Network traffic anomalies automatic detection in DDoS attacks

Distributed denial-of-service attacks (DDoS attacks) are intrusions into computing systems of the Internet. Their purpose is to make Internet inaccessible for users. DDoS attack consist sending many requests a certain resource at same time. As result, server cannot withstand network load. In such situation, provider must determine moment when begins and change traffic management strategy. Detection beginning possible by using unsupervised machine learning methods sequential statistical analysis activity. To activate that, convenient use mathematical models based on discrete random processes with monotonically increasing trajectories. Random functions, which represented in correspondence between generalized time cumulative sum or total number incoming packets processed, their type from linear non-linear. first case, parabolic exponential, second case logarithmic arctangent. going change, one can quadratic forms approximation-estimation tests as rules.