Lexicographic products and lexicographic powers of graphs -- a walk matrix approach

The characteristic polynomial and the spectrum of the lexicographic product of graphs $H[G]$, a specific instance of the generalized composition (also called $H$-join), are explicitly determined for arbitrary graphs $H$ and $G$, in terms of the eigenvalues of $G$ and an $H[G]$ associated matrix $\widetilde{{\bf W}}$, which relates $H$ with $G$. This study also establishes conditions under which a main eigenvalue of $G$ is a main or non-main eigenvalue of the matrix $\widetilde{{\bf W}}$, when the nullity of the graph $H$ is $\eta>0$. In such a case, we prove that every main eigenvalue of $G$ is an eigenvalue of $\widetilde{{\bf W}}$ with multiplicity at least $\eta$ which is non-main for $\bf \widetilde{W}$ if and only if $0$ is a non-main eigenvalue of $H$. Furthermore, the spectra of the lexicographic powers of arbitrary graphs $G$ are analysed by applying the obtained results.