Distributed Denial of Service Attacks Detection in Internet of Things Using the Majority Voting Approach

With the ever-increasing number of Internet of Things devices, their security is becoming a very worrying issue. Weak security measures enable attackers to attack IoT devices. One of these attacks is the distributed denial of service(DDOS) attack. Therefore, the existence of intrusion detection systems in the Internet of Things is of special importance. In this research, the majority voting group approach, which is a subset of machine learning, has been used to detect and predict attacks. The motivation for using this method is to achieve better detection accuracy and a very low false positive rate by combining several machine learning classification algorithms in heterogeneous Internet of Things networks. In this research, the new and improved CICDDOS2019 dataset has been used to evaluate the proposed method. The simulation results show that by applying the majority voting Ensemble method on five attacks from this data set, this method respectively has achieved accuracy of detection 99.9668%, 99.9670%, 100%, 99.9686% and 99.9674% in identifying DNS, NETBIOS, LDAP, UDP and SNMP attacks which better and more stable performance in detecting and predicting attacks have achieved than the basic models .