An Intelligent Mechanism to Detect Cyberattacks of Mirai Botnet in IoT Networks

The evolution of computational resources allowed the innovation and development of new technologies to improve the accessibility and agility of daily tasks. From these new technologies arose the deployment of IoT Networks. However, most of the IoT devices still don’t implement security countermeasures, turn them vulnerable to cyberattacks. One of the most dangerous cyberattacks for IoT networks is the Mirai Botnet, a malware that turns networked consumer devices into a botnet to perform Distributed Denial of Service (DDoS) attacks. Therefore, it is necessary to provide a security solution to detect cyberattacks on IoT networks. A promising approach to improve the IoT network’s detection capacity is the use of Machine Learning (ML) based solutions. This paper presents an intelligent mechanism to detect the Scan, Ack Flooding, Syn Flooding, UDP Flooding, and UDPplain Mirai Botnet attacks on IoT networks using ML techniques, comparing distinct ML approaches (KNN, SVM, and LR). The proposed mechanism was evaluated using a real IoT devices traffic dataset, giving 99% of accuracy to detect Mirai Botnet.