Proofs of two conjectures on the dimensions of binary codes

DOI10.1007/S10623-012-9682-6MaRDI QIDQ2439731zbMATH OpenOpenAlexWikidataFDO

Authors Junhua Wu

Publication date 17 March 2014

Published in Designs, Codes and Cryptography (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1001.5077

zbMATH Keywords

2-rankbinary rankblock idempotentBrauer's theorycharacterconicconic in \(\mathrm{PG}(2, q)\)general linear groupincidence matrixlow-density parity-checkcodemodule

Mathematics Subject Classification ID

Combinatorial aspects of finite geometries (05B25) Modular representations and characters (20C20) Combinatorial structures in finite projective spaces (51E20)

Abstract: Let $mathcalL$ and $mathcal{L}_0$ be the binary codes generated by the column $mathbb{F}_2$-null space of the incidence matrix of external points versus passant lines and internal points versus secant lines with respect to a conic in $PG(2,q)$, respectively. We confirm the conjectures on the dimensions of $mathcalL$ and $mathcal{L}_0$ using methods from both finite geometry and modular representation theory.

Review text

A non-degenerate conic \(\mathbf O\) in \(\mathrm{PG}(2, q)\) is a set of \(q + 1\) points that satisfy a non-degenerate quadratic form. No three points of \(\mathbf O\) are collinear thus making \(\mathbf O\) an oval of \(\mathrm{PG}(2, q)\), and a well-known result of Segre asserts that ovals in \(\mathrm{PG}(2, q)\) with \(q\) odd arise precisely in this manner. Let \(B\) denote the binary matrix corresponding to the incidence between external points (those that lie on two tangent lines) versus passant lines of \(\mathbf O\) and \(B_0\) denote the binary matrix corresponding to the incidence between internal points (those that lie on no tangent line) versus the secant lines. Then \(B\) is a \(\frac{q(q+1)}{2} \times \frac{q(q-1)}{2}\) binary matrix while \(B_0\) is a \(\frac{q(q-1)}{2} \times \frac{q(q+1)}{2}\) binary matrix. Let \(\mathbf L\) and \(\mathbf L_0\) denote the binary linear codes generated by the column null space of \(B\) and \(B_0\) respectively. Confirming a conjecture of Droms et. al. the author proves that the rank of \(\mathbf L\) is \(\frac{q^2-1}{4} - q\) if \(q \equiv 1 \pmod4\) and is \(\frac{q^2-1}{4} - q + 1\) if \(q \equiv 3 \pmod4\) while the rank of \(\mathbf L_0\) is \(\frac{q^2-1}{4}\) if \(q \equiv 1 \pmod4\) and is \(\frac{q^2-1}{4} + 1\) if \(q \equiv 3 \pmod4\). The proof technique is both combinatorial and uses modular forms via Brauer's theory of characters.

Recommendations

• On binary codes from conics in \(\mathrm{PG}(2,q)\)
• Dimensions of some binary codes arising from a conic in \(\mathrm{PG}(2,q)\)
• Conics arising from external points and their binary codes
• Conics arising from internal points and their binary codes
• The dimension of projective geometry codes

Cites work

• Die Zerlegungsmatrizen der Gruppen \(PSL(2,p^f)\)
• Dimensions of some binary codes arising from a conic in \(\mathrm{PG}(2,q)\)
• Fundamentals of Error-Correcting Codes
• Hexagons, Conics, A₅ and PSL₂ (K )
• scientific article; zbMATH DE number 53917 (Why is no real title available?)
• scientific article; zbMATH DE number 1101474 (Why is no real title available?)
• scientific article; zbMATH DE number 1152903 (Why is no real title available?)
• scientific article; zbMATH DE number 205981 (Why is no real title available?)
• scientific article; zbMATH DE number 3252237 (Why is no real title available?)
• scientific article; zbMATH DE number 3421490 (Why is no real title available?)
• LDPC codes generated by conics in the classical projective plane
• On binary codes from conics in \(\mathrm{PG}(2,q)\)
• Ovals In a Finite Projective Plane
• Some applications of the theory of blocks of characters of finite groups. IV
• The principal block of finite groups with dihedral Sylow 2-subgroups

Cited in

(6)

• The dimensions of \(LU(3,q)\) codes
• 2-ranks of incidence matrices associated with conics in finite projective planes
• Dimensions of some binary codes arising from a conic in \(\mathrm{PG}(2,q)\)
• One more proof of the first linear programming bound for binary codes and two conjectures
• On binary codes from conics in \(\mathrm{PG}(2,q)\)
• Incidence matrices of finite quadratic spaces

This page was built for publication: Proofs of two conjectures on the dimensions of binary codes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2439731)