Man-in-the-Middle Intrusion Detection Based on CNN-LSTM Model

Abstract

With the development of the network, the means of network attacks emerge one after another. Man-in-the-middle(MITM) attack is a kind of high threat, difficult to prevent and very common network attack. Aiming at the man-in-the-middle intrusion detection problem in the network, especially in the Internet of Things(lot), this article proposes a man-in-the-middle attack intrusion detection model based on the combination of Convolution Neural Networks(CNN) and Long Short-Term Memory networks(LSTM). The model sets two Convolution layers and two LSTM layers. Firstly, convolutional neural network is used for feature analysis, then the LSTM is used to classify learning. The data set selects "BoTNeTIoT-L01". By model training, the accuracy of prediction results can reach 99.50%. Finally, the confusion matrix and AUC area are used to evaluate the model. The evaluation results show that CNN-LSTM has a very good predictive effect in man-in-the-middle attack intrusion detection.

Tasks

Reproductions