An integrated intelligent paradigm to detect DDoS attack in mobile ad hoc networks

摘要: Various types of routing attacks and their corresponding countermeasures for mobile ad hoc networks (MANETs) have been identified in the literature study. However problems computational complexity false alarms not yet reduced. In this paper, we proposed a proactive detection mechanism distributed denial service (DDoS) which considers feature extraction, reduction entropy, clustering technique ranking. These techniques are approached by statistical analysis involved XOR marking to classify legitimate malicious data packets. Our system applies methodologies on each packet, finds abnormalities during pre-attack phase itself filters them. Experiments done with 2000 DARPA intrusion scenario specific dataset assess time, ratio alarms, complexity. The experimental results show efficiency DDoS attack larger positive