Feature extraction and classification of underwater signals based on Hilbert-Huang transform

Noise radiated from underwater targets is a typical nonlinear and non-stationary signal.Hilbert-Huang Transform is a good method of analyzing this kind of signals com- pared to other traditional methods.Based on this method,a new feature extraction method is proposed in this paper.The targets are divided into four types:cargo ship, fishing boat,marine life and sea noise,and their characteristics are described.Experi- ments on actual noises of underwater targets are carried out,and the results illustrate the method is effective.