Simulation of Rank Correlation Based Detection Mechanism for Distributed Denial of Services Attacks

Abstract

Since the dawn of the Internet, DDoS exhibits a serious threat to the Internet, in which large number of controlled hosts floods the scapegoat or victim site with enormous packets. Furthermore, in Distributed Reflection DoS (DRDoS), invaders bluff or cheat innocent servers into flushing packets to the victim. However, most of current DRDoS detection mechanisms are associated with specific protocols and cannot be used for mysterious or unrecognized protocols. It is learnt that the stimulation by the same attacking flow, the responsive flows from reflectors may have inherent relations: the packet rate of one converged responsive flow may have linear relationships with another. Based on this investigation, the Rank Correlation based Detection (RCD) algorithm is proposed. The primary simulations denote that RCD can differentiate reflection flows from authorized and authenticated ones effectively and efficiently thus, can be utilized as a useable indicator for DRDoS. The paper presents a study of latest botnet attacks and proposes an appropriate guard method for DDoS attacks.