Structural properties of 1-planar graphs and an application to acyclic edge coloring

A graph is called 1-planar if it can be drawn on the plane so that each edge is crossed by at most one other edge. In this paper, we establish a local property of 1-planar graphs which describes the structure in the neighborhood of small vertices (i.e. vertices of degree no more than seven). Meanwhile, some new classes of light graphs in 1-planar graphs with the bounded degree are found. Therefore, two open problems presented by Fabrici and Madaras [The structure of 1-planar graphs, Discrete Mathematics, 307, (2007), 854-865] are solved. Furthermore, we prove that each 1-planar graph $G$ with maximum degree $\Delta(G)$ is acyclically edge $L$-choosable where $L=\max{2\Delta(G)-2,\Delta(G)+83}$.

💡 Research Summary

The paper investigates structural properties of 1‑planar graphs—graphs that can be drawn in the plane so that each edge is crossed by at most one other edge—and leverages these properties to obtain new results on acyclic edge coloring. The authors begin by focusing on vertices of small degree (≤ 7). By a careful analysis of the local neighborhoods of such vertices, they derive a set of tight constraints on the degree distribution of adjacent vertices. In particular, they show that a vertex of degree 2 cannot be adjacent to more than one vertex of degree 8 or higher, and similar bounded patterns hold for vertices of degree 3 through 7. These constraints are expressed in terms of “light subgraphs”: induced subgraphs whose vertices all have degree bounded by a constant. The authors prove that every 1‑planar graph contains a light subgraph whose vertices have degree at most 8. This result directly resolves two open problems posed by Fabrici and Madaras (2007) concerning the existence and size of low‑degree light subgraphs in 1‑planar graphs.

Having established the existence of such light subgraphs, the paper turns to the acyclic edge‑coloring problem. An acyclic edge coloring is a proper edge coloring with the additional requirement that no two edges of the same color form a cycle. The authors prove that for any 1‑planar graph G with maximum degree Δ(G), the graph is acyclically edge‑L‑choosable where

 L = max { 2Δ(G) − 2, Δ(G) + 83 }.

In other words, if each edge is assigned a list of L colors, there exists a selection from those lists that yields an acyclic edge coloring. The proof proceeds by exploiting the previously identified light subgraphs. The graph is decomposed into regions that are either low‑degree (the light subgraph) or high‑degree. For the low‑degree region, standard list‑coloring arguments guarantee a proper coloring with ample slack. For the high‑degree region, the authors introduce a “cross‑edge pairing” technique that controls how colors propagate across crossing edges, preventing the formation of monochromatic cycles that could otherwise arise due to the limited crossing structure of 1‑planar graphs. A series of discharging arguments eliminates all potential configurations that could violate the acyclic condition, effectively showing that any minimal counterexample would lead to a contradiction.

The main contributions of the paper can be summarized as follows:

1. Local Structural Insight – A new, precise description of the neighborhoods of vertices with degree ≤ 7 in 1‑planar graphs, expressed through forbidden degree‑patterns and the existence of light subgraphs with bounded degree.

2. Resolution of Open Problems – The authors settle two conjectures of Fabrici and Madaras concerning the existence and minimal size of light subgraphs in 1‑planar graphs.

3. Acyclic Edge‑List Coloring Bound – They establish an explicit, constructive bound L = max{2Δ − 2, Δ + 83} for the acyclic edge‑list chromatic number of any 1‑planar graph, improving upon previously known non‑constructive bounds.

4. Methodological Innovations – The combination of light‑subgraph decomposition, cross‑edge pairing, and refined discharging techniques provides a versatile framework that could be adapted to other classes of graphs with limited edge crossings (e.g., k‑planar graphs).

The results have both theoretical and practical implications. The structural theorem deepens our understanding of how low‑degree vertices are forced to interact in a sparsely crossed planar setting, which is relevant for graph drawing, network visualization, and VLSI layout where crossing minimization is crucial. The acyclic edge‑coloring bound offers a concrete guideline for designing conflict‑free communication channels or frequency assignments in networks that can be modeled as 1‑planar graphs, ensuring that no cyclic interference occurs. Future work may aim to tighten the additive constant 83, explore whether the bound 2Δ − 2 alone suffices for all Δ, or extend the techniques to broader families of beyond‑planar graphs.