Information Technology Reference
In-Depth Information
suppress the cross terms to a certain extent, it can not eliminate the cross term
completely. Meanwhile, the PWVD reduces the resolution and disrupts many
mathematical properties in WVD.
Compared with the STFT, WT, WVD and PWVD methods, the HHT which has
adaptive characteristic is based on the signal itself. As shown in Figure 5, the signal x(t)
was decomposed into five IMF components and one residue component by the EMD.
As the EMD decomposes signal from high frequency, the
c
1
,
c
2
and
c
3
represent the
linear frequency modulation, sine frequency modulation and cosine components,
respectively. The
c
4
and
c
5
are trend terms, and
r
5
represents the residue component.
Fig. 4.
Comparison of time-frequency analysises of the signal
x(t)
Fig. 5.
The EMD of
x(t)
Fig. 6.
Hilbert time-frequency spectrum
Search WWH ::
Custom Search