UDC 004.724
METHODS FOR OBTAINING PHYSICAL LEVEL DATA OF LATEST GENERATION NETWORKS TO INVESTIGATE CYBER INCIDENTS DURING ADVERSARY ATTACKS
Y. A. Ushakov, Ph.D. (in technical sciences), associate professor, OSU, Orenburg, Russia;
orcid.org/0000-0002-0474-8919, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
M. V. Ushakova, associate professor, OSU, Orenburg, Russia;
orcid.org/0000-0003-4462-9946, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The latest generation of data networks, such as 5G and 6G, is used as part of technical means to provide the core of virtual networks and network functions, while some solutions went further and began to use machine learning to optimize functions including the physical level. Intelligent models used in machine learning are currently modified by adversarial attacks, which result in the removal of minor changes in input data, while the result of the model can be radically changed in contrast to what was expected. Many works are devoted to detecting adversarial attacks, but all of them require internal input information for analysis. Using machine learning in radio path to generate directional radiation in modern ultra-massive MIMO, reconfigurable smart interfaces and other technologies used in networks of latest generation to receive data from the physical layers of sources, as well as subsequently process them due to their volume can be quite hard. The aim of the work is to create a simulation model combined with real machine learning applications to study the issue of obtaining attack data directly from industrial sources of service information.
Key words: : simulation, 6G, OMNeT++, adversarial attacks