On sets of subspaces with restricted hyperplane intersection numbers

Tim Alderson, Simeon Ball · 2026

Read On sets of subspaces with restricted hyperplane intersection numbers by Tim Alderson, Simeon Ball (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-2603-27689.txt.

Introduction

Alderson and Ball bound subspace sets with capped hyperplane hits. Each hyperplane may contain at most t members of the set. They prove a sharp upper bound on the set size. Sets that hit the bound get a full structure portrait. Those maximal sets extend the idea of maximal arcs. For one parameter case, they come from ovoid reductions. Another case yields no maximal sets at all. A ternary Golay code gives rare exceptions.

Catalog record

CREATORTim Alderson, Simeon Ball
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19200
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.27689

Browse the full library · About the Tower of Babel

Full text

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

On sets of subspaces with restricted hyperplane intersection numbers Tim Alderson, Simeon Ball Abstract Let 𝒳 \mathcal{X} be a set of ( h − 1 ) (h-1) -dimensional subspaces of PG ⁡ ( k ​ h − 1 , q ) \mathrm{PG}(kh-1,q) with the property that every hyperplane contains at most t t elements of 𝒳 \mathcal{X} . We prove the upper bound | 𝒳 | ≤ ( t − k + 2 ) ​ q h + t |\mathcal{X}|\leq(t-k+2)q^{h}+t , and characterise the structure of 𝒳 \mathcal{X} in the case of equality. We call sets attaining this bound length-maximal . For k = 3 k=3 , such sets are known as maximal arcs and have been well-studied. They are known to exist for t < q h t<q^{h} if and only if q q is even and t t divides q h q^{h} . For k = 4 k=4 and q > 2 q>2 , we show that any length-maximal set must satisfy t = q h + 1 t=q^{h}+1 and that every hyperplane is either a t t -secant or a 1 1 -secant. For k ≥ 5 k\geq 5 and q > 2 q>2 , no length-maximal set exists. In the language of additive codes, these results assert that additive two-weight codes over 𝔽 q h \mathbb{F}_{q^{h}} attaining the natural Griesmer-type bound do not exist when the code dimension is 5 5 or more and q > 2 q>2 . † † Simeon Ball Departament de Matemàtiques, Universitat Politècnica de Catalunya, Barcelona, Spain. The author acknowledges the support of the Spanish Ministry of Science, Innovation and Universities grant PID2023-147202NB-I00. e-mail: [email protected] Tim Alderson Department of Mathematics and Statistics, University of New Brunswick Saint John, Saint John, NB, Canada e-mail: [email protected] 29 March 2026. † † Mathematics Subject Classification (2010): 51E20, 94B05 Keywords. maximal arc, additive code, two-weight code, projective space, subspaces, Griesmer bound 1 Introduction A maximal arc \mathcal{X} in \mathrm{PG}(2,q) is a set of points with the property that every line is incident with 0 or t points of \mathcal{X} . It is straightforward to prove that |\mathcal{X}|=(t-1)q+t and that t divides q . Denniston [ 7 ] provided examples for q even and all possible values of t . Further examples were discovered by Thas [ 10 , 12 ] , Mathon [ 9 ] , and Hamilton and Mathon [ 8 ] . When q is odd, maximal arcs were shown not to exist by Ball, Blokhuis and Mazzocca [ 2 ] ; see also [ 3 ] . The parameter set (t,q)=(3,9) was ruled out earlier by Cossu [ 5 ] and the parameter set (t,q)=(3,3^{h}) by Thas [ 11 ] . In this article we consider the natural generalisation of maximal arcs to higher dimensions. A maximal arc in \mathrm{PG}(2,q^{h}) yields, by field reduction, a set of (h-1) -dimensional subspaces in \mathrm{PG}(3h-1,q) with the property that every hyperplane contains either 0 or t subspaces. This leads us to study sets \mathcal{X} of (h-1) -dimensional subspaces in \mathrm{PG}(kh-1,q) with the property that every hyperplane contains at most t subspaces of \mathcal{X} . There is an equivalent formulation in terms of additive codes. An additive [n,k,d]_{q}^{h} code C is an \mathbb{F}_{q} -linear subspace of \mathbb{F}_{q^{h}}^{n} of size q^{kh} in which every non-zero vector has at least d non-zero coordinates. Let G be a kh\times n matrix whose row span is C . By fixing a basis \{e_{1},\ldots,e_{h}\} for \mathbb{F}_{q^{h}} over \mathbb{F}_{q} we can write G=\sum_{j=1}^{h}e_{j}G_{j}, where G_{j} is a kh\times n matrix with entries from \mathbb{F}_{q} . For