The ideal circle pattern with boundaries via analytic method
A surface \(S\) is of topological type \((g,n)\) if \(S\) is an oriented compact surface of genus \(g\) with a boundary consisting of \(n\) disjoint simple closed curves. The Euler characteristic of \(S\) is given by \(\chi (S)=2-2g-n\). Let \(\mathcal{D}=(V,E,F)\) be a cellular decomposition of \(S\), where \(V\), \(E \), and \(F\) represent the sets of vertices, edges, and faces of \(\mathcal{D}\), respectively. A circle pattern \(P\) on \(S\) is a collection of disks \( \{D_{v}\}_{v\in V}\). \(P\) is a \(\mathcal{D}\)-effective circle pattern if, for every edge \([v_{i},v_{j}]\in E\), where \(v_{i},v_{j}\in V\), the corresponding disks \(D_{i}\) and \(D_{j}\) intersect. The intersection angle function \(\Theta :E\rightarrow \lbrack 0,\pi )\) of the circle pattern \(P\) is defined as: \( \Theta ([v_{i},v_{j}])=\Theta_{ij}\), the intersection angle between the disks \(D_{i}\) and \(D_{j}\). A circle pattern \(P\) is a \(\mathcal{D}\)-type circle pattern if there exists a geodesic cellular decomposition \(\mathcal{D} (\mu )\) satisfying the following conditions: (1) \(\mathcal{D}(\mu )\) is isotopic to \(\mathcal{D}\); (2) the set of vertices of \(\mathcal{D}(\mu )\) coincides with the set of centers of the disks in \(P\). An interstice is a connected component of \(S\setminus (\bigcup_{v\in V}\mathbb{D}_{v})\), where \( \mathbb{D}_{v}\) is the interior of \(D_{v}\). \(P\) is an ideal circle pattern if there is a one-to-one correspondence between the interstices in \(P\) and the cells in \(\mathcal{D}(\mu )\), and if every interstice contains exactly one point. \N\NFor an ideal circle pattern, the following conditions hold: (1) If \(e_{1}+\dots +e_{m}\) forms the boundary of a face \(f\), then \(\sum_{l=1}^{m}\Theta (e_{l})=(m-2)\pi \); (2) \(\Theta (e)\in (0,\pi )\) for every \( e\in E\). Let \(\sigma (v_{i})\) denote the sum of the cone angles at a primal vertex \(v_{i}\). The vertex curvature at \(v_{i}\) is defined as \( K_{i}=2-\sigma (v_{i})\); if \(v_{i}\in V\setminus V_{\partial }\), \(\pi -\sigma (v_{i})\), if \(v_{i}\in V_{\partial }\), with \(V_{\partial }=\partial S\cap V\). The generalized Gaussian curvature \(R_{i}\) is defined by \(R_{i}= \frac{K_{i}}{\sinh^{2}\frac{r_{i}}{2}}\), in the hyperbolic case, \(R_{i}=\frac{ K_{i}}{r_{i}^{2}}\), in the Euclidean case. The discrete Gaussian curvature map is defined as: \(TH(\cdot ,\cdot ):\mathbb{R}_{+}^{\left\vert V\right\vert }\times \lbrack 0,\pi )^{\left\vert E\right\vert }\rightarrow \mathbb{R}^{\left\vert V\right\vert }\), \(TH(r_{1},\dots ,r_{\left\vert V\right\vert },\Theta_{1},\dots ,\Theta_{\left\vert E\right\vert })=(K_{1},\dots ,K_{\left\vert V\right\vert })\), where \(r_{i}=r_{v_{i}}\) and \(\Theta_{i}=\Theta_{v_{i}}\).The generalized Gaussian curvature map \( GTH_{i}(r,\Theta )\) is defined by \(GTH_{i}(r,\Theta )=\frac{TH_{i}(r,\Theta ) }{\sinh^{2}\frac{r_{i}}{2}}\), in the hyperbolic case, \(GTH_{i}(r,\Theta )= \frac{TH_{i}(r,\Theta )}{r_{i}^{2}}\), in the Euclidean case. The generalized flow equations are \(\frac{dg_{i}}{dt}=-R_{i}g_{i}\), in the hyperbolic case, \( \frac{dg_{i}}{dt}=(R_{c}-R_{i})g_{i}\), in the Euclidean case, where \( g_{i}=\sinh^{2}\frac{r_{i}}{2}\), in the hyperbolic case, \(g_{i}=r_{i}^{2}\), in the Euclidean case, \(R_{c}\) being a constant, and the initial vector \( r(0)\in \mathbb{R}_{+}^{\left\vert V\right\vert }\). A curve in \(S\) is called pseudo-Jordan if \(\gamma \) is closed (not necessarily simple),and \( S\setminus \gamma \) contains a simply-connected component \(\mathbb{K}\) whose boundary \(\partial \mathbb{K}=\gamma \). Additionally, a curve \(\gamma_{2}\) in \(S\) is called semi pseudo-Jordan if there exists an open arc \(\lambda \subset \partial S\) such that \(\gamma_{2}\cup \lambda \) forms a pseudo-Jordan curve in \(S\). \N\NIn a hyperbolic background geometry, for a given surface \(S\) of topological type \((g,n)\), and a cellular decomposition \( \mathcal{D}=(V,E,F)\) of \(S\), if the intersection angle