Searcharxiv⌕ Search

arXiv subjects

Alexander Pott

Publications and source records attributed to Alexander Pott.

27 records · Page 2Linked to original sources

Skew Hadamard Difference Sets from Dickson Polynomials of Order 7

Skew Hadamard difference sets are an interesting topic of study for over seventy years. For a long time, it had been conjectured the classical Paley difference sets (the set of nonzero quadratic residues in $\mathbb{F}_q$ where $q \equiv 3 \bmod{4}$) were the only example in abelian groups. In 2006, the first author and Yuan disproved this conjecture by showing that the image set of $\mathcal{D}_5(x^2,u)$ is a new skew Hadamard difference set in $(\mathbb{F}_{3^m},+)$ with $m$ odd, where $\mathcal{D}_n(x,u)$ denotes the first kind of Dickson polynomials of order $n$ and $u \in \mathbb{F}_q^*$. The key observation in the proof is that $\mathcal{D}_5(x^2,u)$ is a planar function from $\mathbb{F}_{3^m}$ to $\mathbb{F}_{3^m}$ for $m$ odd. Since then a few families of new skew Hadamard difference sets have been discovered. In this paper, we prove that for all $u \in \mathbb{F}_{3^m}^*$, the set $D_u := \{\mathcal{D}_7(x^2,u) : x \in \mathbb{F}_{3^m}^* \}$ is a skew Hadamard difference set in $(\mathbb{F}_{3^m}, +)$, where $m$ is odd and $m \not \equiv 0 \pmod{3}$. The proof is more complicated and different from that of Ding-Yuan skew Hadamard difference sets since $\mathcal{D}_7(x^2,u)$ is not planar in $\mathbb{F}_{3^m}$. Furthermore, we show that such skew Hadamard difference sets are inequivalent to all existing ones for $m = 5, 7$ by comparing the triple intersection numbers.

math.CO↗

Characterization of Negabent Functions and Construction of Bent-Negabent Functions with Maximum Algebraic Degree

We present necessary and sufficient conditions for a Boolean function to be a negabent function for both even and odd number of variables, which demonstrate the relationship between negabent functions and bent functions. By using these necessary and sufficient conditions for Boolean functions to be negabent, we obtain that the nega spectrum of a negabent function has at most 4 values. We determine the nega spectrum distribution of negabent functions. Further, we provide a method to construct bent-negabent functions in $n$ variables ($n$ even) of algebraic degree ranging from 2 to $\frac{n}{2}$, which implies that the maximum algebraic degree of an $n$-variable bent-negabent function is equal to $\frac{n}{2}$. Thus, we answer two open problems proposed by Parker and Pott and by Stǎnicǎ \textit{et al.} respectively.

cs.IT↗

A Note on a Conjecture for Balanced Elementary Symmetric Boolean Functions

In 2008, Cusick {\it et al.} conjectured that certain elementary symmetric Boolean functions of the form $σ_{2^{t+1}l-1, 2^t}$ are the only nonlinear balanced ones, where $t$, $l$ are any positive integers, and $σ_{n,d}=\bigoplus_{1\le i_1<...<i_d\le n}x_{i_1}x_{i_2}...x_{i_d}$ for positive integers $n$, $1\le d\le n$. In this note, by analyzing the weight of $σ_{n, 2^t}$ and $σ_{n, d}$, we prove that ${\rm wt}(σ_{n, d})<2^{n-1}$ holds in most cases, and so does the conjecture. According to the remainder of modulo 4, we also consider the weight of $σ_{n, d}$ from two aspects: $n\equiv 3({\rm mod\}4)$ and $n\not\equiv 3({\rm mod\}4)$. Thus, we can simplify the conjecture. In particular, our results cover the most known results. In order to fully solve the conjecture, we also consider the weight of $σ_{n, 2^t+2^s}$ and give some experiment results on it.

cs.IT↗

A new family of semifields with 2 parameters

A new family of commutative semifields with two parameters is presented. Its left and middle nucleus are both determined. Furthermore, we prove that for any different pairs of parameters, these semifields are not isotopic. It is also shown that, for some special parameters, one semifield in this family can lead to two inequivalent planar functions. Finally, using similar construction, new APN functions are given.

math.CO↗

Non-Boolean almost perfect nonlinear functions on non-Abelian groups

The purpose of this paper is to present the extended definitions and characterizations of the classical notions of APN and maximum nonlinear Boolean functions to deal with the case of mappings from a finite group K to another one N with the possibility that one or both groups are non-Abelian.

cs.CR↗

A new APN function which is not equivalent to a power mapping

A new almost perfect nonlinear function (APN) on the finite field GF(2^10) which is not equivalent to any of the previously known APN mappings is constructed. This is the first example of an APN mapping which is not equivalent to a power mapping.

math.CO↗

On abelian $(2^{2m+1}(2^{m-1}+1), 2^m(2^m+1), 2^m)$-difference sets

In this paper we prove that an abelian group contains $(2^{2m+1}(2^{m-1}+1), 2^m(2^m+1), 2^m)$-difference sets with $m\geqslant 3$ if and only if it contains an elementary abelian 2-group of order $2^{2m}$. Our proof shows that the method of constructing such difference sets is essentially unique.

math.CO↗

A characterization of a class of maximum nonlinear functions

Maximum nonlinear functions on finite fields are widely used in cryptography because the coordinate functions have large distance to linear functions. More precisely, the Hamming distance to the characteristic functions of hyperplanes is large. One class of maximum nonlinear functions are the Gold power functions We characterize these functions in terms of the distance of their coordinate functions to characteristic functions of subspaces of codimension 2.

math.CO↗