Sunteți pe pagina 1din 7

GEOPHYSICS, VOL. 80, NO. 2 (MARCH-APRIL 2015); P. A31–A37, 3 FIGS.

10.1190/GEO2014-0441.1
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

Frequencies of the Ricker wavelet

Yanghua Wang1

derivatives of a Gaussian function have similar amplitude spectra,


ABSTRACT altered from a Gaussian distribution and thus the Ricker wavelet is
often used in seismic analysis, for example, wavefield simulation,
The Ricker wavelet is theoretically a solution of the reflectivity inversion, attenuation evaluation, etc.
Stokes differential equation, which takes into account the In the frequency domain, the amplitude spectrum of the Ricker
effect of Newtonian viscosity, and is applicable to seismic wavelet and the spectra of various derivatives of a Gaussian are
waves propagated through viscoelastic homogeneous media. asymmetric. Physically, asymmetry can represent the frequency-
In this paper, we defined the time-domain breadth and the dependent attenuation behavior of seismic waves in viscoelastic me-
frequency-domain bandwidth of the Ricker wavelet and de- dia. In this paper, we attempt to analyze the character of the Ricker
veloped quantities analytically in terms of the Lambert W wavelet, especially the key parameters describing the frequency
function. We determined that the central frequency, the geo- spectrum. We first define the time-domain wavelet breadth and
metric center of the frequency band, is close to the mean fre- the frequency-domain bandwidth of the Ricker wavelet. These
quency statistically evaluated using the power spectrum, two quantities are the time and frequency durations, defined at a
rather than the amplitude spectrum used in some of the pub- half of the waveform peak and the amplitude spectrum peak, respec-
lished literature. We also proved that the standard deviation tively. The definitions of time breadth and frequency bandwidth
from the mean frequency is not, as suggested by the literature, lead to an inverse exponential equation. The solution of this equa-
the half-bandwidth of the frequency spectrum of the Ricker tion is a special function known as the Lambert W function (Lam-
wavelet. Moreover, we established mathematically the rela-
bert, 1758, 1772; Euler, 1779; Corless et al., 1996). By using the
tionships between the theoretical frequencies (the central
Lambert W function, we are able to analytically present not only the
frequency and the half-bandwidth) and the numerical mea-
frequency bandwidth (Wang, 2015) but also the time-domain
surements (the mean frequency and its standard deviation)
breadth of the Ricker wavelet.
and produced each of these frequency quantities analytically
The geometric center of the frequency band is referred to as the
in terms of the peak frequency of the Ricker wavelet.
central frequency. Once we analytically derive the central fre-
quency of the Ricker wavelet, we are able to find the relationship
between the central frequency and the mean frequency. The latter
INTRODUCTION is a statistical quantity that can be evaluated practically from the
discrete Fourier spectrum of the seismic data. In the published
The Ricker wavelet is a theoretical waveform obtained by solving literature, some works implement the evaluation of the mean fre-
the Stokes differential equation (Ricker, 1943, 1944). Because this quency using the power spectrum (Cohen and Lee, 1989, 1990;
equation takes into account the effect of Newtonian viscosity, the Barnes, 1993; Loughlin and Tacer, 1997; Loughlin and Davidson,
Ricker wavelet is representative of seismic waves propagating 2001; Carter and Kendall, 2006; Wang, 2014), and other works use
through viscoelastic homogeneous media, i.e., the Voigt model. the amplitude spectrum (Quan and Harris, 1997; Hu et al., 2013). We
Mathematically, the Ricker wavelet is the second derivative of a derive expressions analytically for both statistical cases and prove
Gaussian function; therefore, it is symmetric in the time domain. that the central frequency is close to the mean frequency evaluated
However, in reality, seismic signals are often not symmetric and from the power spectrum rather than the amplitude spectrum.
instead are close to the first or one-and-a-half derivatives of a Gaus- The literature also suggests that the standard deviation from
sian function (Ricker, 1953; Hosken, 1988). Nevertheless, various the mean frequency could be considered as the half-bandwidth

Manuscript received by the Editor 20 September 2014; revised manuscript received 11 November 2014; published online 2 February 2015.
1
Imperial College London, Department of Earth Science and Engineering, Centre for Reservoir Geophysics, London, UK. E-mail: yanghua.wang@imperial
.ac.uk.
© 2015 Society of Exploration Geophysicists. All rights reserved.

