A Decentralized Approach to Intrusion Detection in Dynamic Networks of the Internet of Things Based on Multiagent Reinforcement Learning with Interagent Interaction
Maxim Kalinin, E. I. Tkacheva · Automatic Control and Computer Sciences · 2023
Abstract The application of multiagent reinforcement learning technology to solve the problem of intrusion detection in the Internet of Things (IoT) systems is considered. Three models of a multiagent intrusion detection system are implemented: a completely decentralized system, a system with the transfer of forecast data, and a system with the transfer of observation data. The experimental results are given in comparison with the Suricata open-code intrusion detection system. The considered architectures of multiagent systems are shown to be free from the shortcomings of the existing solutions.