A DDoS Detection Algorithm Using Caos Based on Density of Maxima

Abstract

In this paper, a technique is presented to
analyze and detect anomalies in network data ows.
The Density of Maxima is used to measure the chaotic
behavior of network trac in order to detect DDoS
attacks. The experiments were performed using data
containing synthetic trac created with the JMeter
tool, real data from the World Cup 1998, CAIDA
2007 and DARPA 2009 DDoS Malware datasets. All
tests demonstrated the eectiveness of the technique
in identifying when the attacks occurred, as well as in
separating regular from DDoS traffic.