A31
A32 Wang
    pffiffiffi
(Berkhout, 1984; Cohen and Lee, 1989, 1990; Gram-Hansen, 1991; 1 1 2 2 1 1 e
Barnes, 1993). In this paper, however, we prove that the standard − ωp τ exp − ω2p τ2 ¼ : (2)
2 4 2 4 4
deviation is not the half-bandwidth of the Ricker wavelet. We derive
the relationship between the frequency bandwidth, which is a theo- This is the inverse exponential equation,
retical parameter, and the standard deviation, which is a statistical
quantity evaluated from either the power spectrum or the amplitude z exp z ¼ x;
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

(3)
spectrum of the Ricker wavelet.
In geophysical analysis, frequency quantities such as the central fre- pffiffiffi
quency and the bandwidth can be used to describe seismic resolution, where x ¼ 14 e and z ¼ 12 − 14 ω2p τ2 . The solution of this special
and changes in these quantities can be exploited to measure the attenu- equation is
ation coefficients of subsurface media. However, field seismic signa-
z ¼ WðxÞ; (4)
tures might differ from the Ricker wavelet, the second derivative of a
Gaussian. If they are close to fractional derivatives of a Gaussian, their
where WðxÞ is the Lambert W function, displayed in Figure 1
spectra could be similar to the Ricker amplitude spectrum. In this pa-
(Wang, 2015).
per, we will see that the power spectrum, rather than the amplitude
By solving equation 2, we may obtain the half-breadth of the
spectrum, of the Ricker wavelet is close to a Gaussian distribution. Ricker wavelet, as
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffi
THE RICKER WAVELET 2 1 e
τb ¼ −W : (5)
ωp 2 4
The Ricker wavelet is defined in the time domain as
    According to the Lambert W function in Figure 1, when
1 1 pffiffiffi pffiffiffi
rðτÞ ¼ 1 − ω2p τ2 exp − ω2p τ2 ; (1) x ¼ 14 e ≈ 0.41218, Wð14 eÞ ≈ 0.304101. Therefore, the half-
2 4 breadth of the Ricker wavelet is

where τ is time (in seconds) and ωp is the most energetic frequency 0.88521
τb ≈ : (6)
(in radians per second). It is symmetric in the time domain and has a ωp
zero mean, as ∫ ∞ −∞ rðτÞdτ ¼ 0.
The Ricker wavelet rðτÞ is normalized to unity. The breadth of For a sample Ricker wavelet shown in Figure 2a, defined arbi-
this wavelet is defined by the time duration at one-half of the wave trarily to have its most energetic angular frequency at 60π rad∕s
peak, rðτÞ ¼ 12, which leads to the equation (equivalent to the ordinary frequency of 30 Hz), the full-time
breadth is 2τb ≈ 9.4 ms. This time quantity of the arbitrarily chosen
30-Hz wavelet might provide us with an insight on the time-domain
seismic resolution limit.

THE FREQUENCY BAND AND THE CENTRAL


FREQUENCY
The Fourier transform of the Ricker wavelet may be expressed
as
 
2ω2 ω2
RðωÞ ¼ pffiffiffi 3 exp − 2 ; (7)
π ωp ωp

where ω is the angular frequency. Because the frequency spectrum


is real and nonnegative in value, it is just the amplitude spectrum
jRðωÞj ¼ RðωÞ of a Ricker wavelet with any possible time delay in
the time domain.
To verify that the most energetic frequency ωp is the peak fre-
quency corresponding to the maximum amplitude, we can set the
derivative of the amplitude spectrum, with respect to the frequency,
to zero:
   
∂R 4ω ω2 ω2
Figure 1. Lambert W function. The minimal of the real x variable is ¼ pffiffiffi 3 1 − 2 exp − 2 ¼ 0. (8)
∂ω π ωp ωp ωp
−e−1 , where Wð−e−1 Þ ¼ −1. Separated from this point denoted by
a cross, the Lambert W function has two branches, W −1 ðxÞ for
WðxÞ ≤ −1 and W 0 ðxÞ for WðxÞ ≥ −1. This leads to the peak frequency ω ¼ ωp.
Frequencies of the Ricker wavelet A33

