A DDoS Detection Algorithm Using Caos Based on Density of Maxima

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