Character theoretic techniques for nonabelian partial difference sets

Seth R. Nelson, Eric Swartz · 2025

Read Character theoretic techniques for nonabelian partial difference sets by Seth R. Nelson, Eric Swartz (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-23039.txt.

Introduction

Seth Nelson and Eric Swartz develop character methods for partial difference sets in nonabelian groups in a 2025 paper. A partial difference set is a subset of a group where each element factors through the set in one of two counts. Such sets build strongly regular graphs, a prized class in combinatorics.

Character theory worked well for the abelian case, and interest has shifted to harder nonabelian groups. The pair prove analogues of classic results and use them to count how a candidate set meets each conjugacy class. That yields proofs that no such set exists in many cases, tight limits in others, and fresh examples, including an endless family.

Catalog record

CREATORSeth R. Nelson, Eric Swartz
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE122 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-23039

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Character theoretic techniques for nonabelian partial difference sets Seth R. Nelson, Eric Swartz Abstract . A ( v , k , λ , μ ) (v,k,\lambda,\mu) -partial difference set (PDS) is a subset D D of size k k of a group G G of order v v such that every nonidentity element g g of G G can be expressed in either λ \lambda or μ \mu different ways as a product x ​ y − 1 xy^{-1} , x , y ∈ D x,y\in D , depending on whether or not g g is in D D . If D D is inverse closed and 1 ∉ D 1\notin D , then the Cayley graph Cay ⁡ ( G , D ) {\rm Cay}(G,D) is a ( v , k , λ , μ ) (v,k,\lambda,\mu) -strongly regular graph (SRG). PDSs have been studied extensively over the years, especially in abelian groups, where techniques from character theory have proven to be particularly effective. Recently, there has been considerable interest in studying PDSs in nonabelian groups, and the purpose of this paper is develop character theoretic techniques that apply in the nonabelian setting. We prove that analogues of character theoretic results of Ott [ 25 ] about generalized quadrangles of order s s also hold in the general PDS setting, and we are able to use these techniques to compute the intersection of a putative PDS with the conjugacy classes of the parent group in many instances. With these techniques, we are able to prove the nonexistence of PDSs in numerous instances and provide severe restrictions in cases when such PDSs may still exist. Furthermore, we are able to use these techniques constructively, computing several examples of PDSs in nonabelian groups not previously recognized in the literature, including an infinite family of genuinely nonabelian PDSs associated to the block-regular Steiner triple systems originally studied by Clapham [ 6 ] and related infinite families of genuinely nonabelian PDSs associated to the block-regular Steiner 2 2 -designs first studied by Wilson [ 33 ] . 1. Introduction If G is a group, then we say that D\subset G is a (v,k,\lambda,\mu) - partial difference set (PDS) if G has order v , D has size k , and if the product gh^{-1} for all g,h\in D with g\neq h represents all nonidentity elements in D exactly \lambda times and all nonidentity elements in G-D exactly \mu times. We refer to the tuple (v,k,\lambda,\mu) as the parameters of D . In addition, we call a PDS regular if D is closed under taking inverses (in which case we write D=D^{(-1)} ) and 1\not\in D . This definition is not particularly restrictive, for if D is a PDS with 1\in D , then D-\{1\} is a PDS as well, and if \lambda\neq\mu , then D=D^{(-1)} . If we have \lambda=\mu then D is called a difference set, and we say that D is a (v,k,\lambda) - difference set (DS). We additionally call D a reversible (v,k,\lambda) -DS if D=D^{(-1)} . PDSs have been the subject of great interest over the years (see, e.g., [ 24 ] ), especially in abelian groups. In the abelian group setting, character theoretic methods have proven to be particularly effective. For example, if we extend an irreducible character \chi for an abelian group G in the natural way to the group algebra \mathbb{C}[G] and identify an inverse-closed subset D\subset G with \sum_{d\in D}d\in\mathbb{C}[G] , then D is a (v,k,\lambda,\mu) -PDS if and only if (1) \chi(D)=\begin{cases}k,\text{ if $\chi$ is the trivial character,}\\ \frac{(\lambda-\mu)\pm\sqrt{\Delta}}{2},\text{ otherwise,}\end{cases} where \Delta\mathrel{\mathchoice{\vbox{\hbox{$\displaystyle:$}}}{\vbox{\hbox{$\textstyle:$}}}{\vbox{\hbox{$\scriptstyle:$}}}{\vbox{\hbox{$\scriptscriptstyle:$}}}{=}}(\lambda-\mu)^{2}+4(k-\mu) (see [ 24 , Corollary 3.3] ). There has been interest recently in PDSs in nonabelian groups [ 9 , 13 , 31 , 32 ] ; for a recent survey of known results related to PDSs in nonabelian groups, see [ 27 ] . However, most character theoretic results that apply in the abelian case do not hold in the nonabelian case: for example, while linear characters evaluate to the values listed in Equation ( 1 ) above, general irreducible characters do not necessarily need to evaluate to one of these values on a PDS (see Lemma 3.2 ). Perhaps more importantly, the characters evaluating to one of a few specific values on an inverse-closed subset is not a sufficient condition for the subset to be a PDS in the nonabelian case (only a necessary condition); see Remark 4.8 . Up until now, the only restrictions that apply in the nonabelian setting are those of [ 32 ] , which were inspired by the work of Yoshiara on groups acting regularly on the point set of a generalized quadrangle (GQ) [ 34 ] . The main method in each of [ 32 , 34 ] was to constrain the value \Phi(g) , which counts the number of vertices (resp., points) moved to adjacent vertices (resp., points) by an automorphism g in the automorphism group G of the SRG (resp., GQ). Yoshiara was able to utilize Benson’s Lemma [ 1 , Lemma 4.3] , which gave modular constraints on \Phi(g) for GQs. De Winter, Kamischke, and Wang [ 10 ] later proved a generalization of Benson’s Lemma for arbitrary SRGs, and Swartz and Tauscheck [ 32 ] used the generalization of De Winter, Kamischke, and Wang to push Yoshiara’s result into an analogous modular restriction on SRGs with a regular automorphism group. Recently, Ott [ 25 ] studied \Phi – which is indeed a class function on G – from a character-theoretic perspective, and Ott used this to prove that GQs of order s ( s+1 points incident with each line, s+1 lines incident with each point), s even, cannot have a group acting regularly on its point set. As it turns out, the ideas Ott used are applicable in situations far beyond GQs (and reversible difference sets); in fact, one could argue that the PDS setting is in fact where these techniques are most effective. The purpose of this paper is to extend Ott’s results to PDSs in nonabelian groups. (Indeed, just as techniques such as Equation ( 1 ) are only truly effective in the abelian setting, the results outlined here are most effective in certain nonabelian settings.) These techniques have enabled us to rule out the existence of a PDS inside a range of groups and to provide severe restrictions on certain families of PDSs. We also develop new techniques for discovering and constructing new examples of PDSs inside of nonabelian groups. For a given group G of order v , if there exists a parameter set (v,k,\lambda,\mu) which could potentially correspond to a PDS inside G , then these methods use character theory to give information about the size of the intersection of this potential PDS with every conjugacy class of G . We have used these methods to show that if a (v,k,\lambda,\mu) -PDS exists in a group G , where there is a minimal nontrivial normal subgroup N of G such that G/N is abelian and \gcd(|G/N|,\sqrt{\Delta})=1 , then the size of the intersection of the PDS with the cosets of N are known, providing severe restrictions on such PDSs. We have also used our methods

Read the complete text