Hi,
according to Wkipedia, the Mexican hat wavelet is the negative
normalized second derivative  of a Gaussian function. I can verfiy
this analytically and numerically. But I don't understand how to
relate the resulting pulse's frequency to the parameters of the
Gaussian function. Could someone enlighten me?

Kamran

On Apr 12, 4:37=A0pm, Kamran Iranpour <kamran.iranp...@gmail.com> wrote:
> Hi,
> according to Wkipedia, the Mexican hat wavelet is the negative
> normalized second derivative =A0of a Gaussian function. I can verfiy
> this analytically and numerically. But I don't understand how to
> relate the resulting pulse's frequency to the parameters of the
> Gaussian function. Could someone enlighten me?
>
> Kamran

This is not surprising - a gaussian function and its derivatives
consists of an infinite number of frequencies. But if you want the
frequency content of the mexican hat function, just find its fourier
transform. You will have to do integration by parts twice and then
know that the transform of a gaussian is also a gaussian.

But what you are likely wanting is how to compare a windowed fourier
transform kernal to the mexican hat function. This is described in
Daubechies (10 lectures on wavelets).


Clay

Clay wrote:
> On Apr 12, 4:37 pm, Kamran Iranpour <kamran.iranp...@gmail.com> wrote:
>> Hi,
>> according to Wkipedia, the Mexican hat wavelet is the negative
>> normalized second derivative  of a Gaussian function. I can verfiy
>> this analytically and numerically. But I don't understand how to
>> relate the resulting pulse's frequency to the parameters of the
>> Gaussian function. Could someone enlighten me?
>>
>> Kamran
> 
> This is not surprising - a gaussian function and its derivatives
> consists of an infinite number of frequencies. But if you want the
> frequency content of the mexican hat function, just find its fourier
> transform. You will have to do integration by parts twice and then
> know that the transform of a gaussian is also a gaussian.

Or look up the Fourier transform of the Gaussian in a Fourier transform 
table, then observe that given a time-domain signal x and its Fourier 
transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}.  Apply 
that twice, and there you are.

> But what you are likely wanting is how to compare a windowed fourier
> transform kernal to the mexican hat function. This is described in
> Daubechies (10 lectures on wavelets).


-- 
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Tim Wescott <tim@seemywebsite.now> wrote:
(snip)
 
> Or look up the Fourier transform of the Gaussian in a Fourier transform 
> table, then observe that given a time-domain signal x and its Fourier 
> transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}.  Apply 
> that twice, and there you are.

The latter should also be in a Fourier transform table, likely
earlier than the transform of a Gaussian.
 
>> But what you are likely wanting is how to compare a windowed fourier
>> transform kernal to the mexican hat function. This is described in
>> Daubechies (10 lectures on wavelets).

-- glen

On Apr 13, 12:30=A0am, Tim Wescott <t...@seemywebsite.now> wrote:
> Clay wrote:
> > On Apr 12, 4:37 pm, Kamran Iranpour <kamran.iranp...@gmail.com> wrote:
> >> Hi,
> >> according to Wkipedia, the Mexican hat wavelet is the negative
> >> normalized second derivative =A0of a Gaussian function. I can verfiy
> >> this analytically and numerically. But I don't understand how to
> >> relate the resulting pulse's frequency to the parameters of the
> >> Gaussian function. Could someone enlighten me?
>
> >> Kamran
>
> > This is not surprising - a gaussian function and its derivatives
> > consists of an infinite number of frequencies. But if you want the
> > frequency content of the mexican hat function, just find its fourier
> > transform. You will have to do integration by parts twice and then
> > know that the transform of a gaussian is also a gaussian.
>
> Or look up the Fourier transform of the Gaussian in a Fourier transform
> table, then observe that given a time-domain signal x and its Fourier
> transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}. =A0Apply
> that twice, and there you are.
>
> > But what you are likely wanting is how to compare a windowed fourier
> > transform kernal to the mexican hat function. This is described in
> > Daubechies (10 lectures on wavelets).
>
> --
> Tim Wescott
> Control system and signal processing consultingwww.wescottdesign.com

I am a bit slow here. Suppose you wanted to have a pulse with a
certain center frequency and a certain length. How would one then go
from this to deciding what are the appropriate parameter values of
'a', 'b' and 'c' in the Gaussian a*exp^(-(x-b)^2/(2*c^2)) where a, b
and c are some positive numbers ? Or given its statistical equivalent
b would be the mean and c^2 the varaiance.
Sorry about taking your time.

