This is a fairly difficult data set to come by since

Info icon This preview shows pages 72–74. Sign up to view the full content.

active edges. This is a fairly difficult data set to come by, since collection is a chal- lenge for many reasons. At many universities, for example, data from per- sonal machines is not collected for privacy reasons. In addition, sensors in the network that measure this data need to be distributed well to have access to all of the connections, and when they are, the data rate can be very high, so that significant engineering is required to collect and store these measurements. For these reasons, data of this type is not widely available, and there is little published work using it. The LANL has invested sig- nificant resources in collection over the past ten years, and since attacks can and are observed via edge data, we feel it is an extremely fruitful data stream for researchers to collect and analyze. A plot of one of these edges is given in Figure 3.1. There is an enormous variety among the edges in a typical network. Those that are driven by human presence, such as the edge created between a desktop and an email server when a user checks his email, tend to be similar to Figure 3.1. Others, such as edges between load-balancing servers, tend to be much smoother. Several of the edges exhibit periodic behavior, at multiple timescales. Copyright © 2014. Imperial College Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 2/16/2016 3:37 AM via CGC-GROUP OF COLLEGES (GHARUAN) AN: 779681 ; Heard, Nicholas, Adams, Niall M..; Data Analysis for Network Cyber-security Account: ns224671
Image of page 72

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Statistical Detection of Intruders Within Computer Networks 73 Fig. 3.1. Counts of connections per minute between two machines in LANL’s internal network. The connections originate at a specific user’s machine, and the destination is a server providing virtual desktop services. In order to establish a baseline of behavior for each edge, modeling is used. Section 3.4 presents three models, one of which, the hidden Markov model, attempts to capture this human behavior, by explicitly accounting for the burstiness apparent on this edge. A more thorough modeling effort is required to accurately reflect the variety of edges seen in this data, an effort that is ongoing, but the three models presented represent a first step. Attacks create deviations not just on single edges, but across multiple connected edges, a subject discussed in the next section. Under an inde- pendence assumption among the edges in the subgraph, models for edges lead to models for subgraphs as discussed in Section 3.3. 3.1.2. Example traversal attack A common initial stage of attack on computer networks is to infect a machine on the network using malicious software. One method for initial infection is known as a phishing attack, where an email that includes a link to a malicious website is sent to a set of users on a network. When the user clicks on the link, their machine becomes infected, giving the attacker some form of access to the machine.
Image of page 73
Image of page 74
This is the end of the preview. Sign up to access the rest of the document.
  • Spring '12
  • Kushal Kanwar
  • Graph Theory, Statistical hypothesis testing, Imperial College Press, applicable copyright law

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern