Design of Monitoring Algorithm for Abnormal Network Traffic Sequence Based on Wavelet Transform

With the expansion of the scale and application of the Internet, the network data flow presents the characteristics of complexity and multifractality, and at the same time, the network security problems caused by malicious behaviors such as network attacks or network intrusions are becoming increasingly prominent. High-quality feature extraction and anomaly identification of a large number of network traffic data is an important basis for network forensics. In order to effectively locate, identify and extract the transient abnormal characteristics of network traffic sequence, this article proposes an abnormal network traffic sequence monitoring algorithm based on wavelet transform according to the disturbance and transient characteristics of network abnormal traffic characteristics. This method solves some key problems such as wavelet selection, modulus maximum curve attenuation judgment and so on. Experiments show that the accuracy of the proposed algorithm is high, reaching 96.17%, which is about 9.8% higher than that of the multi-scale feature detection method. This method can better find the abnormal events of network traffic and locate the abnormal time, and can provide useful information for subsequent related work. It has important practical value in network management and optimization.