Implementation of Intelligent Agents for Network Traffic and Security Risk Analysis in Cyber-Physical Systems
Igor Vitalievich Kotenko, Sergey Aleksandrovich Ageev, Igor Borisovich Saenko · 2018
The paper offers an approach for implementation of intelligent agents intended for network traffic and security risk analysis in cyber-physical systems. The agents are based on the algorithm of pseudo-gradient adaptive anomaly detection and fuzzy logical inference. The suggested algorithm operates in real time. The fuzzy logical inference is used for regulation of algorithm parameters. The variants of the implementation are proposed. The experimental assessment of the approach confirms its high speed and adequate accuracy for network traffic analysis.