KIT | KIT-Bibliothek | Impressum | Datenschutz

Generative Machine Learning for Resource-Aware 5G and IoT Systems

Piatkowski, Nico; Mueller-Roemer, Johannes S.; Hasse, Peter; Bachorek, Adam; Werner, Tim; Birnstill, Pascal; Morgenstern, Andreas; Stobbe, Lutz

Abstract (englisch):

Extrapolations predict that the sheer number of Internet-of-Things (IoT) devices will exceed 40 billion in the next five years. Hand-crafting specialized energy models and monitoring sub-systems for each type of device is error prone, costly, and sometimes infeasible. In order to detect abnormal or faulty behavior as well as inefficient resource usage autonomously, it is of tremendous importance to endow upcoming IoT and 5G devices with sufficient intelligence to deduce an energy model from their own resource usage data. Such models can in-turn be applied to predict upcoming resource consumption and to detect system behavior that deviates from normal states. To this end, we investigate a special class of undirected probabilistic graphical model, the so-called integer Markov random fields (IntMRF). On the one hand, this model learns a full generative probability distribution over all possible states of the system - allowing us to predict system states and to measure the probability of observed states. On the other hand, IntMRFs are themselves designed to consume as less resources as possible - e.g., faithful modelling of systems with an exponentially large number of states, by using only 8-bit unsigned integer arithmetic and less than 16KB memory. ... mehr


Download
Originalveröffentlichung
DOI: 10.1109/ICCWorkshops50388.2021.9473625
Scopus
Zitationen: 3
Dimensions
Zitationen: 1
Zugehörige Institution(en) am KIT Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung (IOSB)
Institut für Informationssicherheit und Verlässlichkeit (KASTEL)
Publikationstyp Proceedingsbeitrag
Publikationsmonat/-jahr 06.2021
Sprache Englisch
Identifikator ISBN: 978-1-72819-441-7
KITopen-ID: 1000140731
HGF-Programm 46.23.04 (POF IV, LK 01) Engineering Security for Production Systems
Erschienen in 2021 IEEE International Conference on Communications Workshops (ICC Workshops), Montreal, QC, Canada, 14-23 June 2021
Veranstaltung IEEE International Conference on Communications Workshops (ICC 2021), Montreal, Kanada, 14.06.2021 – 23.06.2021
Verlag Institute of Electrical and Electronics Engineers (IEEE)
Seiten 1–6
Schlagwörter 5G core; energy model; generative model; internet of things; probabilistic graphical model
Nachgewiesen in Dimensions
Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page