 
As is known, the Ricker wavelet is the second derivative of a 1
Gaussian function. Although the amplitude spectrum of a Gaussian W0 − ≈ −0.231961;
2e
is still a Gaussian distribution, the amplitude spectrum of the Ricker  
wavelet is the Gaussian distribution multiplied by a factor ω2 and 1
W −1 − ≈ −2.67835. (15)
thus is asymmetric in the frequency domain. Hence, the peak fre- 2e
quency is different from the central frequency, the geometric center
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

of the frequency band. As noted for the first time by Wang (2015), Hence, the frequency quantities are
the definition of the frequency band and the central frequency of a
Ricker wavelet can be expressed in terms of the Lambert W func- ðωl1 ; ωl2 Þ ≈ ð0.481623ωp ; 1.636567ωp Þ;
tion. These results are summarized here for the sake of com- ωc ≈ 1.059095ωp ;
pleteness.
For the amplitude spectrum of the Ricker wavelet, the peak is ωb ≈ 0.577472ωp : (16)
pffiffiffi
Rðωp Þ ¼ 2ðe π ωp Þ−1 . The frequency band is measured at a half
of this maximum as For the sample Ricker wavelet defined with the peak angular
frequency of ωp ≈ 188.5 rad∕s, the full frequency band is 2ωb ≈
1 217.7 rad∕s (Figure 2b).
RðωÞ ¼ pffiffiffi ; (9)
e π ωp
FREQUENCY QUANTITIES EVALUATED USING
which leads to the inverse exponential equation THE POWER SPECTRUM
   
ω2 ω2 1 For a practical application when we have discrete frequency spec-
− 2 exp − 2 ¼ − ; (10) trum of seismic data, we can statistically evaluate the mean fre-
ωp ωp 2e
quency rather than the central frequency. The formulas for
evaluating the mean frequency and the standard deviation, based
with a solution expressed in terms of the Lambert W function: on the power spectrum, are
  R∞
ω2 1 ωR2 ðωÞdω
− 2 ¼W − : (11) ωm ¼ R0 ∞ 2 ;
ωp 2e 0 R ðωÞdω
R ∞ 
ðω − ωm Þ2 R2 ðωÞdω 1∕2
The Lambert W function has two branches W −1 ðxÞ for WðxÞ ≤ −1 ωσ ¼ 0 R ∞ 2 ; (17)
0 R ðωÞdω
and W 0 ðxÞ for WðxÞ ≥ −1, separated at point (−e−1 , −1), denoted
by a cross in Figure 1. Then, the frequency band [ωl1 , ωl2 ] is
given by
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
1
ωl1 ¼ ωp −W 0 − ;
2e
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
1
ωl2 ¼ ωp −W −1 − : (12)
2e

The analytical expression of the central frequency is

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
ωp 1 1
ωc ¼ −W 0 − þ −W −1 − ; (13)
2 2e 2e

and the half-bandwidth is

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
ωp 1 1
ωb ¼ −W −1 − − −W 0 − : (14)
2 2e 2e
Figure 2. (a) Ricker wavelet rðτÞ, where τ is time. The breadth of
the wavelet is 2τb . (b) Frequency spectrum RðωÞ, where ω is the
Numerical approximations of W 0 ðxÞ and W −1 ðxÞ for x ¼ angular frequency of wavelet rðτÞ. The central frequency is ωc ,
−ð2eÞ−1 are and the bandwidth is 2ωb .
A34 Wang

where R2 ðωÞ is the power spectrum of the Ricker wavelet. Because For the sample Ricker wavelet defined with the peak angular
the Ricker wavelet is real valued, we only consider the frequency frequency of ωp ≈ 188.5 rad∕s, the mean frequency is ωm ≈
range from zero to positive infinite. 200.5 rad∕s, in contrast to the central frequency ωc ≈ 199.6 rad∕s,
The three definite integrals in the two expressions above are de- and two times the deviation is 2ωσ ≈ 129.6 rad∕s, in contrast to
rived in Appendix A: the full frequency band is 2ωb ≈ 217.7 rad∕s. Using the two statis-
Z tical parameters (ωm , ωσ ), we can construct an equivalent distribution
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

