Work

Probabilistic Graph Grammars for Generating Realistic Network Topologies

Year: 2026
Type: preprint
Abstract: This paper proposes a novel approach to generating realistic network topologies using Probabilistic Graph Grammars (PGGs). Traditional methods for network topology generation often rely on deterministic rules, leading to networks that lack the complexity and diversity observed in real-world networks. PGGs introduce probabilistic constraints into the rule application process, allowing for a more nuanced and statistically-driven generation of networks. The core claim is that generating realistic network topologies requires sophisticated modeling techniques. The proposed mechanism involves developing a PGG that defines rules for node and edge creation and modification, incorporating probabilities that reflect the expected relationships and dependencies within a given network type. The resulting networks are suitable for simulation and analysis scenarios, offering a significant improvement over purely deterministic approaches. The paper outlines the theoretical framework, key components of the PGG, and discusses potential applications.
Language: English
Open Access status: green
OpenAlex ID: W7208814503
Cites:
Cited by:
Related to: 10