A DDoS Attack Traffic Detection System Based on Deep Learning
Gong Chen, Yumeng Zhao, Lisha Shuai, Jintang Luo, Xiaolong Yang · 2021 7th International Conference on Computer and Communications (ICCC) · 2021
Aiming at the detection of DDoS attack traffic, a detection system based on Autoencoder and GRU is proposed in this paper, which can automatically extract data features for deep learning and effectively avoid the dependence on feature engineering. First of all, the method reconstructs the data packets in the network traffic to form a uniform length of data, then uses autoencoder to reduce the data dimension and extract the content features, and finally trains a two-layer stacked GRU network to extract the time sequence characteristics of the network traffic, so as to make use of the characteristic differences between different kinds of traffic to achieve the purpose of identifying DDoS attack traffic. This paper takes the dataset CIC-IDS 2017 as the experimental object, and the results show that this system can effectively deal with the raw network packets, the precision, accuracy and recall rate of the system are more than 99%.