Evaluación del riesgo de una red de IoT (Internet de las Cosas) en una casa inteligente utilizando redes bayesianas.

Abstract

In the present work, a risk assessment model of an Internet of Things (IoT) network is implemented in a smart home using a Bayesian network. The directed acyclic graph of the Bayesian network is obtained from an attack graph that details the structure of the IoT network and the paths through which different attacks can occur. The parameters of the Bayesian network are estimated with the maximum likelihood method applied to a data set obtained from the simulation of network attacks, in five different scenarios. The risk assessment focuses on the analysis of DoS attacks, MitM attacks and both simultaneously to the devices that allow the automation of the smart home and that are generally the ones that individually have lower levels of security. For this, inferences in the Bayesian network and the impact of the attacks using the simulated data are considered.