A Fourier Series-Based Anomaly Extraction Approach to Access Network Traffic in Power Telecommunications
Honghao Zhao, Bin Lu, Liang Yu, Siwen Zhao, Lingkang Zeng, Zhe Zhang, Ping You · 2017 International Conference on Computer Systems, Electronics and Control (ICCSEC) · 2017
This paper studies anomaly detection approaches about access network traffic in power telecommunications. Firstly, we use the Fourier series expansion theory to analyze the features in access network traffic for power telecommunications. Access network traffic is regarded as a discrete time series. Then we perform the Fourier series expansion process for access network traffic. Secondly, according to the Fourier series expansion of access network traffic, we construct the Fourier series expansion matrix. Thirdly, we the principal component analysis theory to decompose the matrix. The hidden properties about access network traffic are dug out. Then we propose a detection algorithm to find out the abnormal part in access network traffic in power telecommunications. Simulation results show that our approach is effective.