One-Parameter Statistical Methods to Recognize DDoS Attacks

Within our academic high-speed network infrastructure which is used for connecting all universities and high schools in country to the Internet, there are thousands of cybersecurity attacks occurring every day. That why, within SANET II project, an effort has been made create a self-learning system without teacher, would be able quickly adapt arbitrary traffic recognize DDoS on time, even networks, with potential simple implementation into hardware probe. In article, we deal Hurst autoregression coefficients coefficient variation. We test simulated data real records attacks. For early machine recognition attack, propose so-called predicting σ-tunnel. The obtained results can lead investigation other prediction methods that improve attack.