A Fusion Network Intrusion Detection Model Based on CNN-LSTM

Abstract

 The explosive growth of network and Internet traffic has brought unprecedented challenges. Existing network intrusion detection systems are unable to cope with unknown network attacks and achieve real-time network response. To solve this problem, this paper proposes a hybrid deep learning model that integrates convolutional neural network (CNN) and improved long short-term memory network (LSTM). Convolutional neural network is used to obtain the spatial features of network intrusion detection data, and improved long short-term memory network is used to obtain the long-term and short- term temporal features of network intrusion detection data, thereby retaining the spatial and temporal dependencies of the data. The proposed hybrid deep learning model is applied to publicly available datasets to test its performance, and satisfactory results are obtained, which verifies that the model can efficiently detect network intrusions.