Flow and rank correlation based detection against Distributed Reflection Denial of Service attack

Denial of service attack in computing involves blocking of resources to particular user. This kind of blocking may be temporary or indefinite blocking to access web server in the network. This attack is usually done by a victim web server. Distributed denial of service attacks is given by more than two web servers. The flood of incoming network flows to the victim web server forces it to shut down. Although the attack's individual internet packets may be completely not dangerous and flood of such type of packets can overwhelm the victim server's internet connection or other packet-processing resources in the network. The result is that valid traffic, unable to compete with malicious flood, has little chance for obtaining useful service to legitimate users. This paper mainly concentrates on flow and rank correlation based algorithm (FRCD). With rank correlation we can check if there is any attack in the network or not. And having this as the base we could calculate flow correlation to check if this attack is legitimate or not. The proposed system is highly efficient and is a protocol independent method. The computational cost involved does not affect the network's throughput thereby increasing the possibilities of Distributed Reflection Denial of Service attack detection.