Edge-symmetric distance-regular covers of cliques with \(\lambda=\mu\) (Q352231)

The authors deal with undirected graphs without loop or multiple edges. The paper begins with definitions like: \(i\)-neighborhood, degree of a graph, regular graph of degree \(k\), edge-regular graph with parameters \((\nu,k,\lambda)\), amply regular graph with parameters \((\nu,k,\lambda,\mu)\), strongly regular graph, distance-regular graph, intersection number of \(\Gamma\), edge-symmetric graph, antipodal distance-transitive graph of diameter 3 chosen from Godsil et al. As the authors indicate, `a program is proposed' for the study of edge-symmetric antipodal distance-regular graphs of diameter 3 as based on a classification of twice transitive permutation groups. Part of the program for graphs with \(\lambda=\mu\) is implemented. It begins with some equivalent conditions such as: Let \(\Gamma\) be an antipodal distance-regular graph of diameter 3, \(G=\Aut(\Gamma)\), \(\Sigma\) be the set of all antipodal classes of \(\Gamma\), and \(F\in\Sigma\) then \((1) \equiv (2)\), where {\parindent=6mm \begin{itemize}\item[(1)]\(\Gamma\) is an edge-symmetric graph and \(G\) acts twice transitively on \(\Sigma\) \item[(2)]\(\Gamma\) is a vertex-symmetric graph, and for a vertex \(a \in F\), the group \(G_a\) acts transitively on \(\Sigma - \{F\}\). \end{itemize}} The authors cite a proposition from Godsil et al. for a twice transitive permutation group of degree \(n\), with \(H=G_a\) and the socle \(T\) of \(G\). The new infinite series of antipodal distance-regular graphs of diameter 3 are: Suzuki series (in connection with graph \(Suz(q))\) and Ree series (In connection with graph \(Ree(q))\), and unitary series (in connection with graph \({\mathcal V}_3(q)\) and these series come forward for a necessary explanation. The paper carries basically two main theorems, the proofs of which are guided by four lemmas. One of these is already to be found in Cameron's algebraic number theory. The following two main theorems are as follows: Theorem 1. Let \(\Gamma\) be an edge-symmetric distance-regular graph with intersection array \(\{k,(r-1)\mu,1;1,\mu,k\}\) and \(G=\Aut(\Gamma)\). Then the following assertions hold: {\parindent=6mm \begin{itemize} \item[(1)]If \(\Gamma\) is a bipartite graph, then \(\Gamma\) is obtained from \(K_{k+1,k+1}\) by removing the maximal matching, \(G \leq 2\times S_{k+1}\), \item[(2)]If \(r=k\), then either \(k=2\) and \(\Gamma\) is a hexagon or \(k=6\), \(\Gamma\) is a second neighborhood of a vertex in the Hoffman-Singleton graph, and \(G\leq S_7\), \item[(3)]If \(r=2\), then one of the following cases hold: \begin{itemize} \item[(i)]\(k+1= 2^{2m-1}\pm 2^{m-1}\), \(G\leq 2\times Sp_{2m}(2)\), \(m \geq 3\); \item[(ii)]\(k=q^3\), \(G\leq 2\times P\Gamma U_3(q)\), \(q>3\) or \(G\leq 2\times\Aut (^2G_2(q))\), \(q=3^{2e+1}\), \(e\geq 1\): \item[(iii)]\(k=q\), \(G\leq 2\times P\Sigma L_2(q)\), \(q \equiv 1\pmod 4\); \item[(iv)]\(k=175\), \(G=2\times HiSork = 275\), \(G=2\times Co_3\); or \item[(v)]\(k=2^{2t-1}\), \(G=2.AS p_{2t}(2)\) \end{itemize} \end{itemize}} Theorem 2: Let \(\Gamma\) be an edge-symmetric distance-regular graph with intersection array \(\{r\mu+1,(r-1)\mu,1;1, \mu,r\mu\}\) and \(G=\Aut(\Gamma)\). If \(r>2\), then one of the following assertions holds: {\parindent=6mm \begin{itemize}\item[(1)]\(k=q\) is a power of a prime number, \(r\) divides \(q-1\), has the intersection array \(\{q,\frac{(r-1)(q-1)}{r},1;1,\frac{q-1}{R},q\}\), and \(L_2(q)\Delta\mu\) \item[(2)]\(\Gamma\in\{Suz(q,s), Ree(q,s), {\mathcal{V}}_3(q,s)\}\) \end{itemize}} Lemma 1 is a result available in Neukirch's `algebraic number theory', some assertion of Lemma 3 follow from Cameron's `permutation groups' (lemma 1). The proof of theorem 2 relies on Higman's method for handling automorphisms of a distance regular graph as seen in Cameron's `permutation groups' (chapter 3). Lemma 4 is the theorem 2.5 of \textit{C. Godsil} and \textit{A.D. Hensel} [J. Comb. Theory, Ser. B 56, No.\,2, 205--238 (1992; Zbl 0771.05031)], theorem 1 takes help of corollary 2.3 [Godsil et. al., loc.cit.] and a theorem from the authors' result.