University of Ottawa - Carleton University
Ottawa-Carleton Institute for Computer Science (OCICS) Presentation
|
March 15, 2013 @ 10:00a.m.
SWIRL: A ScalableWatermark to Detect Correlated Network Flows
|
Speaker: Zhong Xiabei
Location: 240 TB (Tory Building) |
ABSTRACT
Flow watermarks are active traffic analysis techniques that help establish a causal connection between two network flows by content-independent manipulations. Watermarks provide a much more scalable approach for flow correlation than passive traffic analysis. Previous designs of scalable watermarks, however, were subject to multi-flow attacks. They also introduced delays too large to be used in most environments. SWIRL, a Scalable Watermark is Invisible and resilient to packet Losses. SWIRL is the first watermark that is practical to use for large-scale traffic analysis. SWIRL uses a flow-dependent approach to resist multi-flow attacks, marking each flow with a different pattern. SWIRL is robust to packet losses and network jitter, yet it introduces only small delays that are invisible to both benign users and determined adversaries.
|
| Return to Schedule |
|
