Searcharxiv⌕ Search

arXiv subjects

Konrad J. Swanepoel

Publications and source records attributed to Konrad J. Swanepoel.

34 records · Page 2Linked to original sources
arXiv2011

Absorbing angles, Steiner minimal trees, and antipodality

We give a new proof that a star $\{op_i:i=1,...,k\}$ in a normed plane is a Steiner minimal tree of its vertices $\{o,p_1,...,p_k\}$ if and only if all angles formed by the edges at o are absorbing [Swanepoel, Networks \textbf{36} (2000), 104--113]. The proof is more conceptual and simpler than the original one. We also find a new sufficient condition for higher-dimensional normed spaces to share this characterization. In particular, a star $\{op_i: i=1,...,k\}$ in any CL-space is a Steiner minimal tree of its vertices $\{o,p_1,...,p_k\}$ if and only if all angles are absorbing, which in turn holds if and only if all distances between the normalizations $\frac{1}{\|p_i\|}p_i$ equal 2. CL-spaces include the mixed $\ell_1$ and $\ell_\infty$ sum of finitely many copies of $R^1$.

math.MG↗
arXiv2011

Embedding a Latin square with transversal into a projective space

A Latin square of side n defines in a natural way a finite geometry on 3n points, with three lines of size n and n^2 lines of size 3. A Latin square of side n with a transversal similarly defines a finite geometry on 3n+1 points, with three lines of size n, n^2-n lines of size 3, and n concurrent lines of size 4. A collection of k mutually orthogonal Latin squares defines a geometry on kn points, with k lines of size n and n^2 lines of size k. Extending work of Bruen and Colbourn (J. Combin. Th. Ser. A 92 (2000), 88-94), we characterise embeddings of these finite geometries into projective spaces over skew fields.

math.CO↗
arXiv2011

Favourite distances in high dimensions

Let $S$ be a set of $n$ points in $d$-dimensional Euclidean space. Assign to each $x\in S$ an arbitrary distance $r(x)>0$. Let $e_r(x,S)$ denote the number of points in $S$ at distance $r(x)$ from $x$. Avis, Erdös and Pach (1988) introduced the extremal quantity $f_d(n)=\max\sum_{x\in S}e_r(x,S)$, where the maximum is taken over all $n$-point sets $S$ in $d$-dimensional space and all assignments $r\colon S\to(0,\infty)$ of distances. We give a quick derivation of the asymptotics of the error term of $f_d(n)$ using only the analogous asymptotics of the maximum number of unit distance pairs in a set of $n$ points, which improves on previous results of Avis, Erdös and Pach (1988) and Erdös and Pach (1990). Then we prove a stability result for $d\geq 4$, asserting that if $(S,r)$ with $|S|=n$ satisfies $e_r(S)=f_d(n)-o(n^2)$, then, up to $o(n)$ points, $S$ is a Lenz construction with $r$ constant. Finally we use stability to show that for $n$ sufficiently large (depending on $d$) the pairs $(S,r)$ that attain $f_d(n)$ are up to scaling exactly the Lenz constructions that maximise the number of unit distance pairs with $r\equiv 1$, with some exceptions in dimension 4. Analogous results hold for the furthest neighbour digraph, where $r$ is fixed to be $r(x)=\max_{y\in S} |xy|$ for $x\in S$.

math.CO↗
arXiv2010

Large convexly independent subsets of Minkowski sums

Let $E_d(n)$ be the maximum number of pairs that can be selected from a set of $n$ points in $R^d$ such that the midpoints of these pairs are convexly independent. We show that $E_2(n)\geq Ω(n\sqrt{\log n})$, which answers a question of Eisenbrand, Pach, Rothvoß, and Sopher (2008) on large convexly independent subsets in Minkowski sums of finite planar sets, as well as a question of Halman, Onn, and Rothblum (2007). We also show that $\lfloor\frac{1}{3}n^2\rfloor\leq E_3(n)\leq \frac{3}{8}n^2+O(n^{3/2})$. Let $W_d(n)$ be the maximum number of pairwise nonparallel unit distance pairs in a set of $n$ points in some $d$-dimensional strictly convex normed space. We show that $W_2(n)=Θ(E_2(n))$ and for $d\geq 3$ that $W_d(n)\sim\frac12\left(1-\frac{1}{a(d)}\right)n^2$, where $a(d)\in N$ is related to strictly antipodal families. In fact we show that the same asymptotics hold without the requirement that the unit distance pairs form pairwise nonparallel segments, and also if diameter pairs are considered instead of unit distance pairs.