function \(\Theta :E\rightarrow (0,\pi )\) of the circle pattern \(P\) satisfies the condition: If \(e_{1}+\dots +e_{m}\) is the boundary of a face in \(\mathcal{D}\), then \( \sum_{l=1}^{m}\Theta (e_{l})=(m-2)\pi \), then the following six conditions are equivalent:\N\begin{itemize}\N\item[(D1)] The solution \(r(t)\) to the previous generalized flow equations exists for \(t\geq 0\), and as \(t\rightarrow +\infty \), \(r(t)\) converges;\N\item[(D2)] \((0,0,\dots,0)\) is an image point of the generalized Gaussian curvature mapping \(GTH(\cdot ,\Theta )\);\N\item[(D3)] If \(A\) is a non-empty subset of \(V\), then \(\pi \left\vert A\cap \partial S\right\vert +2\sum_{e\in E_{0}(A)}\Theta (e)+\sum_{e\in E_{\partial }(A)}\Theta (e)>2\pi \left\vert A\right\vert \);\N\item[(D4)] If \(A\) is a non-empty subset of \(V\), then \(\sum_{(e,f)\in Lk(A)}(\Theta (e)-\pi )+2\pi \chi (S(A)\setminus \partial S)+\pi \chi (S(A)\cap \partial S)<0\);\N\item[(D5)] The Euler characteristic number \(\chi (S)\) is negative. If \( e_{1}+\dots +e_{m}\) is a pseudo-Jordan curve and not a boundary of a face in \(\mathcal{D}\), then \(\sum_{i=1}^{m}\Theta (e_{i})<(m-2)\pi \), and if \( e_{1}+\dots +e_{m}\) is a semi pseudo-Jordan curve and does not form the boundary \(P\) of a face in \(\mathcal{D}\) with an open arc \(\gamma \subset \partial S\), we have \(\sum_{i=1}^{m}\Theta (e_{i})<(m-1)\pi \);\N\item[(D6)] There exists a metric \(\mu \) in \(S\) with constant curvature \(-1\), such that \((S,\mu )\) is compatible with a \(\mathcal{D}\)-type circle pattern \(P\), whose intersection angle function is \(\Theta \). \N\end{itemize}\N\NMeanwhile, the vertex curvature of the boundary vertex is 0, and the boundary of \((S,\mu )\) is a geodesic closed curve.\N\NThe second main result analyses the case of an Euclidean background geometry. \N\NFor the proofs, the authors recall properties of the Gaussian curvature, Brouwer invariance of domain theorem, and a generalized version of the results of \textit{A. I. Bobenko} and \textit{B. A. Springborn} in [Trans. Am. Math. Soc. 356, No. 2, 659--689 (2004; Zbl 1044.52009)] in the hyperbolic background geometry, among other tools.
- 3-dimensional combinatorial Yamabe flow in hyperbolic background geometry
- -curvatures and -flows on low dimensional triangulated manifolds
- A combinatorial Yamabe problem on two and three dimensional manifolds
- A Discrete Ricci Flow on Surfaces with Hyperbolic Background Geometry
- A note on circle patterns on surfaces
- A variational principle for circle packings.
- Characterizing the Delaunay decompositions of compact hyperbolic surfaces
- Circle patterns on surfaces of finite topological type
- Combinatorial Ricci flows and the hyperbolization of a class of compact 3-manifolds
- Combinatorial Ricci flows for ideal circle patterns
- Combinatorial Ricci flows on surfaces.
- Combinatorial Ricci flows with applications to the hyperbolization of cusped 3-manifolds
- Differential Topology
- Euclidean structures on simplicial surfaces and hyperbolic volume
- scientific article; zbMATH DE number 4084495 (Why is no real title available?)
- scientific article; zbMATH DE number 3308932 (Why is no real title available?)
- scientific article; zbMATH DE number 3323729 (Why is no real title available?)
- scientific article; zbMATH DE number 3027510 (Why is no real title available?)
- Kreispackungen und Triangulierungen. (Circle packings and triangulations)
- Matrix Analysis
- On a combinatorial curvature for surfaces with inversive distance circle packing metrics
- On the deformation of ball packings
- On the deformation of discrete conformal factors on surfaces
- On the deformation of inversive distance circle packings. II
- On the deformation of inversive distance circle packings. III
- Three-manifolds with positive Ricci curvature
- Variational principles for circle patterns and Koebe’s theorem
This page was built for publication: The ideal circle pattern with boundaries via analytic method
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6873074)
