Fast and Configurable Detection of Device Dependencies in
Network Traffic

D 2025

Fast and Configurable Detection of Device Dependencies in Network Traffic

DUSIL, Jakub; Martin HUSÁK a Lukáš SADLEK

Základní údaje

Originální název

Fast and Configurable Detection of Device Dependencies in Network Traffic

Autoři

DUSIL, Jakub; Martin HUSÁK a Lukáš SADLEK

Vydání

New York, NY, 21st International Conference on Network and Service Management, od s. 1-6, 6 s. 2025

Nakladatel

IEEE

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10200 1.2 Computer and information sciences

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

URL

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14330/25:00142396

Organizační jednotka

Fakulta informatiky

ISBN

978-3-903176-75-1

DOI

https://doi.org/10.23919/CNSM67658.2025.11297543

EID Scopus

2-s2.0-105032094389

Klíčová slova anglicky

network traffic analysis; device dependency; dependency detection; link prediction; graph-based machine learning

Štítky

rivok

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 1. 4. 2026 11:10, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Device dependencies are recurring communication patterns between IP addresses that reveal how networked entities rely on one another. Understanding these relationships is essential for reliability, troubleshooting, and security, yet detecting them efficiently from operational traffic remains challenging. We propose a fast and accurate tool for dependency detection from passive flow-level data using a link prediction approach. In contrast to the prior implementation, the tool introduces a parallelized processing pipeline with early termination of stalled random walks, an expanded feature set that combines embedding-derived and graph-theoretic metrics, and a fully externalized configuration of sampling, embedding, and classification parameters. These design choices enable scalable execution and more reliable identification of dependencies across diverse network environments. Evaluation on synthetic traffic from cyber-defense exercises and real-world campus flows demonstrates up to 100$\times$ faster runtime and markedly higher classification accuracy compared to the prior implementation. Further analysis shows that structural graph features improve stability in sparse settings, while extended embedding training enhances accuracy in low-signal scenarios. Together, these results confirm that the proposed tool advances link prediction-based dependency detection toward practical, near-real-time use.

Přiložené soubory

2025-CNSM-Dependency_detection-paper.pdf Verze souboru

2025-CNSM-Dependency_detection-slides.pdf

Citovat

DUSIL, Jakub; Martin HUSÁK a Lukáš SADLEK. Fast and Configurable Detection of Device Dependencies in Network Traffic. Online. In Walter Cerroni, Mauro Tortonesi, Davide Borsatti, Alberto Egon Schaeffer-Filho, Daphné Tuncer, Jérôme François, Martin Husák. 21st International Conference on Network and Service Management. New York, NY: IEEE, 2025, s. 1-6. ISBN 978-3-903176-75-1. Dostupné z: https://doi.org/10.23919/CNSM67658.2025.11297543.

@inproceedings{2529244,
   author = {Dusil, Jakub and Husák, Martin and Sadlek, Lukáš},
   address = {New York, NY},
   booktitle = {21st International Conference on Network and Service Management},
   doi = {https://doi.org/10.23919/CNSM67658.2025.11297543},
   editor = {Walter Cerroni, Mauro Tortonesi, Davide Borsatti, Alberto Egon Schaeffer-Filho, Daphné Tuncer, Jérôme François, Martin Husák},
   keywords = {network traffic analysis; device dependency; dependency detection; link prediction; graph-based machine learning},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {New York, NY},
   isbn = {978-3-903176-75-1},
   pages = {1-6},
   publisher = {IEEE},
   title = {Fast and Configurable Detection of Device Dependencies in Network Traffic},
   url = {https://opendl.ifip-tc6.org/db/conf/cnsm/cnsm2025/1571195022.pdf},
   year = {2025}
}

TY  - CONF
ID  - 2529244
AU  - Dusil, Jakub - Husák, Martin - Sadlek, Lukáš
PY  - 2025
TI  - Fast and Configurable Detection of Device Dependencies in Network Traffic
PB  - IEEE
CY  - New York, NY
SN  - 9783903176751
KW  - network traffic analysis
KW  - device dependency
KW  - dependency detection
KW  - link prediction
KW  - graph-based machine learning
UR  - https://opendl.ifip-tc6.org/db/conf/cnsm/cnsm2025/1571195022.pdf
N2  - Device dependencies are recurring communication patterns between IP addresses that reveal how networked entities rely on one another. Understanding these relationships is essential for reliability, troubleshooting, and security, yet detecting them efficiently from operational traffic remains challenging. We propose a fast and accurate tool for dependency detection from passive flow-level data using a link prediction approach. In contrast to the prior implementation, the tool introduces a parallelized processing pipeline with early termination of stalled random walks, an expanded feature set that combines embedding-derived and graph-theoretic metrics, and a fully externalized configuration of sampling, embedding, and classification parameters. These design choices enable scalable execution and more reliable identification of dependencies across diverse network environments. Evaluation on synthetic traffic from cyber-defense exercises and real-world campus flows demonstrates up to 100$\times$ faster runtime and markedly higher classification accuracy compared to the prior implementation. Further analysis shows that structural graph features improve stability in sparse settings, while extended embedding training enhances accuracy in low-signal scenarios. Together, these results confirm that the proposed tool advances link prediction-based dependency detection toward practical, near-real-time use.
ER  -

DUSIL, Jakub; Martin HUSÁK a Lukáš SADLEK. Fast and Configurable Detection of Device Dependencies in Network Traffic. Online. In Walter Cerroni, Mauro Tortonesi, Davide Borsatti, Alberto Egon Schaeffer-Filho, Daphné Tuncer, Jérôme Fran\c cois, Martin Husák. \textit{21st International Conference on Network and Service Management}. New York, NY: IEEE, 2025, s.~1-6. ISBN~978-3-903176-75-1. Dostupné z: https://doi.org/10.23919/CNSM67658.2025.11297543.

Přehled o publikaci