English

Node Connectivity Augmentation of Highly Connected Graphs

Data Structures and Algorithms 2023-11-29 v1

Abstract

Node-connectivity augmentation is a fundamental network design problem. We are given a kk-node connected graph GG together with an additional set of links, and the goal is to add a cheap subset of links to GG to make it (k+1)(k+1)-node connected. In this work, we characterize completely the computational complexity status of the problem, by showing hardness for all values of kk which were not addressed previously in the literature. We then focus on kk-node connectivity augmentation for k=n4k=n-4, which corresponds to the highest value of kk for which the problem is NP-hard. We improve over the previously best known approximation bounds for this problem, by developing a 32\frac{3}{2}-approximation algorithm for the weighted setting, and a 43\frac{4}{3}-approximation algorithm for the unweighted setting.

Keywords

Cite

@article{arxiv.2311.17010,
  title  = {Node Connectivity Augmentation of Highly Connected Graphs},
  author = {Waldo Galvez and Dylan Hyatt-Denesik and Afrouz Jabal Ameli and Laura Sanita},
  journal= {arXiv preprint arXiv:2311.17010},
  year   = {2023}
}
R2 v1 2026-06-28T13:34:28.072Z