Scribd
Upload
232 views19 pages

09 Deconvolution

Deconvolution aims to recover the earth's reflectivity series from a recorded seismogram. The seismogram is the convolution of the seismic wavelet with the earth's impulse response. Deconvolution applies an inverse filter to the seismogram to estimate the impulse response. The inverse filter is designed by taking the inverse z-transform of the known seismic wavelet. Least-squares inverse filtering finds the filter coefficients that minimize the error between the actual and desired output, producing a better approximation than simple inverse filtering. There are different types of deconvolution methods, including spike deconvolution and predictive deconvolution.

Uploaded by

Phạm Nam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
232 views19 pages

09 Deconvolution

Deconvolution aims to recover the earth's reflectivity series from a recorded seismogram. The seismogram is the convolution of the seismic wavelet with the earth's impulse response. Deconvolution applies an inverse filter to the seismogram to estimate the impulse response. The inverse filter is designed by taking the inverse z-transform of the known seismic wavelet. Least-squares inverse filtering finds the filter coefficients that minimize the error between the actual and desired output, producing a better approximation than simple inverse filtering. There are different types of deconvolution methods, including spike deconvolution and predictive deconvolution.

Uploaded by

Phạm Nam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

Deconvolution

outline
What is deconvolution
How to apply deconvolution
Optimum Wiener Filter
Kinds of deconvolution

What is deconvolution

seismogram: the convolution of a seismic wavelet


w(t) with the Earths impulse response e(t) yields
the seismogram x(t)

w(t ) e(t ) x(t )

What is deconvolution
Deconvolution tries to recover the
reflectivity series from the
recorded seismogram

wX(t ) e(t ) x(t )


What we want to get is reflectivity series

outline
What is deconvolution
How to apply deconvolution
Optimum Wiener Filter
Kinds of deconvolution

How to apply deconvolutiongeneral idea


Several assumptions

The source waveform is known (in time


domain)
Reflectivity is a random process.

w(t ) e(t ) x(t )

How to apply deconvolutiongeneral idea


A deconvolution operator--inverse filter
Inverse filter * seismic wavelet =spike

f (t ) * w(t ) (t )
Inverse filter * seismogram = earths impulse response

f (t ) x(t ) e(t )
(Provided the source waveform is known)

Therefore, inverse filtering is a method of deconvolution

How to apply deconvolutiongeneral idea


A deconvolution operator--inverse filter

e(t ) f (t ) x(t )
x(t ) w(t ) e(t ) w(t ) f (t ) x(t )
(t ) w(t ) f (t )
f (t ) (t ) 1 / w(t )
Inverse filter operator f(t) converts the basic wavelet to a
spike at t=0.
Likewise, the inverse filter converts the seismogram to a
series of spikes that defines the earths impulse response.

How to apply deconvolutiongeneral idea


A flowchart for inverse filtering
Seismic wavelet
Compute Inverse
(filter operator)
*
Deconvolved
seismogram

Seismogram

How to apply deconvolutionZ-transform


Z-transform
Wavelet w(t): (1, -1/2)
1
W (Z ) 1 Z
2
1
1
1
F (Z )
1 Z Z 2 ...
1
2
4
1 Z
2

Coefficients of F(Z): (1, 1/2, 1/4,)

How to apply deconvolution


the inverse of the Source Wavelet

Filter design 1
Input wavelet: w(t): (1, -1/2)
The Z-transform: W (Z ) 1 1 Z
2
1
1
1
The inverse :
F (Z )
1 Z Z
1
2
4
1

...

The inverse Filter f(t): (1, , , )

Filter application
Truncated inverse Filter f(t): (1, )
Input wavelet w(t): (1, -)
Actual output: (1, 0, -1/4) [f(t)*w(t)]
Desired output: (1, 0, 0)

Filter design 2
Truncated inverse Filter: (1, , )
Input wavelet: (1, -)
Actual output: (1, 0, 0,-1/4)
Desired output: (1, 0, 0, 0)

How to apply deconvolution


the inverse of the Source Wavelet

Filter design 3)
Input wavelet: w(t): (-1/2, 1)
1
W (Z ) Z
The Z-transform:
2
1
F (Z )
2 4Z 8Z
1
The inverse :
Z

...

The inverse Filter f(t): (-2, -4, -8, )

Filter application
Truncated inverse Filter: (-2, -4)
Input wavelet: (-, 1)
Actual output: (1, 0, -4)
Desired output: (1, 0, 0)

Filter design 4
Truncated inverse Filter: (-2,-4,-8 )
Input wavelet: (-, 1)
Actual output: (1, 0, 0,-8)
Desired output: (1, 0, 0, 0)

How to apply deconvolution


Least-Squares Inverse Filtering

The inverse filtering described last slide


yields a good approximation to a zero-lag
spike output (1,0,0). Can we do even
better than that?
Least-squares inverse filtering

How to apply deconvolution


Least-Squares Inverse Filtering
Wavelet (1, -1/2)
Filter (a, b)
Let the error between the actual output and
the desired output(1,0,0) is minimum in the
least-squares sense
L ( a 1) 2 (b
5

,1 a

2
2

5
b

1,
2

a=? b=?

a 2
b
) ( ) 2
2
2

Kinds of deconvolution

Spike deconvolution

Predictive deconvolution

Several assumptions
Allow
formulatin
g the
convolutio
nal model
of the 1-D
seismogra
m by
equation

The earth is made up of horizontal layers of constant


velocity
The source generates a compressional plane wave
that impinges on layer boundaries at normal
incidence. No shear waves are generated
The source waveform does not change as it travels in
the subsurfaceit is stationary
The random ambient noise is zero
The source waveform is known (in time domain)
Reflectivity is a random process.
The seismic wavelet is minimum phase.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy