh x = \cosh^2 n - \sinh^2 n \cos \frac{\pi}{3n} = 1 + \sinh^2 n(1 - \cos \frac{\pi}{3n})\sim \frac{e^n}{n^2} \to \infty.$$
%
%Similarly, any edges connecting vertices in $C_n$ and $C_{n-1}$ is contained in the annul bounded by $C_n$ and $C_{n+1}$. Clearly these edges have lengths $l \geq r_{n+1} - r_n = 1$, and they are also bounded by $\cosh^{-1}(\cosh(n+1)/\cosh n)$. Notice that
%$$\lim_{n\to \infty} \frac{\cosh (n+1)}{\cosh n} = e,$$
%hence all the edges connecting vertices on $C_n$ and $C_{n+1}$ is bounded above. This shows that the triangle inequality will be violated as $n\to \infty$. Hence this construction will not lead to a $6$-regular geometric triangulation on $\mathbb{H}^2$.
%
%
%
%\subsubsection{Case 2: $r_n = \log n$.}
%The previous case shows that $r_n$ can not grow too fast. Here we will show that $r_n$ can not be too slow. The perimeter $l_n$ of $C_n$ is given by $l_n = 2\pi \sinh \log n = \pi(n - 1/n)$. By the same argument, since $b_n = 6n$, the arclength between two consecutive vertices on $C_n$ tends to $\pi/6$.
%
%On the other hand, any edges connecting vertices in $C_n$ and $C_{n-1}$ is again contained in the annul bounded by $C_n$ and $C_{n+1}$. In this case the upper bound is given by $\cosh^{-1}(\cosh(\log(n+1))/\cosh \log n)$. Since
%$$\lim_{n\to \infty} \frac{\cosh \log (n+1)}{\cosh \log n} = 1, $$
%the lengths of the edges connecting vertices in $C_n$ and $C_{n-1}$ tends to $\cosh^{-1}(1) = 0$. This will again leads to a contradiction to the triangle inequality.
%Generalize these two examples and we can show that the growth of $r_n$ should be faster than any power of $\log n$ and slower than any power of $n$ for triangle inequality to be held.
\newpage
\section{6-regular geometric triangulations on the Hyperbolic Plane}
We now give a construction of the desired $6$-regular geometric triangulations with uniform bound in the hyperbolic plane using the Klein disk model.
\begin{figure}[!t]
\includegraphics[width=0.7\linewidth]{triangulation.jpg}
\caption{Tianqi's Picture. }
\end{figure}
\begin{enumerate}
\item Put one vertex $v_0$ in the origin, and shoot out two rays $R_1$ and $R_2$ along the positive $x$-axis and along the direction $e^{i\pi/3}$.
\item Let $r_n = \alpha \log n$ with $0<\alpha<1/2$. Construct a sequence of points $a_n$ and $b_n$ on $R_1$ and $R_2$ with distance $r_n$ from $v_0$ in hyperbolic metric. Connect $a_n$ with $a_{n+1}$ and $b_n$ with $b_{n+1}$ using geodesics. Call these edges of \textit{type 0}.
\item Connect $a_n$ with $b_n$ by circle arcs $C_n$, called the $n$-th layer in this construction. Equally distribute $n-1$ points on $C_n$. Then each $C_n$ are divided to $n$ arcs. Connect the consecutive vertices on $C_n$ to generate $n$ edges. Call these edges of \textit{type 1}. By symmetry, these edge have the same length.
\item Connect points on $C_n$ and $C_{n+1}$ based on the combinatorics of the hexagonal triangulation to generate $2(n+1)$ edges. Call these edges of \textit{type 2}.
\item Rotate the configuration by $\pi/3$ to generate the full hexagon geometric triangulation.
\end{enumerate}
We can give explicit coordinates of the vertices on $C_n$ in this triangulation in the plane
$$v_n^k = \tanh r_ne^{i\frac{k\pi}{3n}}, \quad k = 0, 1, \cdots, n.$$
And the vertex $v_n^k$ is connected with vertices $v_{n+1}^k$ and $v_{n+1}^{k+1}$ in this triangulation.
\begin{proposition}
\ The construction above generates a hexagonal geometric triangulation of $\mathbb{H}^2$ with uniform bound $K>0$ on the lengths of edges.
\end{proposition}
The proof of this proposition consists of two lemmas.
\begin{lemma}
The construction above generates a valid geometric triangulations with no intersection of edges and degenerate triangles.
\end{lemma}
\begin{proof}
The proof is based on induction on $n$. The base case $n = 1$ is trivial. Assume it is true for the triangulation generated up to the $n$-th layer. We will show that the triangles between $C_n$ and $C_{n+1}$ lie in the ring $R_n$ bounded $C_n$ and $C_{n+1}$. There are two types of triangles:
\begin{enumerate}
\item Type 1: two vertices on $C_n$ and one vertex on $C_{n+1}$. The three vertices are
$$v_n^k =\tanh r_ne^{i\frac{k\pi}{3n}}, v_n^{k+1} = \tanh r_ne^{i\frac{(k+1)\pi}{3n}}, \text{and }v_{n+1}^{k+1} =\tanh r_{n+1}e^{i\frac{(k+1)\pi}{3(n+1)}}.$$
Notice that
$$\frac{k}{n} < \frac{k+1}{n+1} < \frac{k+1}{n}, \text{and } r_{n}\leq r_{n+1},$$
\item Type 2: one vertex on $C_n$ and two vertex on $C_{n+1}$. In this case, the three vertices are
$$v_n^k =\tanh r_ne^{i\frac{k\pi}{3n}}, v_{n+1}^{k+1} = \tanh r_{n+1}e^{i\frac{(k+1)\pi}{3n}}, \text{and }v_{n+1}^{k} = \tanh r_{n+1}e^{i\frac{k\pi}{3(n+1)}}.$$
Notice that
$$\frac{k}{n+1} \leq \frac{k}{n} \leq \frac{k+1}{n+1}, \text{and } r_{n}\leq r_{n+1}.$$
\end{enumerate}
{We want to show that the edges of type 2 lie in the ring bounded by two consecutive $C_n$ and $C_{n+1}$, so it does not intersect with previous layers.} If $r_n$ increase too slowly, then it is possible that the edge determined by $v_{n+1}^{k}$ and $v_{n+1}^{k+1}$ might intersect with $v_n^k$. We need to show that if $r_n = \alpha\log n$ with $0<\alpha<1/2$, this will not occur.
% For simplicity, rotate the triangulation so that $v_n^k$ lies on the positive $x$-axis, then $v_{n+1}^k$ lies on the closed lower half-plane, and $v_{n+1}^{k+1}$ lies on the closed upper half-plane.
%
% The shortest distance between $v_n^k$ and points on the edge of $v_{n+1}^kv_{n+1}^{k+1}$ is given in the following symmetric figure, where the intersection of the two chords is $v_n^k$.
%
The critical figures for triangles of type 1 and type 2 are given below, where the angles at $v_n^k$ or $v_{n}^{k+1}$ are right angles.
\begin{figure}[H]
\includegraphics[width=0.5\linewidth]{critical1}
\includegraphics[width=0.5\linewidth]{critical2}
\caption{Critical Cases for Two Types of Triangles}
\end{figure}
In this case, we can show that for sufficiently large $n>N$,
$$\tanh r_n < \tanh r_{n+1} \cos \frac{\pi}{6(n+1)},$$
and
$$\tanh r_n < \tanh r_{n+1} \cos \frac{\pi}{6n},$$
which is equivalent to
$$\sinh(r_{n+1} - r_n) > \cosh r_n\sinh r_{n+1} (1 - \cos \frac{\pi}{6n}).$$
One can show that the left side has order $O(1/n)$ and right side has order $O(1/n^{2-2\alpha})$, so the inequality holds for large $n$ if $0<\alpha<1/2$.
This means that for large $n>N$, the triangles of type 2 will not intersect with the previous layers.
For $0\leq n\leq N$, we can treat it as a polygon in the Euclidean plane, and adjust the radius to avoid the intersections.
\end{proof}
\begin{lemma}
The lengths of edges of the geometric triangulation above are uniformly bounded.