∞ 3 in Gaussian,
D1 ¼ R2 ðωÞdω ¼ pffiffiffiffiffi ;
0 8 2π ωp
Z  
∞ 1 4 1 ðω − ωm Þ2
D2 ¼ ωR2 ðωÞdω ¼ ; PðωÞ ¼ exp − −2 : (21)
0 2π πω2p 2 ω2σ
Z  
∞ ωp 15 3 ωm ω 2
D3 ¼ ðω−ωm Þ2 R2 ðωÞdω ¼ pffiffiffiffiffi þ 2 − m.
0 2π 32 8 ωp π As plotted in the dashed curve in Figure 3a, this Gaussian distribution
is very close to the power spectrum of the Ricker wavelet.
(18)
Once we evaluate the mean frequency ωm and the standard der-
ivation ωσ statistically, we are able to calculate the peak frequency
Then, the mean frequency is ωp and then, following expression 16, the central frequency ωc and
rffiffiffi the half-bandwidth ωb . Because the central frequency, the mean fre-
D 4 2 quency, the half-bandwidth, and the standard derivation are all ex-
ωm ¼ 2 ¼ ω ≈ 1.063846ωp ; (19) pressed in terms of the peak frequency ωp, the relationships for
D1 3 π p
deriving the central frequency and the half-bandwidth are given by

and the standard deviation is


rffiffiffisffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
3 π 1 1
sffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffi ωc ¼ −W 0 − þ −W −1 − ωm ;
D3 5 32 8 2 2e 2e
ωσ ¼
D1
¼ ωp
4 9π
− ≈ 0.343848ωp : (20) pffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
3 π 1 1
ωb ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi −W −1 − − −W 0 − ωσ .
45π −128 2e 2e
(22)

They are approximately ωc ≈ 0:995534ωm and ωb ≈ 1.679438ωσ .


Because the mean frequency ωm, evaluated from the power spec-
trum, is very close to the central frequency ωc, practically, we
can use the evaluated mean frequency as an approximated central
frequency in seismic analysis.

FREQUENCY QUANTITIES EVALUATED USING


THE AMPLITUDE SPECTRUM
In some of the published literature, the mean frequency is evalu-
ated based on the amplitude spectrum, rather than the power spec-
trum. For the Ricker wavelet, the amplitude spectrum jRðωÞj ¼
RðωÞ. The mean frequency and the standard deviation can be evalu-
ated by

R∞
ωRðωÞdω
ωmðaÞ ¼ R0 ∞ ;
0 RðωÞdω
R ∞ 
ðω − ωm Þ2 RðωÞdω 1∕2
ωσðaÞ ¼ 0 R∞ ; (23)
Figure 3. (a) Statistical frequency evaluation using the power spec- 0 RðωÞdω
trum of the Ricker wavelet. The solid vertical lines indicate the cen-
tral frequency ωc and the half-bandwidth ωb . The dashed vertical
lines are the mean frequency ωm and standard deviation ωσ . (b) Stat- where parentheses in the subscripts “ðaÞ” indicate that these
istical frequency evaluation using the amplitude spectrum. The quantities are evaluated from the amplitude spectrum. The three def-
dashed vertical lines are mean frequency ωmðaÞ and standard
deviation ωσðaÞ . The power spectrum is closer to a Gaussian distri- inite integrals in these two expressions are also derived in Appen-
bution (in the dashed curve) than the amplitude spectrum will be. dix A:
Frequencies of the Ricker wavelet A35

Z ∞ 1 CONCLUSIONS
D4 ¼ RðωÞdω ¼ ;
0 2
Z ∞ In this paper, we have analyzed the Ricker wavelet and particu-
ωp larly its frequency attributes. The results of this analysis can be sum-
D5 ¼ ωRðωÞdω ¼ pffiffiffi ;
0 π marized as follows:
Z ∞
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

