An Improved Wavelet Analysis Method for Detecting DDoS Attacks

摘要

Wavelet Analysis method is considered as one of the most efficient methods for detecting DDoS attacks. However, during the peak data communication hours with a large amount of data transactions, this method is required to collect too many samples that will greatly increase the computational complexity. Therefore, the real-time response time as well as the accuracy of attack detection becomes very low. To address the above problem, we propose a new DDoS detection method called Modified Wavelet Analysis method which is based on the existing Isomap algorithm and wavelet analysis. In the paper, we present our new model and algorithm for detecting DDoS attacks and demonstrate the reasons of why we enlarge the Hurstȁ9;s value of the self-similarity in our new approach. Finally we present an experimental evaluation to demonstrate that the proposed method is more efficient than the other traditional methods based on wavelet analysis.