The WV_FN_PAUL function constructs wavelet coefficients for the Paul wavelet function. In real space, the Paul wavelet function is proportional to the complex polynomial (1 – ix / s)^(–m–1), where s is the wavelet scale, m is a non-dimensional parameter, and x is the position.
Result = WV_FN_PAUL( [Order] [, Scale, N] [, /DOUBLE] [, FREQUENCY=variable] [, /SPATIAL] [, WAVELET=variable])
The returned value of this function is an anonymous structure of information about the particular wavelet.
Tag |
Type |
Definition |
FAMILY |
STRING |
‘Paul’ |
ORDER_NAME |
STRING |
‘Parameter’ |
ORDER_RANGE |
DBLARR(3) |
[1, 20, 4] Valid orders [first, last, default] |
ORDER |
DOUBLE |
The chosen Order |
DISCRETE |
INT |
0 [0=continuous, 1=discrete] |
ORTHOGONAL |
INT |
0 [0=nonorthogonal, 1=orthogonal] |
SYMMETRIC |
INT |
1 [0=asymmetric, 1=symm.] |
SUPPORT |
DOUBLE |
Infinity [Compact support width] |
MOMENTS |
INT |
1 [Number of vanishing moments] |
REGULARITY |
DOUBLE |
Infinity [Number of continuous derivatives] |
E_FOLDING |
DOUBLE |
1/sqrt(2) [Autocorrelation e-fold distance] |
FOURIER_PERIOD |
DOUBLE |
Ratio of Fourier wavelength to scale |
A scalar that specifies the non-dimensional order for the wavelet. The default is 4.
A scalar that specifies the scale at which to construct the wavelet function.
An integer that specifies the number of points in the wavelet function. For Fourier space (SPATIAL=0), the frequencies are constructed following the FFT convention:
For real space (/SPATIAL), the spatial coordinates are –(N–1)/2...(N–1)/2.
Note: If none of the above arguments are present then the function will simply return the Result structure using the default Order.
Set this keyword to force the computation to be done in double-precision arithmetic.
Set this keyword to a named variable in which to return the frequency array used to construct the wavelet. This variable will be undefined if SPATIAL is set.
Set this keyword to return the wavelet function in real space. The default is to return the wavelet function in Fourier space.
Set this keyword to a named variable in which to return the wavelet function.
Torrence and Compo, 1998: A Practical Guide to Wavelet Analysis. Bull. Amer. Meteor. Soc., 79, 61–78.
Plot the Paul wavelet function at scale=100:
n = 1000 ; pick a nice number of points
info = WV_FN_PAUL( 6, 100, n, /SPATIAL, $
WAVELET=wavelet)
plot, float(wavelet), THICK=2
oplot, imaginary(wavelet)
Now plot the same wavelet in Fourier space:
info = WV_FN_PAUL( 6, 100, n, $
FREQUENCY=frequency, WAVELET=wave_fourier)
plot, frequency, wave_fourier, $
xrange=[-0.2,0.2], thick=2
5.4 |
Introduced |