D6 ¼ ðω − ωm Þ2 RðωÞdω 1) We have defined the time breadth and the frequency bandwidth
0 and presented them analytically in terms of the Lambert W
   2
3 2 1 2ωp function.
¼ ωp
2 − þ pffiffiffi − ωm : (24) 2) We have proved that the central frequency can be approximated
4 π 2 π
by the mean frequency, evaluated using the power spectrum
rather than the amplitude spectrum as used in some areas of
Then, the mean frequency is the published literature.
3) We have found that the standard deviation about the mean fre-
D5 2ωp quency is not, as suggested by the literature, the half-bandwidth
ωmðaÞ ¼ ¼ pffiffiffi ≈ 1.128379ωp ; (25) of the frequency spectrum.
D4 π
Moreover, we have established mathematical relationships between
and the half-bandwidth is the theoretical parameters (the central frequency and the bandwidth)
and the statistical quantities (the mean frequency and the standard
deviation) evaluated from either the power spectrum or the ampli-
sffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffi tude spectrum of the Ricker wavelet.
D6 3 4
ωσðaÞ ¼ ¼ ωp − ≈ 0.476194ωp : (26)
D4 2 π
ACKNOWLEDGMENTS
The author is grateful to the sponsors of the Centre for Reservoir
For the sample Ricker wavelet with the peak angular frequency of Geophysics, Imperial College London, for supporting this research.
60π rad∕s, the mean frequency is ωmðaÞ ≈ 221:7 rad∕s, and two
times the deviation is 2ωσðaÞ ≈ 179.5 rad∕s. For comparison, Fig-
APPENDIX A
ure 3b plots a Gaussian curve defined with these two parameters:
THE DEFINITE INTEGRALS
 
2 1 ðω − ωmðaÞ Þ2 This appendix summarizes the derivation of the definite integrals
AðωÞ ¼ pffiffiffi exp − − 1 : (27) used in expressions 17 and 23 to evaluate the mean frequency and
π ωp 2 ω2σðaÞ the standard deviation.
The three definite integrals in expression 17, evaluation based on
the power spectrum of the Ricker wavelet, are given as the following:
If we use the amplitude spectrum to evaluate the mean frequency
and the deviation, we have the following two relationships for the Z Z  
∞ 4ω4 ∞ 2ω2
central frequency and the half-bandwidth: D1 ¼ R ðωÞdω ¼
2
exp − 2 dω
0 0 πωp
6 ωp
Z ∞ 4   pffiffiffi 
pffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi! 1 4ω 2ω 2 2ω
π 1 1 ¼ pffiffiffi exp − 2 d
ωc ¼ ωmðaÞ −W 0 − þ −W −1 − ; 2πωp 0 ω 4
p ω p ω p
4 2e 2e Z ∞
pffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ! ¼ pffiffiffi
1 3
x4 expð−x2 Þdx ¼ pffiffiffiffiffi ; (A-1)
π 1 1 2πωp 0 8 2π ωp
ωb ¼ ωσðaÞ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi −W −1 − − −W 0 − :
6π −16 2e 2e
(28) Z ∞
D2 ¼ ωR2 ðωÞdω
They are approximately ωc ≈0.938598ωmðaÞ and ωb ≈1.212682ωσðaÞ . 0
 
Z ∞ 5
As shown in Figure 3 with respect to a comparison between the 4ω 2ω2
power spectrum and the amplitude spectrum, the power spectrum is ¼ exp − 2 dω
0 πωp
6 ωp
more concentrated than the amplitude spectrum. The mean fre- Z ∞  2 2    2
quency evaluated from the power spectrum is closer to the central 1 2ω 2ω2 2ω
¼ exp − d
frequency (ωm ≈ 1.004486ωc ) than that evaluated from the ampli- 4π 0 ωp2 ωp2 ω2p
tude spectrum (ωmðaÞ ≈ 1.065454ωc ). The standard deviation of the Z ∞
1
power spectrum (ωσ ≈ 0.344ωp ) is narrower than that in the ampli- ¼ y2 expð−yÞdy
tude spectrum (ωσðaÞ ≈ 0.476ωp ). Figure 3 also shows that the
4π 0
power spectrum is closer to a Gaussian distribution (in the dashed 1
¼ ; (A-2)
curve) than the amplitude spectrum would be. 2π
A36 Wang
Z ∞
and
D5 ¼ ωRðωÞdω
0
Z  
Z ∞
2ω3 ω2
∞ ¼ pffiffiffi 3 exp − 2 dω
D3 ¼ ðω−ωm Þ2 R2 ðωÞdω 0 π ωp ωp
Z ∞
0
  Z ∞ 2    2
ωp ω ω2 ω
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