\end{lemma}
\begin{proof}
The length of the edge of type 1 is bounded by the length of the arc of the circle connecting $v_n^k$ and $v_n^{k+1}$ given by
$$\sinh |v_n^kv_n^{k+1}| < \frac{\pi}{3n}\sinh r_n \sim O(\frac{1}{n^{1-\alpha}}) \to 0.$$
From the previous lemma, all the edges of type 0 and type 2 in the $n$-th layer lie in the ring $R_n$. The length of the edges in $R_n$ with vertices on $C_n$ is bounded by the following situation
\begin{figure}[!h]
\includegraphics[width=0.5\linewidth]{estimate}
\caption{Estimate for Edges of Type 2. }
\end{figure}
The half length $L$ of this edge is given by hyperbolic Pythagorean theorem
$$L = 2\cosh^{-1}(\frac{\cosh r_{n+1}}{\cosh r_n}) \leq 2\cosh^{-1}(2(1 + \frac{1}{n})^\alpha) < \infty.$$
Hence we can bound the edges of all the three types for large $n>0$. The remaining part is naturally bounded by compactness.
\end{proof}
\section{Related work}
The existence of certain types of geodesic triangulations is a fundamental question. Recently various theory upon geodesic triangle meshes were developed, where people discuss problems about rigidity, convergence, variational principles, discrete maps and discrete geometric structures. See \cite{gaster2018computing}\cite{gaster2019computing}\cite{hass2012simplicial}\cite{luo2021deformation}\cite{luo2021deformation2}\cite{luo2022deformation}\cite{luo2004combinatorial}\cite{bobenko2015discrete}\cite{gu2018discrete}\cite{gu2018discrete2}\cite{sun2015discrete}\cite{gu2019convergence}\cite{luo2020discrete}\cite{wu2015rigidity}\cite{wu2014finiteness}\cite{wu2020convergence}\cite{luo2019koebe}\cite{luo2021convergence}.
for example.
\bibliography{1}
\bibliographystyle{amsplain}
\end{document}
\providecommand{\bysame}{\leavevmode\hbox to3em{\hrulefill}\thinspace}
\providecommand{\MR}{\relax\ifhmode\unskip\space\fi MR }
% \MRhref is called by the amsart/book/proc definition of \MR.
\providecommand{\MRhref}[2]{%
\href{http://www.ams.org/mathscinet-getitem?mr=#1}{#2}
}
\providecommand{\href}[2]{#2}
\begin{thebibliography}{10}
\bibitem{bobenko2015discrete}
Alexander~I Bobenko, Ulrich Pinkall, and Boris~A Springborn, \emph{Discrete
conformal maps and ideal hyperbolic polyhedra}, Geometry \& Topology
\textbf{19} (2015), no.~4, 2155--2215.
\bibitem{datta2019semi}
Basudeb Datta and Subhojoy Gupta, \emph{Semi-regular tilings of the hyperbolic
plane}, Discrete \& Computational Geometry \textbf{65} (2021), no.~2,
531--553.
\bibitem{gaster2018computing}
Jonah Gaster, Brice Loustau, and L{\'e}onard Monsaingeon, \emph{Computing
discrete equivariant harmonic maps}, arXiv preprint arXiv:1810.11932 (2018).
\bibitem{gaster2019computing}
\bysame, \emph{Computing harmonic maps between riemannian manifolds}, arXiv
preprint arXiv:1910.08176 (2019).
\bibitem{gu2019convergence}
David Gu, Feng Luo, and Tianqi Wu, \emph{Convergence of discrete conformal
geometry and computation of uniformization maps}, Asian Journal of
Mathematics \textbf{23} (2019), no.~1, 21--34.
\bibitem{gu2018discrete2}
Xianfeng Gu, Ren Guo, Feng Luo, Jian Sun, and Tianqi Wu, \emph{A discrete
uniformization theorem for polyhedral surfaces ii}, Journal of differential
geometry \textbf{109} (2018), no.~3, 431--466.
\bibitem{gu2018discrete}
Xianfeng~David Gu, Feng Luo, Jian Sun, and Tianqi Wu, \emph{A discrete
uniformization theorem for polyhedral surfaces}, Journal of Differential
Geometry \textbf{109} (2018), no.~2, 223--256.
\bibitem{hass2012simplicial}
Joel Hass and Peter Scott, \emph{Simplicial energy and simplicial harmonic
maps}, arXiv preprint arXiv:1206.2574 (2012).