each coordinate i\in\{1,\ldots,n\} we define a subspace \pi_{i} as the span of the i -th column of G_{j} , j\in\{1,\ldots,h\} . Thus \pi_{i} is a subspace of \mathrm{PG}(kh-1,q) of projective dimension at most h-1 . The property that every non-zero vector of C has at least d non-zero coordinates is equivalent to the property that every hyperplane of \mathrm{PG}(kh-1,q) contains at most t=n-d elements of the multi-set \mathcal{X}=\{\pi_{1},\ldots,\pi_{n}\} . In this setting the length bound |\mathcal{X}|\leq(t-k+2)q^{h}+t is a Griesmer-type bound for additive codes. Codes attaining this bound are additive two-weight codes whose non-zero weights are n and n-t . An additive permutation on each coordinate produces an equivalent code containing the all-one vector. Although maximal arcs do not exist in planes of odd characteristic, Mathon [ 6 ] discovered an example of 21 lines in \mathrm{PG}(5,3) (equivalently, k=3 , h=1 , q=3 , t=3 ) with the property that every hyperplane contains 0 or 3 lines. This example shows that a direct generalisation of the Ball–Blokhuis–Mazzocca nonexistence theorem to higher dimensions is not possible. To our knowledge it is the only known example in odd characteristic, and the uniqueness question remains open. The main results of this paper are as follows. Theorem 2 establishes the bound |\mathcal{X}|\leq(t-k+2)q^{h}+t and characterises equality. The divisibility condition t-k+3\mid q^{h} (Corollary 5 ) provides a further necessary condition for length-maximal sets. Lemma 7 shows that for k=4 , any length-maximal set must satisfy t=q^{h}+1 and have no zero-secant hyperplanes. Lemma 8 shows that no length-maximal set exists for k=5 and q>2 , and by induction this extends to all k\geq 5 with q>2 . The case q=2 is discussed in Section 4 . Throughout, k\geq 2 , and dimension refers to projective dimension. 2 An upper bound on the size of \mathcal{X} We first prove a divisibility condition on t in the extremal case. Lemma 1 . Let \mathcal{X} be a multi-set of (h-1) -dimensional subspaces in \mathrm{PG}(kh-1,q) of size (t-k+2)q^{h}+t with the property that every hyperplane contains either 0 or t subspaces of \mathcal{X} . Let \pi be an (s-1) -dimensional subspace skew to \mathcal{X} . Then \pi is contained in exactly \frac{q^{h}-q^{kh-s-h}}{q-1}+\frac{(k-2)(q^{kh-s}-q^{h})}{t(q-1)} zero-secant hyperplanes. In particular, if there exists a point skew to \mathcal{X} , then t divides (k-2)q^{h} . Proof. Let a_{0} and a_{t} denote the number of zero-secant and t -secant hyperplanes containing \pi , respectively. Counting hyperplanes through \pi , a_{0}+a_{t}=\frac{q^{kh-s}-1}{q-1}, and counting pairs (x,H) with x\in\mathcal{X} and H a hyperplane through both \pi and x , ta_{t}=\bigl((t-k+2)q^{h}+t\bigr)\frac{q^{kh-s-h}-1}{q-1}. Solving for a_{0} gives the stated formula. Applying this for s=0 and s=1 yields t\mid(k-2)q^{h}(1+q+\cdots+q^{kh-h-1})\quad\text{and}\quad t\mid(k-2)q^{h}(1+q+\cdots+q^{kh-h-2}), from which t divides (k-2)q^{h} . ∎ Theorem 2 . Let \mathcal{X} be a multi-set of (h-1) -dimensional subspaces in \mathrm{PG}(kh-1,q) with the property that every hyperplane contains at most t subspaces of \mathcal{X} . Then |\mathcal{X}|\leq(t-k+2)q^{h}+t, (1) with equality if and only if 1. every set of (k-1) elements of \mathcal{X} spans a ((k-1)h-1) -dimensional subspace, and 2. every hyperplane contains either 0,1,\ldots,k-3 or t elements of \mathcal{X} . Proof. Let x_{1},\ldots,x_{k-2} be elements of \mathcal{X} and let \pi=\langle x_{1},\ldots,x_{k-2}\rangle . The dimension of \pi is (k-2)h-1-e_{1} for some e_{1}\geq 0 . Let (k-1)h-1-e_{1}-e_{2} denote the maximum dimension of \langle\pi,x\rangle over x\in\mathcal{X}\setminus\{x_{1},\ldots,x_{k-2}\} . We count pairs (x,H) where x\in\mathcal{X}\setminus\{x_{1},\ldots,x_{k-2}\} and H is a hyperplane containing both x and \pi . By hypothesis each such hyperplane contains at most t-k+2 elements of \mathcal{X}\setminus\{x_{1},\ldots,x_{k-2}\} , so this count is at most \frac{q^{2h+e_{1}}-1}{q-1}(t-k+2). Since (k-1)h-1-e_{1}-e_{2} is the maximum dimension of \langle\pi,x\rangle , each element of \mathcal{X}\setminus\{x_{1},\ldots,x_{k-2}\} lies in at least \frac{q^{h+e_{1}+e_{2}}-1}{q-1} hyperplanes through \pi , so the count is at least \frac{q^{h+e_{1}+e_{2}}-1}{q-1}(n-k+2). Combining these

Read the complete text