Searcharxiv⌕ Search

arXiv · 2610.24408

A Recursive Construction of Perfect $K^n\times K^n$ Arrays over $K$ Roots of Unity

Abstract

We give a uniform, explicit and elementary recursive construction of two-dimensional arrays with perfect (ideal) periodic autocorrelation over a \emph{fixed} alphabet of $K$ roots of unity and \emph{unbounded} size: for every integer $K\ge2$ and every $n\ge1$ a perfect $K^n\times K^n$ array whose entries are $K$-th roots of unity, given by a nested sum of floor terms. The construction works with \emph{twisted} (quasi-perfect) arrays indexed by a type $v\in\Z_K^2$, of which ordinary perfect arrays are the type $v=0$ members. For $K=2$ the types are Jedwab's generalised perfect binary arrays and the recursion specialises to the doubling constructions of Jedwab--Mitchell and Wild; for prime $K$ the arrays are the relative difference sets whose existence in abelian $p$-groups was settled by Ma and Schmidt; and the bare existence of perfect $K^n\times K^n$ arrays over $K$ roots of unity follows, for every $K$, from recent constructions of group-invariant Butson Hadamard matrices. The contribution of this note is therefore not existence but the calculus: types with arbitrary entries in $\Z_K^2$; three spectral lemmas (a twisted periodic extension of a type-$v$ array has a flat spectrum supported on one coset of $K\Z^2$; interleaving arrays whose spectra are flat on disjoint cosets gives a perfect array; and a shear of the index plane with a floor-function carry permutes types through the action of $SL_2(\Z_K)$ on $\Z_K^2$, whose orbits are the classes $\gcd(v_1,v_2,K)=g$); a one-directional lift that produces every type at side $KM$ from every type at side $M$ for arbitrary $K$; and the resulting closed index function, which is the recursive form of the floor-function arrays of \cite{Blake2017}.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Sam Blake. 2026-10-17. A Recursive Construction of Perfect $K^n\times K^n$ Arrays over $K$ Roots of Unity. https://arxiv.org/abs/2610.24408

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

KEEP EXPLORING

Related papers

Two Erdos-Hajnal-type theorems for forbidden order-size pairs

The celebrated Erdős-Hajnal conjecture says that any graph without a fixed induced subgraph $H$ contains a very large homogeneous set. A direct analog of this conjecture is not true for hypergraphs. In this paper we present two natural variants of this problem which do hold for hypergraphs. We show that for every $r \geq 3$, $m \geq m_0(r)$ and $0 \leq f \leq \binom{m}{r}$, if an $r$-graph $G$ does not contain $m$ vertices spanning exactly $f$ edges, then $G$ contains much larger homogeneous sets than what is guaranteed to exist in general $r$-graphs. We also prove that if a $3$-graph $G$ does not contain homogeneous sets of polynomial size, then for every $m \geq 3$ there are $Ω(m^3)$ values of $f$ such that $G$ contains $m$ vertices spanning exactly $f$ edges. This makes progress on a problem of Axenovich, Bradač, Gishboliner, Mubayi and Weber.

math.CO↗

Unexpectedly, a symmetry on unlabeled graphs

We exhibit the joint symmetric distribution of the following two parameters on the set of unlabeled, simple, connected graphs with $n$ vertices. The first parameter is the maximal number of leaves attached to a vertex. The second parameter is the size of the largest set of vertices sharing the same closed neighborhood minus $1$. Apparently, this is the first example of a natural, non-trivial equidistribution of graph parameters on unlabeled connected graphs on a fixed set of vertices. Our proof is enumerative, using the theory of species. Exhibiting an explicit bijection interchanging the two parameters remains an open problem.

math.CO↗

Cycle lengths in the percolated hypercube

Let $Q^d_p$ be the random subgraph of the $d$-dimensional binary hypercube obtained after edge-percolation with probability $p$. It was shown recently by the authors that, for every $\varepsilon > 0$, there is some $c = c(\varepsilon)>0$ such that, if $pd\ge c$, then typically $Q^d_p$ contains a cycle of length at least $(1-\varepsilon)2^d$. We strengthen this result to show that, under the same assumptions, typically $Q^d_p$ contains cycles of all even lengths between $4$ and $(1-\varepsilon)2^d$.

math.CO↗