Network Based Intrusion Detection using Convolutional Neural Network

Abstract

The diversification of wireless network traffic attack characteristics has led to the problems what traditional intrusion detection technology with high false positive rate, low detection efficiency, and poor generalization ability. In order to enhance the security and improve the detection ability of malicious intrusion behaviour in a wireless network, this paper proposes a wireless network intrusion detection method based on convolutional neural network (CNN). First, the network traffic data is characterized and pre-processed, then modelled the network intrusion traffic data by CNN. The low-level intrusion traffic data is abstractly represented as advanced features by CNN, which extracted autonomously the sample features, and optimizing network parameters to converge the model. Finally, we conducted a sample test to detect the intrusion behaviour of the network. during the training process, and the key feature information loss and parameter tuning difficulty are easily caused during the training process. This paper considers using the end-to end semi-supervised network training classifier of convolutional neural network (CNN), and the multi-layer feature of CNN to detect network, learn the sample features and discover the rules in the data training process to simplify the implementation process. Key Words: intrusion detection, anomaly detection, deep learning, convolution neural network, network security.