Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Industrial Network Topology Analysis with Episode Mining

: Meshram, Ankush

Fulltext urn:nbn:de:0011-n-5521928 (4.7 MByte PDF)
MD5 Fingerprint: 51d8b308ab8156d92b1345057a77ea60
Created on: 19.7.2019

Beyerer, Jürgen (Ed.); Taphanel, Miro (Ed.); Taphanel, Miro (Ed.):
Proceedings of the 2018 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory
Karlsruhe: KIT Scientific Publishing, 2019 (Karlsruher Schriften zur Anthropomatik 40)
ISBN: 978-3-7315-0936-3
ISBN: 3-7315-0936-9
DOI: 10.5445/KSP/1000094782
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation and Institute for Anthropomatics, Vision and Fusion Laboratory (Joint Workshop) <2018, Triberg-Nussbach>
Conference Paper, Electronic Publication
Fraunhofer IOSB ()

Industrial network communication is highly deterministic as result of availability requirement of control systems in automated industrial production systems. This deterministic character helps with initial step of self-learning anomaly detection systems to detect periodic production cycle in industrial network communication. The methods for frequent episode mining in event sequences fits well to solve the challenge of production cycle detection for self-learning system. We encode the network communication events to serial and parallel episodes. Methods for discovery of frequent episodes in event sequences are briefly explained. These methods would be further adapted in future to our encoded network communication traffic to extract production cycle comprised of serial and parallel episodes.