Extremal graphs for the identifying code problem
arXiv:1004.5230 · doi:10.1016/j.ejc.2011.01.002
Abstract
An identifying code of a graph G is a dominating set C such that every vertex x of G is distinguished from all other vertices by the set of vertices in C that are at distance at most 1 from x. The problem of finding an identifying code of minimum possible size turned out to be a challenging problem. It was proved by N. Bertrand that if a graph on n vertices with at least one edge admits an identifying code, then a minimum identifying code has size at most n-1. Some classes of graphs whose smallest identifying code is of size n-1 were already known, and few conjectures were formulated to classify all these graphs. In this paper, disproving these conjectures, we classify all finite graphs for which all but one of the vertices are needed to form an identifying code. We also classify all infinite graphs needing the whole set of vertices in any identifying code. New upper bounds in terms of the number of vertices and the maximum degree of a graph are also provided.
Cited by in corpus (11)
- Locating-dominating sets in twin-free graphs
- Identifying codes in line graphs
- On the size of identifying codes in triangle-free graphs
- Location-domination and matching in cubic graphs
- Characterizing extremal graphs for open neighbourhood location-domination
- Characterizing extremal digraphs for identifying codes and extremal cases of Bondy's theorem on induced subsets
- Revisiting and improving upper bounds for identifying codes
- Bounds and extremal graphs for total dominating identifying codes
- Random subgraphs make identification affordable
- The RED-BLUE SEPARATION problem on graphs
- Identifying codes in graphs of given maximum degree: Characterizing trees