Table of Contents
Fetching ...

On the $Q$-polynomial property of bipartite graphs admitting a uniform structure

Blas Fernández, Roghayeh Maleki, Štefko Miklavič, Giusy Monzillo

TL;DR

The paper develops a framework to certify Q-polynomial structure in graphs with a uniform bipartite framework by introducing dual adjacency matrix candidates $A^*(\theta^*_0,\dots,\theta^*_\varepsilon)$ and a tridiagonal relation linking $A$ and $A^*$. It leverages Terwilliger algebras, the lowering/raising operators, and a uniform structure $(U,f)$ to derive sufficient conditions under which such an $A^*$ exists, along with explicit parameter relations $\beta,\gamma,\rho$. As a key application, it proves that the full bipartite graphs of dual polar graphs are $Q$-polynomial, providing explicit eigenvalues and dual eigenstructure. The results extend Terwilliger's approach to broader classes of graphs beyond distance-regular ones and offer a concrete method to certify the $Q$-polynomial property for bipartite, uniformly structured graphs.

Abstract

Let $Γ$ denote a finite, connected graph with vertex set $X$. Fix $x \in X$ and let $\varepsilon \ge 3$ denote the eccentricity of $x$. For mutually distinct scalars $\{θ^*_i\}_{i=0}^\varepsilon$ define a diagonal matrix $A^*=A^*(θ^*_0, θ^*_1, \ldots, θ^*_{\varepsilon}) \in M_X(\mathbb{R})$ as follows: for $y \in X$ we let $(A^*)_{yy} = θ^*_{\partial(x,y)}$, where $\partial$ denotes the shortest path length distance function of $Γ$. We say that $A^*$ is a dual adjacency matrix candidate of $Γ$ with respect to $x$ if the adjacency matrix $A \in M_X(\mathbb{R})$ of $Γ$ and $A^*$ satisfy $$ A^3 A^* - A^* A^3+(β+1)( A A^* A^2 - A^2 A^* A)= γ(A^2A^*-A^*A^2)+ρ( A A^* - A^* A) $$ for some scalars $β, γ, ρ\in \mathbb{R}$. Assume now that $Γ$ is uniform with respect to $x$ in the sense of Terwilliger [Coding theory and design theory, Part I, IMA Vol. Math. Appl., 20, 193-212 (1990)]. In this paper, we give sufficient conditions on the uniform structure of $Γ$, such that $Γ$ admits a dual adjacency matrix candidate with respect to $x$. As an application of our results, we show that the full bipartite graphs of dual polar graphs are $Q$-polynomial.

On the $Q$-polynomial property of bipartite graphs admitting a uniform structure

TL;DR

The paper develops a framework to certify Q-polynomial structure in graphs with a uniform bipartite framework by introducing dual adjacency matrix candidates and a tridiagonal relation linking and . It leverages Terwilliger algebras, the lowering/raising operators, and a uniform structure to derive sufficient conditions under which such an exists, along with explicit parameter relations . As a key application, it proves that the full bipartite graphs of dual polar graphs are -polynomial, providing explicit eigenvalues and dual eigenstructure. The results extend Terwilliger's approach to broader classes of graphs beyond distance-regular ones and offer a concrete method to certify the -polynomial property for bipartite, uniformly structured graphs.

Abstract

Let denote a finite, connected graph with vertex set . Fix and let denote the eccentricity of . For mutually distinct scalars define a diagonal matrix as follows: for we let , where denotes the shortest path length distance function of . We say that is a dual adjacency matrix candidate of with respect to if the adjacency matrix of and satisfy for some scalars . Assume now that is uniform with respect to in the sense of Terwilliger [Coding theory and design theory, Part I, IMA Vol. Math. Appl., 20, 193-212 (1990)]. In this paper, we give sufficient conditions on the uniform structure of , such that admits a dual adjacency matrix candidate with respect to . As an application of our results, we show that the full bipartite graphs of dual polar graphs are -polynomial.

Paper Structure

This paper contains 10 sections, 19 theorems, 63 equations.

Key Result

Lemma 2.3

(See projective.) We have $E^*_i A E^*_j = 0$ if $|i - j| > 1 \; (0 \le i, j \le \varepsilon)$. Assume that $\Gamma$ has a dual adjacency matrix $A^*$ with respect to $x$ and $\{E_i\}_{i=0}^{\cal D}$. Then, $E_i A^* E_j = 0$ if $|i - j| > 1 \; (0 \le i, j \le {\cal D})$.

Theorems & Definitions (28)

  • Definition 2.1
  • Definition 2.2
  • Lemma 2.3
  • Definition 3.1
  • Lemma 3.2
  • Theorem 3.3
  • Definition 4.1
  • Definition 4.2
  • Lemma 4.3
  • Definition 5.1
  • ...and 18 more