Classificação de tráfego e dos dispositivos de IoT através do fluxo de rede e inspeção da carga útil dos pacotes

Abstract

Internet of things arises as a computational paradigm that promotes the interconnection of intelligent objects to the Internet and allows interaction, operational efficiency, and communication. With the growing inclusion, in the network, of intelligent objects that have characteristics such as diversity, heterogeneity, mobility, and low computational power, it is essential to develop mechanisms that allow management and control. In addition, it is important to identify whether the assets are functioning properly or have anomalies. Traffic classification techniques are important to assist network analysis and to handle many other key aspects, such as security, management, access control, and resource supply. Traffic classification mechanisms still present difficulties when applied in dynamic environments and without knowledge of services, especially with cryptography. In order to promote the classification of the network devices and traffic, especially IoT, a technique is presented that uses the random forest (random forest), an automated learning algorithm supervised, together with the inspection of the contents of the packages To this end. Additionally, the same algorithm is used to perform the classification of network traffic through the characteristics extracted from the network flow. At the end of this dissertation, the proposed strategy will be presented in IoT scenarios and the results.