Network intrusion detection method based on one-dimensional CNN and GWO-SVM
Chen Chen, Yajiang Qi, Lintao Yang, Guanghua Wang, Xiaoyan Ye, Dan Wei · 2022
Most networks mainly use firewall and other devices to isolate from external networks. However, with the application of new technologies such as cloud computing and Internet of things, the degree of interconnection between networks is deepening, and the difficulty of security protection is greatly improved. How to effectively detect network intrusion has become very important. Compared with traditional intrusion detection technology, convolutional neural network has better ability to extract intrusion features. This paper proposes a network intrusion detection method based on one-dimensional convolutional neural network and grey wolf optimization algorithm to optimize support vector machine. Firstly, one-dimensional convolutional neural network is used to extract high-level features from intrusion detection data, and then support vector machine is used to classify and detect the extracted high-level features, in which the parameters of support vector are optimized by grey wolf optimization algorithm. Through simulation experiments, the proposed method can effectively improve the detection accuracy and model balance.