Anomaly Detection in IP Networks
Description |
Publications |
People |
Collaborators |
Funding
Description
Statistical techniques for detecting anomalous events can be an
invaluable tool for the operators of large IP networks. In this
project, we explore the use of statistical techniques, such as
wavelets, Principal Component Analysis, and Kalman-based filters, in
automatically detecting and diagnosing anomalies. The research draws
on analyzing large volumes of traffic and routing data collected from
many vantage points.
Publications
- Haakon Ringberg, Augustin Soule, Jennifer Rexford, and Christophe Diot,
"Sensitivity of PCA for traffic anomaly detection,"
to appear in Proc. ACM SIGMETRICS, June 2007.
- Augustin Soule, Haakon Ringberg, Fernando Silveira, Jennifer
Rexford, and Christophe Diot,
"Detectability of traffic anomalies in two
adjacent networks,"
in Proc. Passive and Active Measurement Conference, April 2007
(Augustin's slides).
- Jian Zhang, Jennifer Rexford, and Joan Feigenbaum,
"Learning-based anomaly detection in
BGP updates," Proc. ACM SIGCOMM MineNet workshop, August 2005
(Jian's slides).
A longer version is available as
Yale University Technical Report YALEU/DCS/TR-1318,
April 2005.
People
Collaborators
Funding
The research on BGP anomaly detection is part of the project on "Incrementally Deployable Secure Interdomain
Routing" funded by HSARPA. The research on detecting traffic
anomalies is partially funded by Thomson Technology Paris Lab.