Simulation procedure the steps for generating

Info icon This preview shows pages 91–93. Sign up to view the full content.

Simulation Procedure. The steps for generating simulated data are: (1) For each edge, estimate historic parameters from the full 30 days of LANL data (see Section 3.4). (2) Fit models for the distribution of the λ γ scores collected on the 30 days of data (see Section 3.4.5). (3) Obtain a p -value threshold from ten days of simulated, non-anomalous data (see Section 3.4.6). (4) Simulate 100 days of minute data on each edge according to the his- toric estimates, except for the set of anomalous edges, where the model parameters are adjusted to introduce an anomaly (see below). Anomalous Shapes Introduced into the Simulation. Three anoma- lous shapes were used in the simulation, on two different areas of the LANL network. The shapes are visualized in Figure 3.5. On each edge in each shape, the model parameters were adjusted from their historic settings to mimic an attacker, while all other edges (approximately 550k edges) in the network are left at their historic settings. In both subgraphs, a red path is highlighted. These paths will form the path anomalies . The purple edges, in addition to the red edges, form a more general attack, the caterpillar anomalies , designed to mimic the attack described in Figure 3.2. Finally, the star anomaly was introduced as the set of outgoing edges from the yellow-circled node in subgraph B. 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 91

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

92 J. Neil, C. Storlie, C. Hash and A. Brugh (a) Subgraph A (b) Subgraph B Fig. 3.5. Star, path, and caterpillar anomalies. Each subgraph has a core path, around which anomalous shapes were introduced. For each core path, the anomaly is plotted, along with the directly connected edges not involved in the anomaly, which are provided for context. Red edges and nodes give the core path, and additional caterpillar nodes and edges are plotted in purple. Every edge in subgraph B is part of the caterpillar for this subgraph. The yellow circle indicates the node at the center of the anomalous out-star. The additional green edges in subgraph A are to give context for the embedded anomalous subgraph within the larger network. One key differ- ence between subgraph A and subgraph B is that in subgraph A, only two additional purple edges where chosen to be anomalous for the caterpillar anomaly, whereas in subgraph B, every outgoing edge of each node in the path was made anomalous. Yet the red path in subgraph A, call it path A, traverses a much more central part of LANL’s network, whereas path B traverses a much less connected area. While these subgraphs do not come close to examining all of the possibilities of traversal in the network, we chose them as exemplars of the interplay between the underlying graph topology and the attack path taken over that topology.
Image of page 92
Image of page 93
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