Research Article
Network Security Situation Prediction Based on Improved WGAN
183 downloads
@INPROCEEDINGS{10.1007/978-3-030-32216-8_64, author={Jiang Zhu and Tingting Wang}, title={Network Security Situation Prediction Based on Improved WGAN}, proceedings={Simulation Tools and Techniques. 11th International Conference, SIMUtools 2019, Chengdu, China, July 8--10, 2019, Proceedings}, proceedings_a={SIMUTOOLS}, year={2019}, month={10}, keywords={Situational awareness Situation prediction Generative adversarial network Difference Wasserstein-GAN}, doi={10.1007/978-3-030-32216-8_64} }- Jiang Zhu
Tingting Wang
Year: 2019
Network Security Situation Prediction Based on Improved WGAN
SIMUTOOLS
Springer
DOI: 10.1007/978-3-030-32216-8_64
Abstract
The current network attacks on the network have become very complex. As the highest level of network security situational awareness, situation prediction provides effective information for network administrators to develop security protection strategies. The generative adversarial network (GAN) is a popular generation model, which is difficult to train, collapse mode and gradient instability in this network. A Wasserstein distance as a loss function of GAN is proposed. And a difference term is added on the loss function. The improved Wasserstein-GAN (IWGAN) is to improve the classification precision of the situation value. Compared with other forecasting methods, the results show that the method has obvious advantages.
Copyright © 2019–2024 ICST
