5G slice network anomaly detection method based on virtual network flow analysis

The invention relates to a 5G slice network anomaly detection method based on virtual network flow analysis, and the method comprises the steps: designing and obtaining a normal cumulative data volume fluctuation range set corresponding to each collection node, and carrying out the anomaly analysis of each data flow passing through each collection node through each collection node according to the normal cumulative data volume fluctuation range set; end-to-end security detection of the 5G slicing service is realized, and parallelization of a network anomaly detection algorithm is realized; on the basis of OD shunting of the data traffic by each acquisition node, the central server counts the transmission of each data traffic in the network based on a network topology structure, and performs abnormal analysis on the quantity traffic passing through the central server by each node; according to the whole scheme design, localization of mass data storage and calculation is achieved, and a high-performance extensi