We prove local L p -Poincaré inequalities, p ∈ [1, ∞], on quasiconvex sets in infinite graphs endowed with a family of locally doubling measures, and global L p -Poincaré inequalities on connected sets for flow measures on trees. We also discuss the optimality of our results.
Introduction and notationPoincaré inequalities have played an important role in analysis on noncompact manifolds, Lie groups, infinite graphs, and more generally on metric spaces. A complete overview of the literature on this subject would go out of the scope of this paper, so we refer the reader to [6,8] and the references therein.While the key role of these inequalities is largely understood, and their validity is considered a natural geometric assumption in the context of analysis in metric spaces, on the other hand, quite surprisingly, in the discrete setting there is a lack of concrete examples of graphs fulfilling these inequalities in the literature. In this note we focus on infinite graphs and we prove that some concrete common families of measures and graphs actually fulfil the desired property of supporting Poincaré-type inequalities.(Local) Poincaré inequalities combined with the (local) doubling condition, are a standard tool to obtain (local) Harnack inequalities both in continuous and discrete settings (see [4,5,14,16]). Applications of the Poincaré inequalities discussed in this note in such direction will be object of further investigation in [11].Let X be an infinite, locally finite, connected, and undirected graph. We identify X with its set of vertices and we write x ∼ y whenever x, y ∈ X are neighbors, namely, when they are connected by an edge. We denote by deg(x) the number of neighbors of x. We say that the graph X has bounded degree b + 1 if deg(x) ≤ b + 1, for some b ≥ 1 and every x ∈ X. A path of length n ∈ N connecting two vertices x and y is a sequence {x 0 , x 1 , . . . , x n } ⊂ X, with no repeated vertices, such that x 0 = x, x n = y, and x i ∼ x i+1 for every i = 0, . . . , n − 1. The distance d(x, y) is defined as the minimum of the lengths of the paths connecting x and y. For any x ∈ X and r ≥ 0, the ball of radius r and center x is B r (x) = {y ∈ X : d(x, y) ≤ r}. For every subset E of X the diameter of E is diam(E) = sup{d(x, y) : x, y ∈ E}.