Statistical measures: Promising features for time series based DDoS attack detection

摘要: The pervasive use of communication technologies increases the demand for high quality and reliable services which guarantees availability a system. However, providing is challenging issue due to existence Distributed Denial Service (DDoS) attacks. In DDoS attacks, an attacker, who masquerade itself as legitimate user, tries increase in volume traffic degrade Quality between hosts server. Although intrusion detection systems are used detect they impotent since packets similar normal ones dispatched by attacker. Therefore, transferring from conventional packet-based analysis methods time series based (flow-based) algorithms would be better promising alternative spot this study, we kurtosis skewness measures investigate performance these parameters distinguishing attack traffic.