Searcharxiv⌕ Search

arXiv · 2609.35689

Linearity bounds for APN functions

Abstract

For $n\ge5$, let $F\colon\mathbb{F}_2^n\to \mathbb{F}_2^n$ be almost perfect nonlinear and write $N=2^n$. It is proven that the linearity $\mathcal{L}(F)$ of $F$, i.e., the largest absolute Walsh coefficient of a nonzero component, is at most $N-10$ in even dimension and at most $N-6$ in odd dimension. This improves the general upper bound of $N-6$ in even dimension and $N-4$ in odd dimension. It is further proven that, for each fixed $k$, the $k$-th largest absolute Walsh coefficient among nonzero components, counted with multiplicity, is at most $(1+O_k(2^{-n}))N/\sqrt{k}$ as $n\to\infty$. The second and fourth largest coefficients are at most $2\lfloor N/3\rfloor$ and $N/2$, respectively. Finally, a bound on the linearity in terms of the number $q$ of nonplateaued nonzero components is derived. In odd dimension, for $q>0$, we have $\mathcal{L}(F)^2\le N(1+\sqrt{q(N-1)})$, so that $\mathcal{L}(F)/N\to1$ implies $q/N\to1$. In even dimension, for every $1/2\le C<1$, the condition $q\le(4C-C^2-1)N/4+1$ implies $\mathcal{L}(F)\le CN$. In particular, $q\le3N/16+2$ implies $\mathcal{L}(F)\le N/2$.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Christof Beierle. 2026-09-28. Linearity bounds for APN functions. https://arxiv.org/abs/2609.35689

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

The Rainbow Saturation Number of Cycles

An edge-colored graph $(G,\mathcal C)$ is $F$-rainbow saturated if it contains no rainbow copy of $F$, but adding any nonedge in any prescribed color creates one. The rainbow saturation number $\operatorname{rsat}(n,F)$ is the minimum number of edges in such a graph of order $n$. In this paper we study $\operatorname{rsat}(n,C_r)$. We determine the value for $C_4$ exactly, and establish bounds for every fixed $C_r$ with $r\geq5$.

math.CO↗

Moreira's Theorem for Image Partition Regular Matrices

The famous Hindman conjecture says that for any finite coloring of natural numbers, there exists a monochromatic copy of the form $\{x,x+y,xy\}.$ In a celebrated article, Moreira gave a partial answer to this conjecture by showing that every finite coloring of the natural numbers contains a monochromatic configuration of the form $\{x, x+y, xy\}$. In this article we prove matrix versions (both finite and infinite) of Moreira's theorem. A matrix $A$ is said to be an image partition regular matrix if for any finite coloring of naturals, there exists a monochromatic image of $A,$ i.e. there exists a vector $\vec X$ such that all the entries of $A\vec X$ are monochromatic. From a recent paper of Bowen, one can derive the finite matrix version of the Moreira theorem: if $A$ and $B$ are two finite image partition regular matrices of the same order, then under any finite coloring of $\mathbb{N}$, there exist vectors $\vec{X}$ and $\vec{Y}$ such that all entries in the union of $A\vec{X}, A\vec{X} + B\vec{Y}, A\vec{X} \cdot B\vec{Y}$ are monochromatic, where $A\vec{X} \cdot B\vec{Y}$ denote the vector each of its entries are pointwise multiplication of the coordinates of $A\vec{X} \text{ and } B\vec{Y}$. In this article, we give a short combinatorial proof of this result, and then we extend it to infinite image partition regular matrices.

math.CO↗

Explicit enumeration and large-valence asymptotics of even-valent maps

Let $\mathscr{N}_g(2ν,j)$ denote the number of connected labelled $2ν$-valent maps of genus $g$ with $j$ vertices. Explicit bivariate formulae for $\mathscr{N}_g(2ν,j)$ have been available only in the planar and toroidal cases. Using a structural formula of Ercolani et al (2023), we translate the problem of determining an explicit bivariate formula for $\mathscr{N}_g(2ν,j)$, $g \geq 2$, to finding finitely many counts with a fixed number of vertices. For $g=2,3$ and $4$ we determine these counts using the associated orthogonal polynomials, yielding explicit bivariate formulae for $\mathscr{N}_g(2ν,j)$ in these genera. Furthermore, the same method applies for every $g\ge5$ at the cost of additional computation. From these formulae we obtain the leading-order asymptotics of $\mathscr{N}_g(2ν,j)$ as $ν\to\infty$ for $g=2,3,4$, and we conjecture the structure of these formulae in general genus. In addition, we establish an analogous reduction to finitely many counts, derive explicit formulae and large-valence asymptotics, and formulate corresponding conjectures for two-legged even-valent maps.

math.CO↗