Comparison and Application of Time-Frequency Analysis Methods for Nonstationary Signal Processing - Advanced Research on Computer Education, Simulation and Modeling - page 291

The Hilbert time-frequency spectrum of x(t) shown in Figure 6 can be obtained by
performing Hilbert transform to the IMF components, following equation (9).
Compared with the time-frequency maps in Figure 4, the Hilbert time-frequency
spectrum with good time-frequency resolution can reflect all frequency components of
the signal x(t) and give a clear rule of these components changing with time axis. Thus,
in the view of applications, the HHT may be a better method for time-frequency
analysis of nonstationary signals.
4 Conclusions
In this paper, five time-frequency analysis methods, i.e., the STFT, WT, WVD, PWVD
and HHT, were compared and discussed in nonstationary signal processing. The
advantages and shortcomings of each method were investigated. Among the five
methods in discussion, the HHT which has adaptive characteristic can exactly express
the local information of nonstationary signals in a high time-frequency resolution. It
overcomes the irreconcilable contradiction between the time-frequency aggregation
and cross term. A conclusion can be drawn that, since the HHT can exactly reflect
instantaneous frequency components, it is quite suitable for nonstationary signal
analysis in engineering.
Acknowledgments
This work was supported by the NSFC (grant no. 51045007), the Program for Professor
of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning,
and partly supported by graduate student scientific innovation project of SUES (grant
no. A-0530-10-15).
References
1. Zhang, X.: The analysis and processing of nonlinear signals. National Defence Industry
Press, Beijing (1998)
2. Hu, G.: Modern signal processing course. Tsinghua University Press, Beijing (2004)
3. Mallat, S.: A Theory for Multi-Resolution Signal Decomposition Wavelet Representation.
IEEE Trans. on Pattern Analysis and Machine Intell. 11(7), 674-693 (1989)
4. Huang, N.E., Shen, Z.: The empirical mode decomposition and the hilbert spectrum for
non-linear non-stationary time series analysis. Proc. R. Soc., London (1998)
5. Huang, N.E.: A new view of nonlinear waves: the hilbert spectrum. Annual Review of Fluid
Mechanics 31(5), 417-457 (1999)