Source author record

Kyle Pacheco

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

Catalog footprint

Research overview

1works
6topics
3close collaborators

Published work

1 publications

preprint2025arXiv

Almost perfect nonlinear power functions with exponents expressed as fractions

Let FF be a finite field, let ff be a function from FF to FF, and let aa be a nonzero element of FF. The discrete derivative of ff in direction aa is Δaf ⁣:F→FΔ_a f \colon F \to F with (Δaf)(x)=f(x+a)−f(x)(Δ_a f)(x)=f(x+a)-f(x). The differential spectrum of ff is the multiset of cardinalities of all the fibers of all the derivatives ΔafΔ_a f as aa runs through F∗F^*. An almost perfect nonlinear (APN) function is one for which the largest cardinality in its differential spectrum is 22. Almost perfect nonlinear functions are of interest as cryptographic primitives. If dd is a positive integer, then the power function over FF with exponent dd is the function f ⁣:F→Ff \colon F \to F with f(x)=xdf(x)=x^d for every x∈Fx \in F. There is a small number of known infinite families of APN power functions. In this paper, we re-express the exponents for one such family in a more convenient form. This enables us not only to obtain the differential spectrum of each power function ff with an exponent in our family, but also to determine the elements that lie in an arbitrary fiber of the discrete derivative of ff. This differential analysis, which is far more detailed than previous results, is achieved by composing the discrete derivative of ff with some permutations and a double covering of its domain to obtain a function whose fibers can more readily be analyzed.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.