We extract statistical features and build models to identify anomalous bgp traffic based on several typical anomaly event data sets. This package includes multi-scale LSTM, attention based multi-scale LSTM, hierarchy attention based multi-scale LSTM as well as baselines (1-layer LSTM, 2-Layer LSTM, RNN and traditional machine learning methods).
numpy, PyWavelets, scikit-leran, tensorflow, matplotlib, Keras (Tensorflow backend).
The original BGP data set is from RIPE Network Coordination Center: RIPE RIS raw data If you find either the codes or the results are helpful to your work, please kindly cite our paper
Cheng, M., Xu, Q., Lv, J., Liu, W., Li, Q., & Wang, J. (2016, November). MS-LSTM: A multi-scale LSTM model for BGP anomaly detection. In Network Protocols (ICNP), 2016 IEEE 24th International Conference on (pp. 1-6).
This project is licensed under the MIT License.
If had any problem, please send me email: mc.cheng@my.cityu.edu.hk