Network traffic control system based on ethically-social intruders detection in multi-agent systems

This article presents methods derived from ethically-social domain, which enables detection of intruders in a multi-agent system. Proposed mechanisms may enable to recognize and block actions of entities undesirable or harmful in the computer system on the basis of behavior observation (actions which they undertake). These mechanisms were applied to the network traffic control system and tested in simulations. The results of tests are presented and discussed.