Securing ZigBee IoT Network Against HULK Distributed Denial of Service Attack

Abstract

In recent years, cases of Distributed Denial of Service (DDoS) campaigns have been increasingly launched by hackers to exploit different Internet of Things (IoT) installations. Of the different strategies used to launch these attacks, the HTTP Unbearable Load King (HULK) DDoS attack method has been known to have devastating consequences when pulled off, because it is made to evade most firewall rules by its form of execution. The ZigBee network, which has existing security features to guard against cyber-attacks, will require extra measures to augment the AES-128 encryption standard it currently implements. This work investigates the HULK threat against a ZigBee network, with a goal to implement a security method that uses the machine learning algorithms such as Support Vector Machines (SVM), Random Forest (RF), Naive Bayes (NB), and K-Nearest Neighbor (KNN) are tested to identify the best algorithm in detecting anomalies in traffic to fortify the ZigBee network framework.