Distributed Denial of Service Attack Detection using Deep Learning Approaches

In this paper, the Deep Learning based approaches (Convolutional Neural Network and variants of Recurrent Neural Networks, i.e. Long Short-Term Memory, Bidirectional Long Short-Term Memory, Stacked Long Short-Term Memory and Gated Recurrent Units) have been used to detect Distributed Denial of Service (DDoS) attacks. The Deep Learning approaches have been evaluated using the Portmap.csv file of recent DDoS dataset, i.e. CICDDoS2019. Before giving input to the Deep Learning approaches, the data is pre-processed. The Deep Learning approaches are trained and tested using the pre-processed dataset. The reporting results show that RNN based Stacked-LSTM Deep Learning approach produces the best results in detecting Portmap DDoS attack in comparison to other Deep Learning based algorithms.