Detection of Distributed Denial of Service Prevention (DDoSP): An Overview

Abstract

The present study was carried out to describe the browsing habits of web Searchers and attacks can be prevented. There are many solution based methods created against Distributed Denial Of Service (DDoS) attacks are focused on the Transmission Control Protocol and Internet Protocol layers as a substitute of the high layer. An extended hidden semi-Markov model is proposed to describe the browsing habits of web searchers. A forward algorithm is derived for the online implementation of the model based on the M-algorithm in order to reduce the computational amount introduced by the model’s large state space. Entropy of the user’s HTTP request sequence accurate to the replica is used as a principle to measure the user’s normality. Finally, experiments are conducted to validate our model and algorithm. A new on-line algorithm based on the M-algorithm was designed for the anomaly detection. A set of real traffic data collected from an educational website and a generated App-DDoS attack traffic were used to validate our model.