Kamran

On Apr 13, 12:15=A0pm, Kamran Iranpour <kamran.iranp...@gmail.com>
wrote:
> On Apr 13, 12:30=A0am, Tim Wescott <t...@seemywebsite.now> wrote:
>
>
>
>
>
> > Clay wrote:
> > > On Apr 12, 4:37 pm, Kamran Iranpour <kamran.iranp...@gmail.com> wrote=
:
> > >> Hi,
> > >> according to Wkipedia, the Mexican hat wavelet is the negative
> > >> normalized second derivative =A0of a Gaussian function. I can verfiy
> > >> this analytically and numerically. But I don't understand how to
> > >> relate the resulting pulse's frequency to the parameters of the
> > >> Gaussian function. Could someone enlighten me?
>
> > >> Kamran
>
> > > This is not surprising - a gaussian function and its derivatives
> > > consists of an infinite number of frequencies. But if you want the
> > > frequency content of the mexican hat function, just find its fourier
> > > transform. You will have to do integration by parts twice and then
> > > know that the transform of a gaussian is also a gaussian.
>
> > Or look up the Fourier transform of the Gaussian in a Fourier transform
> > table, then observe that given a time-domain signal x and its Fourier
> > transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}. =A0Appl=
y
> > that twice, and there you are.
>
> > > But what you are likely wanting is how to compare a windowed fourier
> > > transform kernal to the mexican hat function. This is described in
> > > Daubechies (10 lectures on wavelets).
>
> > --
> > Tim Wescott
> > Control system and signal processing consultingwww.wescottdesign.com
>
> I am a bit slow here. Suppose you wanted to have a pulse with a
> certain center frequency and a certain length. How would one then go
> from this to deciding what are the appropriate parameter values of
> 'a', 'b' and 'c' in the Gaussian a*exp^(-(x-b)^2/(2*c^2)) where a, b
> and c are some positive numbers ? Or given its statistical equivalent
> b would be the mean and c^2 the varaiance.
> Sorry about taking your time.
>
> Kamran- Hide quoted text -
>
> - Show quoted text -

I don't think you are understanding the problem. The center frequency
of the gaussian pulse is determined by "b." But it contains much more
than just one frequency.  The pulse length is infinite.

On Apr 13, 9:14=A0pm, Clay <c...@claysturner.com> wrote:
> On Apr 13, 12:15=A0pm, Kamran Iranpour <kamran.iranp...@gmail.com>
> wrote:
>
>
>
> > On Apr 13, 12:30=A0am, Tim Wescott <t...@seemywebsite.now> wrote:
>
> > > Clay wrote:
> > > > On Apr 12, 4:37 pm, Kamran Iranpour <kamran.iranp...@gmail.com> wro=
te:
> > > >> Hi,
> > > >> according to Wkipedia, the Mexican hat wavelet is the negative
> > > >> normalized second derivative =A0of a Gaussian function. I can verf=
iy
> > > >> this analytically and numerically. But I don't understand how to
> > > >> relate the resulting pulse's frequency to the parameters of the
> > > >> Gaussian function. Could someone enlighten me?
>
> > > >> Kamran
>
> > > > This is not surprising - a gaussian function and its derivatives
> > > > consists of an infinite number of frequencies. But if you want the
> > > > frequency content of the mexican hat function, just find its fourie=
r
> > > > transform. You will have to do integration by parts twice and then
> > > > know that the transform of a gaussian is also a gaussian.
>
> > > Or look up the Fourier transform of the Gaussian in a Fourier transfo=
rm
> > > table, then observe that given a time-domain signal x and its Fourier
> > > transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}. =A0Ap=
ply
> > > that twice, and there you are.
>
> > > > But what you are likely wanting is how to compare a windowed fourie=
r
> > > > transform kernal to the mexican hat function. This is described in
> > > > Daubechies (10 lectures on wavelets).
>
> > > --
> > > Tim Wescott
> > > Control system and signal processing consultingwww.wescottdesign.com
>
> > I am a bit slow here. Suppose you wanted to have a pulse with a
> > certain center frequency and a certain length. How would one then go
> > from this to deciding what are the appropriate parameter values of
> > 'a', 'b' and 'c' in the Gaussian a*exp^(-(x-b)^2/(2*c^2)) where a, b
> > and c are some positive numbers ? Or given its statistical equivalent
> > b would be the mean and c^2 the varaiance.
> > Sorry about taking your time.
>
> > Kamran- Hide quoted text -
>
> > - Show quoted text -
>
> I don't think you are understanding the problem. The center frequency
> of the gaussian pulse is determined by "b." But it contains much more
> than just one frequency. =A0The pulse length is infinite.

