A Multi-Agent System for Detecting Attacks on Connected Objects
Guyot Tifaine, Florent Carlier, Valérie Renault, Leroux Pascal · 2019
Internet of Things (IoT) is more and more present in our daily lives. The information is often of a personal and sensitive nature. The IoT's security is often too low or non-existent in response to the threats to which it is exposed. It is therefore critical to improve this security. We present a model of embedded security agents that allows to secure connected devices without increasing their resource consumption excessively, while remaining as adaptable as possible. Through a situational analysis, we show the possibility of detecting anomalies in the functioning of objects by ensuring that they respect "proper functioning" patterns.