An Intrusion Detection System Based on Convolutional Neural Network

Abstract

In order to achieve rapid response, network intrusion detection system needs the ability to quickly learn new network behavior characteristics. At present, there are many machine learning methods applied in the field of intrusion detection. However, with the increase of the amount of data in the network and the diversification of network intrusion methods, the traditional intrusion detection system has been unable to meet the current needs of the complex network environment. But with the development of Alphago, FaceID and auto-driving technology, we have seen the successful application of convolutional neural network in current complex network environment. Therefore, this paper proposes a network intrusion detection system based on the convolutional neural network model Lenet-5. In addition, in order to improve the stability, increase the accuracy of the model and speed up the converge rate of the model, we also introduce the OHE coding and normalization method to deal with the feature matrix, making it easier for the model to detect the characteristics of the traffic data packet.