DDoS Attacks Detection in ‘Internet of Medical Things’ Using Machine Learning Techniques

IoT technology plays a significant role in advancing the healthcare sector. IoT in healthcare enables real-time patient monitoring, improved treatment, and faster disease diagnosis. However, IoMT consists of various devices with different attributes and computation capabilities, and the data collected and transferred is susceptible. Therefore, it is a very challenging task for one to encapsulate the data and conserve data privacy. As the Internet of Things has grown in popularity, so have cyber-attacks. The ‘MIRAI’ and ‘BASHLITE’ are infamous attacks that took advantage of insecure IoT devices. Each layer in the IoT creates an extensive network for properly functioning interconnected devices. Thus, it is crucial to secure all the layers. As existing research demonstrates, the most affected layer in the IoT is the network layer, and the attack is a DDoS attack. In order to mitigate this type of attack, this paper focuses on mitigating the DDoS attack on IoMT. By utilizing real-time traffic data collected from commercial IoT devices affected by the MIRAI-BASHLITE attack, we proposed a machine learning model to detect the DDoS attack. The experimental results show that the KNN algorithm has an accuracy of 99.97%, the Logistic Regression algorithm has an accuracy of 99.98%, and the SVM linear algorithm has an accuracy of 99.97%. The SVM-RBF algorithm performs with 99.94% accuracy.