Research on deep learning method of intrusion attack prediction in software defined network
Jing Wang, Hongyan Liu, Fangfang Liu · ITM Web of Conferences · 2022
At present, with the increase of the number of network attacks, in the software defined network, the controller is equivalent to the brain, which is an entity with a complete view of the network. When the attacker directs the malicious traffic to the controller, it may lead to the paralysis of the whole network. Therefore, although there are many solutions for intrusion detection, the attack prediction of network intrusion is still a problem worthy of study. This paper proposes a deep learning model based on gating loop unit (Gru) to identify and prevent intrusion attacks. The model can deeply learn the dependencies of security alarm sequences, and use data sets to evaluate the model. Experiments show that it can show the significant improvement of attack detection.