Using Hilbert-Huang Transform to Characterize Intrusions in Computer Networks

As there is a rapid rampancy of malicious intrusion, people pay more attentions to the network security. This paper presents a new method based on Hilbert-Huang trans- form (HHT) to characterize intrusions that can detect many kinds of intrusion according to the number of packets. This paper demonstrates the usefulness of such a method to differentiate abnormal data from normal data. Experiments are performed using intrusion detection data sets and tested for validation. Some results are reported along with analysis and concluding remarks.