\bibitem{jucovivc1973theorem}
Ernest Jucovi{\v{c}} and Mari{\'a}n Trenkler, \emph{A theorem on the structure
of cell--decompositions of orientable 2--manifolds}, Mathematika \textbf{20}
(1973), no.~1, 63--82.
\bibitem{luo2004combinatorial}
Feng Luo, \emph{Combinatorial yamabe flow on surfaces}, Communications in
Contemporary Mathematics \textbf{6} (2004), no.~05, 765--780.
\bibitem{luo2020discrete}
Feng Luo, Jian Sun, and Tianqi Wu, \emph{Discrete conformal geometry of
polyhedral surfaces and its convergence}, arXiv preprint arXiv:2009.12706
(2020).
\bibitem{luo2019koebe}
Feng Luo and Tianqi Wu, \emph{Koebe conjecture and the weyl problem for convex
surfaces in hyperbolic 3-space}, arXiv preprint arXiv:1910.08001 (2019).
\bibitem{luo2021convergence}
Yanwen Luo, Tianqi Wu, and Xiaoping Zhu, \emph{The convergence of discrete
uniformizations for genus zero surfaces}, arXiv preprint arXiv:2110.08208
(2021).
\bibitem{luo2021deformation}
\bysame, \emph{The deformation space of geodesic triangulations and generalized
tutte's embedding theorem}, arXiv preprint arXiv:2105.00612 (2021).
\bibitem{luo2021deformation2}
\bysame, \emph{The deformation spaces of geodesic triangulations of flat tori},
arXiv preprint arXiv:2107.05159 (2021).
\bibitem{luo2022deformation}
\bysame, \emph{The deformation space of delaunay triangulations of the sphere},
arXiv preprint arXiv:2202.06402 (2022).
\bibitem{martelli2016introduction}
Bruno Martelli, \emph{An introduction to geometric topology}, arXiv preprint
arXiv:1610.02592 (2016).
\bibitem{sun2015discrete}
Jian Sun, Tianqi Wu, Xianfeng Gu, and Feng Luo, \emph{Discrete conformal
deformation: algorithm and experiments}, SIAM Journal on Imaging Sciences
\textbf{8} (2015), no.~3, 1421--1456.
\bibitem{wu2014finiteness}
Tianqi Wu, \emph{Finiteness of switches in discrete yamabe flow}, Ph.D. thesis,
Master Thesis, Tsinghua University, Beijing, 2014.
\bibitem{wu2015rigidity}
Tianqi Wu, Xianfeng Gu, and Jian Sun, \emph{Rigidity of infinite hexagonal
triangulation of the plane}, Transactions of the American Mathematical
Society \textbf{367} (2015), no.~9, 6539--6555.
\bibitem{wu2020convergence}
Tianqi Wu and Xiaoping Zhu, \emph{The convergence of discrete uniformizations
for closed surfaces}, arXiv preprint arXiv:2008.06744 (2020).
\end{thebibliography}
\providecommand{\bysame}{\leavevmode\hbox to3em{\hrulefill}\thinspace}
\providecommand{\MR}{\relax\ifhmode\unskip\space\fi MR }
% \MRhref is called by the amsart/book/proc definition of \MR.
\providecommand{\MRhref}[2]{%
\href{http://www.ams.org/mathscinet-getitem?mr=#1}{#2}
}
\providecommand{\href}[2]{#2}
\begin{thebibliography}{10}
\bibitem{bobenko2015discrete}
Alexander~I Bobenko, Ulrich Pinkall, and Boris~A Springborn, \emph{Discrete
conformal maps and ideal hyperbolic polyhedra}, Geometry \& Topology
\textbf{19} (2015), no.~4, 2155--2215.
\bibitem{datta2019semi}
Basudeb Datta and Subhojoy Gupta, \emph{Semi-regular tilings of the hyperbolic
plane}, Discrete \& Computational Geometry \textbf{65} (2021), no.~2,
531--553.
\bibitem{gaster2018computing}
Jonah Gaster, Brice Loustau, and L{\'e}onard Monsaingeon, \emph{Computing
discrete equivariant harmonic maps}, arXiv preprint arXiv:1810.11932 (2018).