I know there are more than one frequencies, thats why I called it
center frequency. For me it appears to be decided not only by 'b' but
by 'c' too.

On Apr 14, 2:45=A0am, Kamran Iranpour <kamran.iranp...@gmail.com> wrote:
> On Apr 13, 9:14=A0pm, Clay <c...@claysturner.com> wrote:
>
>
>
>
>
> > On Apr 13, 12:15=A0pm, Kamran Iranpour <kamran.iranp...@gmail.com>
> > wrote:
>
> > > On Apr 13, 12:30=A0am, Tim Wescott <t...@seemywebsite.now> wrote:
>
> > > > Clay wrote:
> > > > > On Apr 12, 4:37 pm, Kamran Iranpour <kamran.iranp...@gmail.com> w=
rote:
> > > > >> Hi,
> > > > >> according to Wkipedia, the Mexican hat wavelet is the negative
> > > > >> normalized second derivative =A0of a Gaussian function. I can ve=
rfiy
> > > > >> this analytically and numerically. But I don't understand how to
> > > > >> relate the resulting pulse's frequency to the parameters of the
> > > > >> Gaussian function. Could someone enlighten me?
>
> > > > >> Kamran
>
> > > > > This is not surprising - a gaussian function and its derivatives
> > > > > consists of an infinite number of frequencies. But if you want th=
e
> > > > > frequency content of the mexican hat function, just find its four=
ier
> > > > > transform. You will have to do integration by parts twice and the=
n
> > > > > know that the transform of a gaussian is also a gaussian.
>
> > > > Or look up the Fourier transform of the Gaussian in a Fourier trans=
form
> > > > table, then observe that given a time-domain signal x and its Fouri=
er
> > > > transform F{x}, the Fourier transform of dx/dt is j*omega*F{x}. =A0=
Apply
> > > > that twice, and there you are.
>
> > > > > But what you are likely wanting is how to compare a windowed four=
ier
> > > > > transform kernal to the mexican hat function. This is described i=
n
> > > > > Daubechies (10 lectures on wavelets).
>
> > > > --
> > > > Tim Wescott
> > > > Control system and signal processing consultingwww.wescottdesign.co=
m
>
> > > I am a bit slow here. Suppose you wanted to have a pulse with a
> > > certain center frequency and a certain length. How would one then go
> > > from this to deciding what are the appropriate parameter values of
> > > 'a', 'b' and 'c' in the Gaussian a*exp^(-(x-b)^2/(2*c^2)) where a, b
> > > and c are some positive numbers ? Or given its statistical equivalent
> > > b would be the mean and c^2 the varaiance.
> > > Sorry about taking your time.
>
> > > Kamran- Hide quoted text -
>
> > > - Show quoted text -
>
> > I don't think you are understanding the problem. The center frequency
> > of the gaussian pulse is determined by "b." But it contains much more
> > than just one frequency. =A0The pulse length is infinite.
>
> I know there are more than one frequencies, thats why I called it
> center frequency. For me it appears to be decided not only by 'b' but
> by 'c' too.- Hide quoted text -
>
> - Show quoted text -

Your "a" is linked to "c" via normalization. "b" of course translates
(moves back and forth) the gaussian. If your gaussian is in the time
domain, then "b" does not affect the frequency content other than its
phase info. If your gaussian is in the freq domain, then "b" clearly
translates in the frequency domain and thus sets the center frequency.
Back in the time domain, this results in a modulated gaussian. Look up
Heaviside's shifting theorem for details. Your "c" does not affect
your center frequency - it sets the variance of the gaussian and the
normalization.

Try to find Ingrid Daubechies' book "Ten Lectures on Wavelets," She
really does cover the similarities and differences between the Mexican
Hat Function and using a Windowed Fourier transform. You question
strikes me as asking something akin to "What is the frequency of a
parabola?" There is not a good answer because the question is ill
posed.