A New DDoS Attack Detection Algorithm
Pin Zhou · Communications technology · 2014
In order to detect the DDoS attacks accurately and timely,a new DDoS attack detection algorithm is proposed. The algorithm,based on traditional wavelet analysis method,integrates the DDoS detection method of principal component analysis and wavelet analysis,and the corresponding model and algorithm are built up to detect the DDoS attack based on this algorithm. The detection method could be applied to design a model and an algorithm,and the information entropy of information theory is also used to measure the dispersion degree of the source IP address. The proposed algorithm could set the threshold self-adaptively according to the initial-stage variation of Husrt index and the entropy,thus to detect the occurrence of attacks. Experimental result shows that this method could significantly improve the detection rate of DDoS attacks.