\bibitem{gaster2019computing}
\bysame, \emph{Computing harmonic maps between riemannian manifolds}, arXiv
preprint arXiv:1910.08176 (2019).
\bibitem{gu2019convergence}
David Gu, Feng Luo, and Tianqi Wu, \emph{Convergence of discrete conformal
geometry and computation of uniformization maps}, Asian Journal of
Mathematics \textbf{23} (2019), no.~1, 21--34.
\bibitem{gu2018discrete2}
Xianfeng Gu, Ren Guo, Feng Luo, Jian Sun, and Tianqi Wu, \emph{A discrete
uniformization theorem for polyhedral surfaces ii}, Journal of differential
geometry \textbf{109} (2018), no.~3, 431--466.
\bibitem{gu2018discrete}
Xianfeng~David Gu, Feng Luo, Jian Sun, and Tianqi Wu, \emph{A discrete
uniformization theorem for polyhedral surfaces}, Journal of Differential
Geometry \textbf{109} (2018), no.~2, 223--256.
\bibitem{hass2012simplicial}
Joel Hass and Peter Scott, \emph{Simplicial energy and simplicial harmonic
maps}, arXiv preprint arXiv:1206.2574 (2012).
\bibitem{jucovivc1973theorem}
Ernest Jucovi{\v{c}} and Mari{\'a}n Trenkler, \emph{A theorem on the structure
of cell--decompositions of orientable 2--manifolds}, Mathematika \textbf{20}
(1973), no.~1, 63--82.
\bibitem{luo2004combinatorial}
Feng Luo, \emph{Combinatorial yamabe flow on surfaces}, Communications in
Contemporary Mathematics \textbf{6} (2004), no.~05, 765--780.
\bibitem{luo2020discrete}
Feng Luo, Jian Sun, and Tianqi Wu, \emph{Discrete conformal geometry of
polyhedral surfaces and its convergence}, arXiv preprint arXiv:2009.12706
(2020).
\bibitem{luo2019koebe}
Feng Luo and Tianqi Wu, \emph{Koebe conjecture and the weyl problem for convex
surfaces in hyperbolic 3-space}, arXiv preprint arXiv:1910.08001 (2019).
\bibitem{luo2021convergence}
Yanwen Luo, Tianqi Wu, and Xiaoping Zhu, \emph{The convergence of discrete
uniformizations for genus zero surfaces}, arXiv preprint arXiv:2110.08208
(2021).
\bibitem{luo2021deformation}
\bysame, \emph{The deformation space of geodesic triangulations and generalized
tutte's embedding theorem}, arXiv preprint arXiv:2105.00612 (2021).
\bibitem{luo2021deformation2}
\bysame, \emph{The deformation spaces of geodesic triangulations of flat tori},
arXiv preprint arXiv:2107.05159 (2021).
\bibitem{luo2022deformation}
\bysame, \emph{The deformation space of delaunay triangulations of the sphere},
arXiv preprint arXiv:2202.06402 (2022).
\bibitem{martelli2016introduction}
Bruno Martelli, \emph{An introduction to geometric topology}, arXiv preprint
arXiv:1610.02592 (2016).
\bibitem{sun2015discrete}
Jian Sun, Tianqi Wu, Xianfeng Gu, and Feng Luo, \emph{Discrete conformal
deformation: algorithm and experiments}, SIAM Journal on Imaging Sciences
\textbf{8} (2015), no.~3, 1421--1456.
\bibitem{wu2014finiteness}
Tianqi Wu, \emph{Finiteness of switches in discrete yamabe flow}, Ph.D. thesis,
Master Thesis, Tsinghua University, Beijing, 2014.
\bibitem{wu2015rigidity}
Tianqi Wu, Xianfeng Gu, and Jian Sun, \emph{Rigidity of infinite hexagonal
triangulation of the plane}, Transactions of the American Mathematical
Society \textbf{367} (2015), no.~9, 6539--6555.
\bibitem{wu2020convergence}
Tianqi Wu and Xiaoping Zhu, \emph{The convergence of discrete uniformizations
for closed surfaces}, arXiv preprint arXiv:2008.06744 (2020).
\end{thebibliography}