4 2ω2
¼ ðω−ωm Þ 2 4ω
exp − 2 dω ¼ p ffiffiffi exp − 2 d 2
πω6p ωp π 0 ω2p ωp ωp
0 Z
Z ∞  pffiffiffi 6 pffiffiffi 4  ωp ∞
ωp 1 2ω ω2m 2ω ¼ pffiffiffi y expð−yÞdy
¼ p ffiffi
ffi þ 2 π 0
2π 0 2 ω p ω p ω p ωp
  pffiffiffi  ¼ pffiffiffi ; (A-6)
2ω 2 2ω π
×exp − 2 d
ωp ωp
Z      2 and
ωm ∞ 2ω2 2 2ω2 2ω
− exp − 2 d Z ∞
2π 0 ωp2 ωp ω2p
Z ∞  D6 ¼ ðω − ωm Þ2 RðωÞdω
ωp 1 6 ω2m 4 0
Z  
¼ pffiffiffi x þ 2 x expð−x2 Þdx ∞ 2ω2 ω2
2π 0 2 ωp ¼ ðω − ωm Þ2 pffiffiffi 3 exp − 2 dω
Z ∞ π ωp ωp
ω 0
− m y2 expð−yÞdy Z ∞     
2π 0 2 ω 4
ω 2
ω2 ω
   rffiffiffi 2 ¼ pffiffiffi ωp 4 þ ωm 2 exp − 2 d
2 2

3ωp 5 32 3 4 2 π 0 ωp ωp ωp ωp
¼ pffiffiffiffiffi − þ pffiffiffiffiffi ωm − ω ; (A-3) Z    2
8 2π 4 9π 8 2π 3 π p 2ωm ωp ∞ ω2 ω2 ω
− pffiffiffi exp − d 2
π 0 ωp
2
ω2p ωp
Z ∞
pffiffiffi 2
where x ¼ 2ω∕ωp and y ¼ 2ω2 ∕ω2p . The derivations above have ¼ pffiffiffi ðω2p x4 þ ω2m x2 Þ expð−x2 Þdx
π 0
applied the following definite integral formulas: Z
2ωm ωp ∞
− p ffiffiffi y expð−yÞdy
π 0
Z pffiffiffi    
∞ 3 π 2 3 2 1 2ωp 2
x4 expð−x2 Þdx ¼
; ¼ ωp − þ ωm − pffiffiffi ; (A-7)
0 8 4 π 2 π
Z ∞ pffiffiffi
3 π
ðax þ bx Þ expð−x Þdx ¼
6 4 2
ð5a þ 2bÞ; where x ¼ ω∕ωp and y ¼ ω2 ∕ω2p . In the derivation above, we have
0 16
Z ∞ used the following definite integrals:
y2 expð−yÞdy ¼ 2: (A-4) Z pffiffiffi
∞ π
0
x2 expð−x2 Þdx ¼
;
0 4
Z ∞ pffiffiffi pffiffiffi
3 π π
ðax þ bx Þ expð−x Þdx ¼
4 2 2

ffiffiffiffiffiffiffiffiffiffiffiffiffiffi is ωm ¼ D2 ∕D1 , and the standard
Then, the mean p frequency b;
0 8 4
deviation is ωσ ¼ D3 ∕D1 . Z ∞
The three definite integrals in expression 23, evaluation based on y expð−yÞdy ¼ 1: (A-8)
the amplitude spectrum, are given as the following: 0

