An empirical study for the traffic flow rate prediction-based anomaly detection in software-defined networking: a challenging overview

Currently, there is an enormous disturbance regarding privacy in information and communication technology around the scientific community. Since any assault or abnormality in the network can seriously disturb numerous realms like national security, private data storage, social welfare, economic issues, and so on. Consequently, one of the domains for detecting intrusion in the network is anomaly detection domain and it is a wide probe area. Various numerous methods and approaches have developed for anomaly detection. In the network security field, traffic anomaly detection has been a main aspect. The network security domain recognizes assaults in terms of significant deviations from the entrenched regular usage profiles. Nowadays, software-defined networking (SDN) is a new networking model has developed to ease effectual network control and management. This view investigates 50 probe papers focused on traffic flow rate prediction-based anomaly detection in SDN. Furthermore, it presents technique wise classifications like flow counting-based techniques, information theory-based approaches, entropy-based techniques, deep learning (DL)-based approaches, hybrid methods and network methods. An examination includes in an overview based on classification research techniques, toolset used, years of publication, datasets, and evaluation metrics for predicting anomaly in the SDN environment. Lastly, the limitations of surveyed techniques are explained, that encourage investigators for inventing more new techniques for predicting anomaly in SDN.