MODELING NETWORK TRAFFIC AND ATTACKS FOR SECURITY TESTING

摘要

In response to the threat posed by malicious code and by potential system and network intruders, the information assurance community continues to work on countervailing techniques and tools. Designers must test their detection algorithms under realistic conditions; however, traffic collected from real networks may contain any number of unknown attacks whose presence may compromise test results. Ideally researchers and developers would be able to introduce specific attacks into an isolated test-bed network that includes realistic attack-free background traffic. We present such a system for generating synthetic network traffic based on models of behavior observed in real networks.