Then, the mean frequency is ωmðaÞ ¼ D5 ∕D4 and the standard


pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z ∞ deviation is ωσðaÞ ¼ D6 ∕D4 .
D4 ¼ RðωÞdω
0
Z  
∞2ω2 ω2 REFERENCES
¼ pffiffiffi 3 exp − 2 dω
0 π ωp ωp Barnes, A. E., 1993, Instantaneous spectral bandwidth and dominant fre-
Z ∞ 2     quency with applications to seismic reflection data: Geophysics, 58,
2 ω ω2 ω 419–428, doi: 10.1190/1.1443425.
¼ pffiffiffi exp − d Berkhout, A. J., 1984, Seismic resolution: Resolving power of acoustical
π 0 ω2p ω2p ωp echo techniques: Geophysical Press Handbook of Geophysical Explora-
Z ∞ tion Series 1.
2
¼ p ffiffiffi x2 expð−x2 Þdx Carter, A. J., and M. J. Kendall, 2006, Attenuation anisotropy and the rel-
π 0 ative frequency content of split shear waves: Geophysical Journal
International, 165, 865–874, doi: 10.1111/j.1365-246X.2006.02929.x.
1 Cohen, L., and C. Lee, 1989, Standard deviation of instantaneous frequency:
¼ ; (A-5) IEEE Proceedings of International Conference on Acoustics, Speech, and
2 Signal Processing, 2238–2241, doi: 10.1109/ICASSP.1989.266910.
Frequencies of the Ricker wavelet A37

Cohen, L., and C. Lee, 1990, Instantaneous bandwidth for signals and Lambert, J. H., 1772, Observations analytiques: Nouveaux Mémoires de
spectrogram: IEEE Proceedings of International Conference on Acous- l’Académie Royale des Sciences et Belles-lettres Berlin, 1, for 1770.
tics, Speech and Signal Processing, 2451–2454, doi: 10.1109/ICASSP Loughlin, P. J., and K. L. Davidson, 2001, Modified Cohen-Lee time-fre-
.1990.116086. quency distributions and instantaneous bandwidth of multicomponent sig-
Corless, R. M., G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey, and D. E. Knuth, nals: IEEE Transactions on Signal Processing, 49, 1153–1165, doi: 10
1996, On the Lambert W function: Advances in Computational Math- .1109/78.923298.
ematics, 5, 329–359, doi: 10.1007/BF02124750. Loughlin, P. J., and B. Tacer, 1997, Comments on the interpretation of in-
Euler, L., 1779, De serie Lambertina plurimisque eius insignibus proprieta- stantaneous frequency: IEEE Signal Processing Letters, 4, 123–125, doi:
Downloaded 02/06/15 to 129.31.136.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

tibus (On the remarkable properties of a series of Lambert and others): 10.1109/97.575553.
Opera Omnia (Series 1) (Originally published in Acta Academiae Scien- Quan, Y., and J. M. Harris, 1997, Seismic attenuation tomography using the
tarum Imperialis Petropolitinae, 1779, 29–51), 6, 350–369. frequency shift method: Geophysics, 62, 895–905, doi: 10.1190/1.1444197.
Gram-Hensen, K., 1991, A bandwidth concept for CPB time-frequency analy- Ricker, N., 1943, Further developments in the wavelet theory of seismogram
sis: IEEE Proceedings of International Conference on Acoustics, Speech, structure: Bulletin of the Seismological Society of America, 33, 197–228.
and Signal Processing, 2033–2036, doi: 10.1109/ICASSP.1991.150803. Ricker, N., 1944, Wavelet functions and their polynomials: Geophysics, 9,
Hosken, J. W. J., 1988, Ricker wavelets in their various guises: First Break, 314–323, doi: 10.1190/1.1445082.
6, 24–33, doi: 10.3997/1365-2397.1988002. Ricker, N., 1953, The form and laws of propagation of seismic wavelets:
Hu, C., N. Tu, and W. Lu, 2013, Seismic attenuation estimation using an Geophysics, 18, 10–40, doi: 10.1190/1.1437843.
improved frequency shift method: IEEE Geoscience and Remote Sensing Wang, Y., 2014, Stable Q analysis on vertical seismic profiling data: Geo-
Letters, 10, 1026–1030, doi: 10.1109/LGRS.2012.2227933. physics, 79, no. 4, D217–D225, doi: 10.1190/geo2013-0273.1.
Lambert, J. H., 1758, Observationes variae in mathesin puram: Acta Helve- Wang, Y., 2015, The Ricker wavelet and the Lambert W function: Geophysi-
tica: Physico-mathematico-anatomico-botanico-medica, Band III, 128– cal Journal International, 200, 111–115, doi: 10.1093/gji/ggu384.
168.

S-ar putea să vă placă și