Anomaly Analysis Based on Entropy Measurement Network Flow Order
Guoqiang Zhang, Liang Chen, Yongfeng Lin, Jiankuan Wang · 2021
According to the current power Internet of Things security situation, we hope to have a detection method that can quickly, efficiently and accurately identify abnormal traffic and network attacks. This paper proposes an information entropy-based network flow order abnormal detection method. This method makes full use of the distribution characteristics of network flow parameters, and constructs network flow order with information entropy. Through entropy analysis under normal and abnormal conditions, real-time abnormal detection of network data is realized. By simulating four mainstream attacks, most of the network flow order characteristics reflect the impact of the attack on the network. Entropy accurately describes the impact of the attack, so it can be used as an effective means to detect attacks. The method in this paper provides effective ideas for the establishment of network flow order and anomaly detection and analysis.