Code and Data for: Safe Fusion in Self-Organizing Networks
- Daniel Nicolas Hofstadler(Creator),
- Daniel Reisinger(Creator),
- Thomas Schmickl(Creator),
- Payam Zahadat(Creator)
- ,
- ,
- ,
- ,
- University of Graz
Dataset:
Dataset Types
Dataset
Dataset Information
Date made available
07/10/2026Publisher
ZENODODescription
This repository contains the complete supplementary source code, pre-computed datasets, mathematical derivations, and batch simulation scripts required to reproduce the findings presented in our manuscript: "Safe Fusion in Self-Organizing Networks: Stabilizing Multi-Parent Connectivity via Hysteresis".
Abstract
The evolution of resilient, self-organizing physical networks relies on decentralized algorithms to optimize resource distribution. While existing paradigms excel at routing through strict spanning trees, they struggle with anastomosis: the fusion of independent branches into multi-parent pathways. Within adaptive flow networks, unbuffered fusions risk severe symmetry breaking, where positive-feedback loops quickly starve redundant infrastructure. To address this, we introduce FusionVMC, extending the bio-inspired Vascular Morphogenesis Controller to safely manage multi-parent topologies. By analyzing an exemplary minimal fusion system, we demonstrate that a cross-coupled Resource-Dependent Distribution (RDD) scheme achieves profound dynamic resilience. RDD links demand with physical delivery, creating a mathematical hysteresis buffer that prevents instantaneous symmetry breaking and enables stable, proportional load-sharing. Furthermore, by modulating the algorithmic competition exponent, FusionVMC allows networks to seamlessly tune their physical topology between structural redundancy (exploration) and decisive pruning (exploitation).
File Manifest
fusion-vmc-mfs.zip: The complete, ready-to-run repository. It is logically separated into mathematical derivations (Jupyter notebook with sympy proofs) and numerical simulations mapping the bistability and transient dynamics of the Minimal Fusion System (MFS).
README.md: The top-level documentation detailing the repository structure and setup instructions.
environment.yml: The conda/mamba environment configuration file containing all required dependencies to ensure strict reproducibility.
LICENSE: The MIT License explicitly permitting open academic use and modification.
Quick Start
To run the code, download and extract the .zip archive, navigate to the root folder in your terminal, and create the environment using mamba env create -f environment.yml. Refer to the included README.md for detailed execution instructions.
Cite this dataset
Brief description
Daniel Nicolas Hofstadler, Daniel Reisinger, Thomas Schmickl, Payam Zahadat. (2026). Code and Data for: Safe Fusion in Self-Organizing Networks. 10.5281/zenodo.20415386
Funding Details
FunderFunding number
European Commission
640959

