A necessary condition for the logarithmic Minkowksi problem in higher dimension

Mijia Lai, Zixiao Wang · 2026

Read A necessary condition for the logarithmic Minkowksi problem in higher dimension by Mijia Lai, Zixiao Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-18167.txt.

Introduction

Mijia Lai and Zixiao Wang establish a necessary condition for the logarithmic Minkowski problem in higher dimensions. Their 2026 preprint extends a condition first proposed for the two-dimensional case by Liu, Lu, Sun, and Xiong. The new result also refines the subspace concentration condition. The Minkowski problem asks when a given measure comes from the surface of a convex body.

Catalog record

CREATORMijia Lai, Zixiao Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-18167

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A necessary condition for the logarithmic Minkowksi problem in higher dimension Thanks: This paper grows out from a research project in the National Scientific and Technological Innovation Talent Training Program of Shanghai Municipality. Mijia Lai, Zixiao Wang Abstract . In this paper, we establish a necessary condition for the logarithmic Minkowski problem in higher dimensions. This result generalizes a necessary condition proposed by Liu, Lu, Sun, and Xiong in their investigation of the two-dimensional case, and also refines the so-called subspace concentration condition. 1. Introduction The classical Minkowski problem, dating back to 1897 by Hermann Minkowski [ 15 ] , asks for the existence of a strictly convex closed surface S in \mathbb{R}^{3} whose Gaussian curvature, when expressed as a function of the unit normal, coincides with a prescribed positive function f on the sphere \mathbb{S}^{2} . Reflecting on its profound influence, Calabi once remarked that ”from the geometric viewpoint, it [the Minkowski problem] is the Rosetta Stone, from which several related problems can be solved.” In its modern, measure-theoretic formulation, the Minkowski problem seeks necessary and sufficient conditions for a given Borel measure \mu on \mathbb{S}^{n-1} to be the surface area measure of a convex body K\subset\mathbb{R}^{n} . For a convex body K with boundary \partial K , denote by \nu:\partial^{\prime}K\rightarrow\mathbb{S}^{n-1} the Gauss map, which is defined a.e.\mathcal{H}^{n-1} on \partial K where the normal is unique. The surface area measure \mu_{K} of K is then defined on Borel subsets E\subset\mathbb{S}^{n-1} by \mu_{K}(E):=\mathcal{H}^{n-1}\left(\nu^{-1}(E)\right), where \mathcal{H}^{n-1} denotes the (n-1) -dimensional Hausdorff measure. Minkowski himself first solved the polytopal case of the problem [ 15 ] . Let n_{1},\ldots,n_{m} be distinct, non-coplanar unit vectors in \mathbb{R}^{3} and a_{1},\ldots,a_{m}>0 be positive real numbers. Then there exists a closed convex polyhedral surface in \mathbb{R}^{3} , unique up to translation, whose faces have precisely these vectors as outer unit normals and these numbers as their respective surface areas, if and only if \sum_{i=1}^{m}a_{i}n_{i}=0. The condition \sum_{i}a_{i}n_{i}=0 admits a clear geometric interpretation: the total signed area of the projection of any closed polyhedral surface onto an arbitrary plane must be zero. Subsequently, Bonnesen-Fenchel [ 2 ] and Alexandrov [ 1 ] extended the existence and uniqueness result to general convex hypersurfaces in all dimensions. Later, the regularity of solutions-especially the question of when a weak (generalized) convex surface is in fact smooth-was established through seminal contributions of (among many others) Lewy [ 11 ] , Nirenberg [ 16 ] , Cheng-Yau [ 6 ] and Pogorelov [ 17 ] . Lutwak [ 14 ] , building on earlier ideas of Firey [ 8 ] , introduced the notion of L_{p} surface area measure for a general convex body K\subset\mathbb{R}^{n} . For a Borel set E\subset\mathbb{S}^{n-1} , the L_{p} surface area measure \mu_{(K,p)} is defined by \mu_{(K,p)}(E):=\int_{\nu_{K}^{-1}(E)}\left(x\cdot\nu_{K}(x)\right)^{1-p}d\mathcal{H}^{n-1}(x). Note that when p=1 , \mu_{(K,1)} is just the surface area measure of K mentioned above. Analogous to the classical Minkowski problem, the L_{p} Minkowski problem asks for necessary and sufficient conditions on a given Borel measure \mu on \mathbb{S}^{n-1} to be the L_{p} surface area measure of some convex body. In the past two decades, significant