math.CO↗
arXiv2010

Midpoint sets contained in the unit sphere of a normed space

The midpoint set M(S) of a set S of points is the set of all midpoints of pairs of points in S. We study the largest cardinality of a midpoint set M(S) in a finite-dimensional normed space, such that M(S) is contained in the unit sphere, and S is outside the closed unit ball. We show in three dimensions that this maximum (if it exists) is determined by the facial structure of the unit ball. In higher dimensions no such relationship exists. We also determine the maximum for euclidean and sup norm spaces.

math.MG↗
arXiv2010

Sets of unit vectors with small pairwise sums

We study the sizes of delta-additive sets of unit vectors in a d-dimensional normed space: the sum of any two vectors has norm at most delta. One-additive sets originate in finding upper bounds of vertex degrees of Steiner Minimum Trees in finite dimensional smooth normed spaces (Z. Füredi, J. C. Lagarias, F. Morgan, 1991). We show that the maximum size of a delta-additive set over all normed spaces of dimension d grows exponentially in d for fixed delta>2/3, stays bounded for delta<2/3, and grows linearly at the threshold delta=2/3. Furthermore, the maximum size of a 2/3-additive set in d-dimensional normed space has the sharp upper bound of d, with the single exception of spaces isometric to three-dimensional l^1 space, where there exists a 2/3-additive set of four unit vectors.

math.MG↗
arXiv2009

Helly-type Theorems for Hollow Axis-aligned Boxes

A hollow axis-aligned box is the boundary of the cartesian product of $d$ compact intervals in R^d. We show that for d\geq 3, if any 2^d of a collection of hollow axis-aligned boxes have non-empty intersection, then the whole collection has non-empty intersection; and if any 5 of a collection of hollow axis-aligned rectangles in R^2 have non-empty intersection, then the whole collection has non-empty intersection. The values 2^d for d\geq 3 and 5 for d=2 are the best possible in general. We also characterize the collections of hollow boxes which would be counterexamples if 2^d were lowered to 2^d-1, and 5 to 4, respectively.

math.CO↗
arXiv2008

Simultaneous packing and covering in sequence spaces

We adapt a construction of Klee (1981) to find a packing of unit balls in $\ell_p$ ($1\leq p<\infty$) which is efficient in the sense that enlarging the radius of each ball to any $R>2^{1-1/p}$ covers the whole space. We show that the value $2^{1-1/p}$ is optimal.

math.MG↗
arXiv2008

Balancing unit vectors

Theorem A. Let $x_1,...,x_{2k+1}$ be unit vectors in a normed plane. Then there exist signs $\epsi_1,...,\epsi_{2k+1}\in\{\pm 1\}$ such that $\norm{\sum_{i=1}^{2k+1}\epsi_i x_i}\leq 1$. We use the method of proof of the above theorem to show the following point facility location result, generalizing Proposition 6.4 of Y. S. Kupitz and H. Martini (1997). Theorem B. Let $p_0,p_1,...,p_n$ be distinct points in a normed plane such that for any $1\leq i<j\leq n$ the closed angle $\angle p_ip_0p_j$ contains a ray opposite some $\overrightarrow{p_0p_k}, 1\leq k\leq n$. Then $p_0$ is a Fermat-Toricelli point of $\{p_0,p_1,...,p_n\}$, i.e. $x=p_0$ minimizes $\sum_{i=0}^n\norm{x-p_i}$. We also prove the following dynamic version of Theorem A. Theorem C. Let $x_1,x_2,...$ be a sequence of unit vectors in a normed plane. Then there exist signs $\epsi_1,\epsi_2,...\in\{\pm 1\}$ such that $\norm{\sum_{i=1}^{2k}\epsi_i x_i}\leq 2$ for all $k\in\N$. Finally we discuss a variation of a two-player balancing game of J. Spencer (1977) related to Theorem C.

