Active Defense Oriented Malicious Behavior Detection And Visualization Research

With the continuous development of computer and network technology,the scale of networks is constantly expanding,and the data generated by network communication is growing exponentially.The network security situation is becoming increasingly severe,and the issue of network defense has received widespread attention.Traditional passive defense strategies are increasingly limited,and honeypot technology based on active defense can achieve flexible trapping of network attacks,capturing multidimensional correlations and temporal correlations of attack behavior information and its traffic data.In recent years,the rapidly developing deep learning framework in the field of image has provided new ideas for the development of network security.This article constructs an intelligent detection model for network traffic based on temporal correlations,which maps traffic behavior data based on time series into a visualization expression of traffic parallel coordinates,and applies the correlation information of traffic time context to intelligent classification of traffic.At the same time,using active defense honeypot technology and combining it with the VGG-16 deep learning model,a flow classification visualization intelligent detection system for active defense is constructed.It realizes the trapping,classification and attribution discrimination of network attacks.Experimental results show that the flow classification visualization intelligent detection system for active defense can effectively detect and classify network malicious traffic,and visualize the statistical information of the classification results.The accuracy of the proposed visualization deep learning intelligent detection model based on parallel coordinates can reach 97.2%,and ultimately achieve intuitive expression and effective identification of malicious traffic information.