A Method of Malicious Data Flow Detection Based on Convolutional Neural Network
Binghao Jia, Yanhui Guo, Hongmei Li, Chunlai Du · 2021
Network intrusion detection can protect the devices and data in a network from the threat of network attacks. Nowadays, network intrusion detection research based on traffic data mostly uses the statistical features of traffic data to design algorithms, which has problems such as heavy dependence on expert experience and low accuracy. This paper has proposed a CNN model using the characteristics of traffic packet contents to detect network intrusions. The CNN model is used to extract content features from the traffic packets and then identify the malicious payloads in the packets. The performance of this method is evaluated by the experiments of 9 common categories of malicious payloads and normal traffic. and the results show that the model has the ability to accurately detect network intrusions with high performance.