Thus one may use changepoint detection methods to

Info icon This preview shows pages 61–63. Sign up to view the full content.

Thus, one may use changepoint detection methods to detect hosts that send email blocks of a similar size. (2) Dropped connections. Block lists that refuse connections from sus- pected spammers will be detected in network traces. Keeping track of such events can help detect spammers, and changepoint detection tech- niques can detect a change in dropped connections from a particular IP address. (3) Connection patterns. Spammers typically send very few emails to a particular domain to avoid being detected. Network monitoring, how- ever, which monitors many domains at once, can detect this pattern. Changepoint detection can detect a spammer touching many different domains. 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 61

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

62 A. G. Tartakovsky 0 500 1000 0 100 200 Time, sec Packets/sec Packet Rate 0 500 1000 0 2 4 n, sec (sample) W n CUSUM 0 500 1000 0 5 Shiryaev-Roberts n, sec (sample) R n SPAM MESSAGE SENT SPAMMER DETECTED SPAMMER DETECTED Fig. 2.12. Spam detection. Top – raw data; middle – CUSUM statistic; bottom – SR statistic. We now illustrate rapid spam detection for a particular real-world data set to prove the feasibility of change detection methods. The data set was obtained from a regional ISP. The trace contains email flows to a mail server from a number of hosts. The records are sorted by the source IP address. Our objective is to isolate suspicious hosts and extract the typ- ical behavioral pattern. Examining how the email size changes with time shows that it is very stable, with some occasional bursts. The individual producing such bursts is very likely to be a spammer. Figure 2.12 shows the detection of a real-world spammer using the AbIDS. The email (SMTP) traffic generated by a certain host is under surveillance. Ordinarily, SMTP traffic produced by a user sending legitimate messages is characterized by a relatively steady intensity, i.e., the number of messages sent per unit time remains more or less constant, with no major bursts or drops. However, the behavior changes abruptly once a spam attack begins: the number of sent messages explodes, possibly for a very short period of time. The top-most plot in Figure 2.12 illustrates this typical behavior. The spike in the traffic intensity that appears in the far right of the plot can eas- ily be detected by changepoint detection methods. The behavior of the linear score-based CUSUM and SR procedures (i.e., C 2 = 0 in (2.15)) is plotted in the middle and bottom pictures, respectively. Both statistics behave similarly – an alarm is raised as soon as the traffic intensity blunder caused by the spam attack is encountered. The spammer is detected imme- diately after he/she starts activity. The difference is mainly prior to the
Image of page 62
Image of page 63
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