Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Load characterization and anomaly detection for voice over IP traffic.

Michel Mandjes1, Iraj Saniee, Alexander L Stolyar

  • 1Center for Mathematics and Computer Science (CWI), 1090 GB Amsterdam, The Netherlands. michel@cwi.nl

IEEE Transactions on Neural Networks
|October 29, 2005
PubMed
Summary

We developed new methods for detecting traffic anomalies in IP networks by analyzing cumulative traffic over time. This approach effectively identifies overloads and failures, improving network management beyond traditional methods.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Parameter estimation in a dynamic Chung-Lu random graph.

Physical review. E·2025
Same author

Interpersonal trust: Asymptotic analysis of a stochastic coordination game with multi-agent learning.

Chaos (Woodbury, N.Y.)·2024
Same author

Statistical inference for a quasi birth-death model of RNA transcription.

BMC bioinformatics·2022
Same author

Approximations for reflected fractional Brownian motion.

Physical review. E·2019
Same author

Queues on a Dynamically Evolving Graph.

Journal of statistical physics·2019
Same author

Rare event simulation for steady-state probabilities via recurrency cycles.

Chaos (Woodbury, N.Y.)·2019

Area of Science:

  • Computer Science
  • Network Engineering
  • Data Analysis

Background:

  • Traffic anomalies in IP networks often stem from focused overload or network element failures.
  • Current IP network management relies on first-order statistics and fixed thresholds, which are insufficient for detecting complex anomalies.

Purpose of the Study:

  • To develop novel analytical methods for traffic anomaly detection in IP networks.
  • To specifically address anomaly detection in voice over IP (VoIP) traffic.
  • To propose simple yet effective anomaly detection tests for over/underload conditions.

Main Methods:

  • Derivation of general formulae for the variance of cumulative traffic over fixed time intervals.
  • Simplification of analytical expressions for voice over IP (VoIP) traffic.

Related Experiment Videos

  • Proposal of anomaly detection tests based on cumulative traffic analysis over extended periods (e.g., 5 minutes).

    • Main Results:

    • Cumulative traffic analysis over longer intervals (e.g., 5 minutes) is sufficient for detecting load anomalies.
    • The derived analytical expressions provide a robust method for identifying traffic anomalies.
    • The proposed approach significantly enhances anomaly detection capabilities compared to existing methods.

    Conclusions:

    • The developed scheme offers a substantial improvement over current IP network management practices.
    • The proposed methods are effective for detecting both overloads and underloads in network traffic.
    • Successful application to field data from an operational network validates the proposed anomaly detection scheme.