A subspace code of size 333 in the setting of a binary \(q\)-analog of the Fano plane (Q2311610)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    A subspace code of size 333 in the setting of a binary \(q\)-analog of the Fano plane
    scientific article

      Statements

      A subspace code of size 333 in the setting of a binary \(q\)-analog of the Fano plane (English)
      0 references
      0 references
      0 references
      0 references
      0 references
      4 July 2019
      0 references
      The subspace distance between two subspaces \(U,W\) of \(\mathbb{F}_q^n\) is defined by \[ d_S(U,W) = \operatorname{dim}(U+W) - \operatorname{dim}(U\cap W).\] Let \(A_q(n,d;k)\) to be the maximum cardinality of a constant dimension subspace code in \(\mathbb{F}_q^n\) with subspaces of dimension \(k\) and minimum subspace distance \(d\). The smallest case for binary constant dimension codes in which this value is unknown is \(A_2(7,4;3)\). Prior to this work, it was known that \(329 \leq A_2(7,4;3) \leq 381\). Every previously known code of size \(329\) has trivial automorphism group. The authors of this paper find a code of size \(329\) with automorphism group of size \(16\) and are able to modify it to find a code of size \(333\). The main idea is to consider possible subgroups of the automorphism group of the code, building on an approach described by \textit{A. Kohnert} and \textit{S. Kurz} [Lect. Notes Comput. Sci. 5393, 31--42 (2008; Zbl 1178.94239)]. Define \(A_q(n,d;k;U)\) as we defined \(A_q(n,d;k)\), except with the additional requirement that \(U\) is a subgroup of the automorphism group of the code. The authors describe how to give upper bounds for \(A_q(n,d;k;U)\) by solving an integer linear program. In order to understand the subgroups of \(\operatorname{GL}(\mathbb{F}_2^7)\), the authors use the Small Groups library contained in Magma. While the paper focuses on a specific set of parameters for subspace codes, the methods developed should be useful in a broad range of combinatorial problems.
      0 references
      0 references
      finite groups
      0 references
      finite projective spaces
      0 references
      constant dimension codes
      0 references
      subspace codes
      0 references
      subspace distance
      0 references
      combinatorics
      0 references
      computer search
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references