progress has been made on this problem-see, for example, the foundational works of Lutwak [ 14 ] , Chou-Wang [ 7 ] , Lutwak-Yang-Zhang [ 13 ] , and Böröczky-Lutwak-Yang-Zhang [ 4 ] . The reader is also referred to the excellent survey by Huang-Yang-Zhang [ 10 ] on the theory of Minkowski problems for variety of geometric measures including the above mentioned L_{p} surface area measure. An important special case of the L_{p} Minkowski problem arises when p=0 , known as the logarithmic Minkowski problem. The associated L_{0} surface measure-commonly referred to as the cone-volume measure (up to a factor of \frac{1}{n} )-is geometrically natural. For a polytope P containing the origin in its interior, the cone-volume measure of a face is simply the volume of the convex hull of that face and the origin. In a seminal work [ 4 ] , Böröczky-Lutwak-Yang-Zhang established that the logarithmic Minkowski problem admits a solution for an even measure if and only if the measure satisfies the subspace concentration condition. Later, Chen-Li-Zhu [ 5 ] extended this result by proving that the same subspace concentration condition remains sufficient for general (non-symmetric) measures. The discrete version of the logarithmic Minkowski problem is to prescribe a disrete cone volume measure of the form \mu=\sum_{i=1}^{m}v_{i}\delta_{n_{i}} . Geometrically, this amounts to finding necessary and sufficient conditions on a set of unit vectors n_{1},\dots,n_{m}\in\mathbb{R}^{n} and a set of positive numbers v_{1},\cdots,v_{m}>0 that ensure the existence of an m -faced polytope P , containing the origin in its interior, whose faces have outer unit normals n_{1},\cdots,n_{m} and corresponding cone-volumes v_{1},\cdots,v_{m} . Stancu [ 18 ] investigated the discrete logarithmic Minkowski problem in \mathbb{R}^{2} , i.e., the case of polygons. She established a general existence result for n -gons that have no pair of parallel sides. Additionally, she provided a sufficient condition for existence in two further cases: either when n>4 and the polygon has at least two pairs of parallel sides, or when the polygon is a trapezoid. Zhu [ 20 ] solved the discrete logarithmic Minkowski problem for measures whose supports are in general position. Here, a set of unit vectors in \mathbb{R}^{n} is said to be in general position if they are not contained in any closed hemisphere and any n of them are linearly independent. Later, Böröczky-Hegedűs-Zhu [ 3 ] proved that the essential subspace concentration condition is also sufficient for the existence of solutions to the discrete logarithmic Minkowski problem in \mathbb{R}^{n} . Liu-Lu-Sun-Xiong [ 12 ] recently discovered a necessary condition for two dimensional convex bodies whose centroid lies at the origin. They also completely solved the logarithmic Minkowski problem for quadrilaterals. The main result of the present paper is to extend above-mentioned Liu-Lu-Sun-Xiong’s condition to higher dimensions. Theorem 1 . Let K\subset\mathbb{R}^{n},(n\geq 3) be a convex body with centroid at the origin , \mu be the associated cone volume measure. Set V:=\mu(\mathbb{S}^{n-1})=V(K) , then for any u\in\mathbb{S}^{n-1} , there holds (1.1) \displaystyle n\frac{\mu(u)+\mu(-u)}{V}+(n+1)^{n-1}|\frac{\mu(u)-\mu(-u)}{V}|^{n}\leq 1. Moreover if the equality holds for some u , then K is • either an oblique prism of the form K=\operatorname{conv}\{P\cup P+\vec{l}\} , • or a cone of the form K=\operatorname{conv}\{P\cup\{p\}\} . Here P is an (n-1) -dimensional convex body lying in an (n-1) -dimensional hyperplane H perpendicular to u , and \vec{l} is a translation vector and p a point both not contained in H . \operatorname{conv}\{A\} represents the convex hull of A , i.e., the smallest convex set containing A . Figure 1. Left: oblique prism, Right: cone We make some remarks on our condition. First, for n=2 , condition ( 1.1 ) coincides with that of Liu–Lu–Sun–Xiong. Moreover, if the equality holds for some u , then K is either a trapezoid (including a parallelogram or rectangle as a special case) with its two parallel sides perpendicular to u , or a triangle with one side perpendicular to u . This situation differs slightly from the equality case for n\geq 3 . While the cone case is analogous to the triangle, the prism case is more strict than the trapezoid case. An oblique prism has two congruent

Read the complete text