Denial of Service Attack Detection Based on Feature Extraction and Supervised Techniques

Internet of Things systems (IoT) is expanding exponentially, providing expanded services in different environments. The wide variety of these systems makes security an increasingly important challenge, several Malware, such as Mirai or Dark Nexus, demonstrate an increase in attacks based on IoT. One of the most used protocols in the application layer is the Message Queuing Telemetry Transport (MQTT), these systems can be attacked by Denial of Service attacks. This paper presents a framework for detecting MQTT protocol attacks based on automatic learning, using a dataset formed by all the network traffic generated in an environment that uses an IoT system with the MQTT protocol on which several DoS attacks are performed.