math.MG↗
arXiv2007

Extremal Problems in Minkowski Space related to Minimal Networks

We solve the following problem of Z. Füredi, J. C. Lagarias and F. Morgan [FLM]: Is there an upper bound polynomial in $n$ for the largest cardinality of a set S of unit vectors in an n-dimensional Minkowski space (or Banach space) such that the sum of any subset has norm less than 1? We prove that |S|\leq 2n and that equality holds iff the space is linearly isometric to \ell^n_\infty, the space with an n-cube as unit ball. We also remark on similar questions raised in [FLM] that arose out of the study of singularities in length-minimizing networks in Minkowski spaces.

math.MG↗
arXiv2007

Elementary incidence theorems for complex numbers and quaternions

We present some elementary ideas to prove the following Sylvester-Gallai type theorems involving incidences between points and lines in the planes over the complex numbers and quaternions. (1) Let A and B be finite sets of at least two complex numbers each. Then there exists a line l in the complex affine plane such that l intersects AxB in exactly two points. (2) Let S be a finite noncollinear set of points in the complex affine plane. Then there exists a line l intersecting S in 2, 3, 4 or 5 points. (3) Let A and B be finite sets of at least two quaternions each. Then there exists a line l in the quaternionic affine plane such that l intersects AxB in 2, 3, 4 or 5 points. (4) Let S be a finite noncollinear set of points in the quaternionic affine plane. Then there exists a line l intersecting S in at least 2 and at most 24 points.

math.CO↗
arXiv2006

A problem of Kusner on equilateral sets

R. B. Kusner [R. Guy, Amer. Math. Monthly 90 (1983), 196--199] asked whether a set of vectors in a d-dimensional real vector space such that the l-p distance between any pair is 1, has cardinality at most d+1. We show that this is true for p=4 and any d >= 1, and false for all 1 0$ depends on p.

math.MG↗
arXiv2006

Sylvester-Gallai Theorems for Complex Numbers and Quaternions

A Sylvester-Gallai (SG) configuration is a finite set S of points such that the line through any two points in S contains a third point of S. According to the Sylvester-Gallai Theorem, an SG configuration in real projective space must be collinear. A problem of Serre (1966) asks whether an SG configuration in a complex projective space must be coplanar. This was proved by Kelly (1986) using a deep inequality of Hirzebruch. We give an elementary proof of this result, and then extend it to show that an SG configuration in projective space over the quaternions must be contained in a three-dimensional flat.

math.MG↗
arXiv2004

Antinorms and Radon curves

In this paper we consider two notions that have been discovered and rediscovered by geometers and analysts since 1917 up to the present day: Radon curves and antinorms. A Radon curve is a special kind of centrally symmetric closed convex curve in the plane. A Radon plane is a normed plane obtained by using a Radon curve as (the boundary of) the unit ball. Many known results in Euclidean geometry also hold for Radon planes, for example the triangle and parallelogram area formulas, certain theorems on angular bisectors, the area formula of a polygon circumscribed about a circle, certain isoperimetric inequalities, and the non-expansive property of certain non-linear projections. These results may be further generalized to arbitrary normed planes if we formally change the statement of the result by referring in some places to the antinorm instead of the norm. The antinorm is a norm dual to the norm of an arbitrarily given normed plane, although it lives in the same plane as the original norm. It is the purpose of this mainly expository paper to give a list of results on antinorms that generalize results true for Radon norms, and in many cases characterize Radon norms among all norms in the plane. Many of the results are old, well-known, and have often been rediscovered. However, for most of the results we give streamlined proofs. Also, some of the characterizations of Radon curves given here seem not to have appeared previously in print.

math.FA↗
arXiv2004

Largest family without A union B contained in C intersect D

Let F be a family of subsets of an n-element set not containing four distinct members such that A union B is contained in C intersect D. It is proved that the maximum size of F under this condition is equal to the sum of the two largest binomial coefficients of order n. The maximum families are also characterized. A LYM-type inequality for such families is given, too.

math.CO↗