3D Network Traffic Monitoring Based on an Automatic Attack Classifier

In the last years, the exponential growth of computer networks has created an incredibly increase of network data traffic. The management becomes a challenging task, requesting a continuous monitoring of the network to detect and diagnose problems, and to fix problems and to optimize performance. Tools, such as Tcpdump and Snort are commonly used as network sniffer, logging and analysis applied on a dedicated host or network segment. They capture the traffic and analyze it for suspicious usage patterns, such as those that occur normally with port scans or Denial-of-service attacks. These tools are very important for the network management, but they do not take advantage of human cognitive capacity of the learning and pattern recognition. To overcome this limitation, this paper aims to present a visual interactive and multiprojection 3D tool with automatic data classification for attack detection.