Degree-based approximations for network reliability polynomials

Journal Article (2025)
Author(s)

P.F.A. Van Mieghem (TU Delft - Network Architectures and Services)

X. Liu (TU Delft - Network Architectures and Services)

Research Group
Network Architectures and Services
DOI related publication
https://doi.org/10.1093/comnet/cnaf001
More Info
expand_more
Publication Year
2025
Language
English
Research Group
Network Architectures and Services
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository as part of the Taverne amendment. More information about this copyright law amendment can be found at https://www.openaccess.nl. Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. @en
Issue number
2
Volume number
13
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Two approximations for network reliability polynomials, only based upon the knowledge of the degree vector of the graph, are compared: the first-order approximation by Brown et al. and our stochastic approximation. Our method is an extension of the connectivity probability of Erdős–Rènyi random graphs. Both approximations are shown to upper bound the actual reliability polynomial and are increasingly accurate for dense and large graphs. Moreover, the first-order approximation is always sharper or at least as good as the stochastic approximation, whereas the stochastic approximation is computationally easier. Our stochastic approximation (2.2) can determine the critical operational probability under which the graph is disconnected almost surely for any graph and an approximation for the number Fj of sets of j links whose removal retains the graph G connected, which is helpfull because the exact computation of Fj is NP-hard.

Files

JCN2024_ReliabilityPolynomials... (pdf)
(pdf | 12.2 Mb)
- Embargo expired in 29-09-2025
License info not available