2008•Problems of Information TransmissionRequires access

Bent and hyper-bent functions over a field of 2ℓ elements

Alexander S. Kuzmin, Viktor Timofeevich Markov, Alexander A. Nechaev, В. А. Шишкин, Алексей Борисович Шишков

Open publisher page 8 citations

Abstract

We study the parameters of bent and hyper-bent (HB) functions in n variables over a field $$ P = \mathbb{F}_q $$ with q = 2ℓ elements, ℓ > 1. Any such function is identified with a function F: Q → P, where $$ P < Q = \mathbb{F}_{qn} $$ . The latter has a reduced trace representation F = tr (Φ), where Φ(x) is a uniquely defined polynomial of a special type. It is shown that the most accurate generalization of results on parameters of bent functions from the case ℓ = 1 to the case ℓ > 1 is obtained if instead of the nonlinearity degree of a function one considers its binary nonlinearity index (in the case ℓ = 1 these parameters coincide). We construct a class of HB functions that generalize binary HB functions found in [1]; we indicate a set of parameters q and n for which there are no other HB functions. We introduce the notion of the period of a function and establish a relation between periods of (hyper-)bent functions and their frequency characteristics.

About this research paper

What this paper is about

We study the parameters of bent and hyper-bent (HB) functions in n variables over a field $$ P = \mathbb{F}_q $$ with q = 2ℓ elements, ℓ > 1. Any such function is identified with a function F: Q → P, where $$ P < Q = \mathbb{F}_{qn} $$ . The latter has a reduced trace representation F = tr (Φ), where Φ(x) is a uniquely defined polynomial of a special type. It is shown that the most accurate generalization of results on parameters of bent functions from the case ℓ = 1 to the case ℓ > 1 is obtained if instead of the nonlinearity degree of a function one considers its binary nonlinearity index (in the case ℓ = 1 these parameters coincide). We construct a class of HB functions that generalize binary HB functions found in [1]; we indicate a set of parameters q and n for which there are no other HB functions. We introduce the notion of the period of a function and establish a relation between periods of (hyper-)bent functions and their frequency characteristics.

Why it matters

OpenAlex reports 8 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We study the parameters of bent and hyper-bent (HB) functions in n variables over a field $$ P = \mathbb{F}_q $$ with q = 2ℓ elements, ℓ > 1. Any such function is identified with a function F: Q → P, where $$ P < Q = \mathbb{F}_{qn} $$ . The latter has a reduced trace representation F = tr (Φ), where Φ(x) is a uniquely defined polynomial of a special type. It is shown that the most accurate generalization of results on parameters of bent functions from the case ℓ = 1 to the case ℓ > 1 is obtained if instead of the nonlinearity degree of a function one considers its binary nonlinearity index (in the case ℓ = 1 these parameters coincide). We construct a class of HB functions that generalize binary HB functions found in [1]; we indicate a set of parameters q and n for which there are no other HB functions. We introduce the notion of the period of a function and establish a relation between periods of (hyper-)bent functions and their frequency characteristics.

Key concepts: Bent molecular geometry, Mathematics, Generalization, Bent function, Binary number, Function (biology), Polynomial, Combinatorics

Related papers

Back to paper searchBrowse research topicsOriginal source
Bent and hyper-bent functions over a field of 2ℓ elements — Research Paper | ScholarLens