Abstract
This paper proposes a new intrusion detection system (IDS) for Software Defined 5G Networking using machine learning techniques to tackle the increasing cybersecurity threats in future 5G networks. The authors proposed and designed dual IDS architectures; one was a combined model of autoencoder and random forest classifier. By testing and evaluation on the benchmark datasets, the proposed models achieved detection rate above 99%. In addition, we implemented our IDS into a real SDN-based network and developed an iOS application that would allow authorized users to view the network’s current security situation easily. This work is important because it offers a large-scale and efficient security solution for 5G that is currently threatened by more vulnerabilities.