English

Broadcast Dimension of Graphs

Combinatorics 2020-05-18 v1 Discrete Mathematics

Abstract

In this paper we initiate the study of broadcast dimension, a variant of metric dimension. Let GG be a graph with vertex set V(G)V(G), and let d(u,w)d(u,w) denote the length of a uwu-w geodesic in GG. For k1k \ge 1, let dk(x,y)=min{d(x,y),k+1}d_k(x,y)=\min \{d(x,y), k+1\}. A function f:V(G)Z+{0}f: V(G) \rightarrow \mathbb{Z}^+ \cup \{0\} is called a resolving broadcast of GG if, for any distinct x,yV(G)x,y \in V(G), there exists a vertex zV(G)z \in V(G) such that f(z)=i>0f(z)=i>0 and di(x,z)di(y,z)d_{i}(x,z) \neq d_{i}(y,z). The broadcast dimension, bdim(G)bdim(G), of GG is the minimum of cf(G)=vV(G)f(v)c_f(G)=\sum_{v \in V(G)} f(v) over all resolving broadcasts of GG, where cf(G)c_f(G) can be viewed as the total cost of the transmitters (of various strength) used in resolving the entire network described by the graph GG. Note that bdim(G)bdim(G) reduces to adim(G)adim(G) (the adjacency dimension of GG, introduced by Jannesari and Omoomi in 2012) if the codomain of resolving broadcasts is restricted to {0,1}\{0,1\}. We determine its value for cycles, paths, and other families of graphs. We prove that bdim(G)=Ω(logn)bdim(G) = \Omega(\log{n}) for all graphs GG of order nn, and that the result is sharp up to a constant factor. We show that adim(G)bdim(G)\frac{adim(G)}{bdim(G)} and bdim(G)dim(G)\frac{bdim(G)}{dim(G)} can both be arbitrarily large, where dim(G)dim(G) denotes the metric dimension of GG. We also examine the effect of vertex deletion on the adjacency dimension and the broadcast dimension of graphs.

Cite

@article{arxiv.2005.07311,
  title  = {Broadcast Dimension of Graphs},
  author = {Jesse Geneson and Eunjeong Yi},
  journal= {arXiv preprint arXiv:2005.07311},
  year   = {2020}
}
R2 v1 2026-06-23T15:33:46.719Z