English

Generalized spectral characterization of mixed graphs

Combinatorics 2019-12-02 v1

Abstract

A mixed graph GG is a graph obtained from a simple undirected graph by orientating a subset of edges. GG is self-converse if it is isomorphic to the graph obtained from GG by reversing each directed edge. For two mixed graphs GG and HH with Hermitian adjacency matrices A(G)A(G) and A(H)A(H), we say GG is R\mathbb{R}\emph{-cospectral} to HH if, for any yRy\in \mathbb{R}, yJA(G)yJ-A(G) and yJA(H)yJ-A(H) have the same spectrum, where JJ is the all-one matrix. A self-converse mixed graph GG is said to be determined by its generalized spectrum, if any self-converse mixed graph that is RR-cospectral with GG is isomorphic to GG. Let GG be a self-converse mixed graph of order nn such that 2n/2detW2^{-\lfloor n/2\rfloor}\det W (which is always a real or pure imaginary Gaussian integer) is square-free in Z[i]\mathbb{Z}[i], where W=[e,Ae,,An1e]W=[e,Ae,\ldots,A^{n-1}e], A=A(G)A=A(G) and ee is the all-one vector. We prove that, for any self-converse mixed graph HH that is R\mathbb{R}-cospectral to GG, there exists a Gaussian rational unitary matrix UU such that Ue=eUe=e, UA(G)U=A(H)U^*A(G)U=A(H) and (1+i)U(1+i)U is a Gaussian integral matrix. In particular, if GG is an ordinary graph (viewed as a mixed graph) satisfying the above condition, then any self-converse mixed graph HH that is R\mathbb{R}-cospectral to GG is GG itself (in the sense of isomorphism). This strengthens a recent result of the first author.

Keywords

Cite

@article{arxiv.1911.13004,
  title  = {Generalized spectral characterization of mixed graphs},
  author = {Wei Wang and Lihong Qiu and Jianguo Qian and Wei Wang},
  journal= {arXiv preprint arXiv:1911.13004},
  year   = {2019}
}

Comments

22 pages,1 figure

R2 v1 2026-06-23T12:30:47.344Z