Design And Implementation Of Real-time Traffic Monitoring System Based On Streaming Computing

As a new generation of information and communication technology,the fifth-generation mobile communication technology(5G)has become a hot topic of great concern from all walks of life around the world.Compared with 4G,5G network provides ten times the information transmission rate of 4G network,millisecond delay and large-scale equipment connection.The huge traffic generated by massive equipment makes the attack and defense situation of the entire network more uneven.Effective use of original traffic data packets to monitor abnormal behavior in the network so as to provide timely and accurate basis for security incident response is an important issue.This thesis designs and implements a real-time traffic monitoring system based on stream computing to solve the problem of real-time network monitoring and abnormal traffic detection under super large network data traffic.The main research contents of this thesis are as follows:(1)Designed and implemented a two-stage abnormal traffic detection model based on the SMOTE algorithm.The model can not only distinguish between normal traffic and abnormal traffic,but also performs fine-grained attack classification on abnormal traffic.The model uses SMOTE algorithm to randomly oversample the abnormal traffic data set.And generate artificially synthesized abnormal flow data to solve the problem of model overfitting caused by the imbalance of abnormal traffic data.The model is simple to construct and easy to understand,and to a certain extent solves the problem of easy overfitting of the decision tree algorithm,and can quickly and accurately identify abnormal network traffic.(2)Combined with the abnormal traffic detection model implemented in this article,a real-time traffic monitoring system based on stream computing is designed and implemented.In the scenario of super large network data traffic,the system can perform real-time network monitoring,network protocol analysis,and abnormal traffic detection,which is a safe operation for the network.Maintenance personnel provide real-time visualized network operation status information,improve the protocol and network operation and maintenance efficiency of network security researchers,and strengthen the ability to perceive abnormal network behaviors.