Identifying and tracking suspicious activities through IP gray space analysis

Abstract

Campus or enterprise networks often have many unassigned IP addresses that collectively form IP gray space within the address blocks of such networks. Using one-month traffic data collected in a large campus network, we have monitored a significant amount of unwanted traffic towards IP gray space in various forms, such as worms, port scanning, and denial of service attacks. In this paper, we apply a heuristic algorithm to extract the IP gray space in our campus network. Subsequently, we analyze the behavioral patterns such as dominant activities and target randomness, of the gray space traffic for individual outside hosts. By correlating and contrasting the traffic towards IP gray addresses and live end hosts, we find the gray space traffic provides unique insight for uncovering the behavior, and intention,of anomalous traffic towards live end hosts. Finally, we demonstrate the applications of gray space traffic for identifying SPAM behavior, detecting malicious scanning and worm activities that successfully compromise end hosts.