A novel method of time-frequency representation and its application to biomedical signal processing

Hilbert spectrum was employed to investigate the time-varying frequency characteristics of the practical heart sound signals. The aim is to explore the role that both empirical mode decomposition and Hilbert transform can be used to play in such heart sound signals. Hilbert transform is applied to each intrinsic mode function to obtain the global time-frequency distribution of the underlying signal with a point of view of instantaneous frequency. The instantaneous frequency distributions of heart sound signals were also compared with the results by using the wavelet transform. Both simulation and experimental results were presented and discussed to demonstrate the power and effectiveness of the proposed new approach.