Re: Interpolation

Eric Jacobsen wrote:
> 
> On Tue, 01 Apr 2008 19:37:07 -0600, jim <"sjedgingN0sp"@m@mwt.net>
> wrote:
> 
> >
> >
> >Eric Jacobsen wrote:
> >
> >> >If the process requires an AA filter to reduce bandwidth they
> >> >"necessarily" will.
> >>
> >> Back to my previous example with a solitary tone in a unity gain
> >> passband.   If the decimation rate is integer and the samples are not
> >> shifted in time, what will be different about the output samples as
> >> compared to the corresponding input samples?
> >
> >That is why the word "requires" is in the sentence. Your example does not
> >require reducing bandwidth therefore it has already been excluded from
> >consideration.
> 
> It seems that you're saying that the case to be considered is one
> where energy is removed by the antialiasing filter.

	I wasn't the one who said it, but yes,  that was the case that was
offered for consideration. For some reason considering that case has
proven to be extremely difficult for some folks.


>  If that's the
> case, then I'd go back to my previous point that I don't think anybody
> here has said that an output waveform should look like the input
> waveform if energy has been removed.   Clearly that's the normal
> function of a filter.   That's a bit orthogonal, I think, to the
> definition of "interpolation", though, unless you want the definition
> of "interpolation" to be conditional on whether or not energy was
> removed during the process.   Is that what you're suggesting?

I wasn't the one who made the suggestion originally. But understanding the
simple concept that was suggested is no where near as difficult as you are
making it out to be. 


> 
> >> >> And since the topic is also terminology "anti aliasing" filters are
> >> >> usually analog.
> >> >
> >> > If your point is digital AA filters do not exist - then why not just say
> >> >that?
> >>
> >> I wasn't making a point, I was asking a question, which you snipped:
> >>
> >> " A decimating filter may need to be frequency
> >> selective in order to prevent aliasing in the output, but is that
> >> really an anti-aliasing filter or just a properly designed decimating
> >> (or downsampling) filter?"
> >
> >I didn't snip that question - I answered it. I responded: "prevent
> >aliasing" = "anti-aliasing".
> >
> >       You are asking if a filter designed to prevent aliasing is really an
> >anti-alias filter. How much of an answer do you want to a question like
> >that?
> 
> It's the same sort of question of whether or not an upsampling filter
> should be called an interpolator, and you've been happy to participate
> in that discussion.   As I've already said, a common definition of
> "anti-alias filter" is the analog filter preceding an ADC, and for a
> lot of folks that's the only way the term is used.  Using it
> generically to describe a frequency-selective decimating filter could
> cause the same confusion if the majority of relevant people thought of
> the analog filter when you meant the digital.  Do you not think that's
> worth clearing up?   Or do you think that the terminology confusion is
> okay?

The confusion you refer to, as far as I can tell, is not something that is
common to the world at large. it only exists in this thread. The use of
anti-alias filters for downsampling is not, as far as I know, something
that causes confusion.



> 
> >> Since a fair portion of the discussion is about terminology, and a few
> >> folks have been using "anti-aliasing" together with decimating
> >> filters, I thought it was a pertinent question.  Since "anti-aliasing"
> >> is usually an analog filter function preceding an ADC in my
> >> experience, I'm curious how many people also apply the term to
> >> decimating filters?   It seems unusual to me.
> 
> >
> >Well the point wasn't to spawn another tangential discussion about what
> >name you give something. The point being made was why the word
> >interpolation isn't used when the process involves reducing the frequency
> >content.
> >
> >-jim
> 
> And I think the word "interpolation" as commonly understood can
> certainly be applied to frequency-selective filters, whether they're
> upsampling or downsampling or preserving the sample rate.   If you're
> saying that "interpolation" should only be conditionally applied based
> on whether or not frequency selectivity is implemented I'd be
> interested in any arguments for why that should be the case.
> 
> After all, linear interpolation adds distortion to a signal, so a
> sample rate change (or even just time shifter)  that uses linear
> interpolation would change the frequency content of the signal, so is
> it not interpolation when used in that manner?


 The process called interpolation is simply to find new points on a
continuous curve that passes thru the input samples. The original question
was why the word "interpolation" is used by some to mean upsampling but
not downsampling. A simple explanation was offered. The explanation sounds
perfectly reasonable to me, but confusing to you. And as near as I can
tell there is nothing that can be done about that.

-jim


----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.pronews.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= - Total Privacy via Encryption =---

Eric Jacobsen wrote:
> 
> On Tue, 01 Apr 2008 19:37:07 -0600, jim <"sjedgingN0sp"@m@mwt.net>
> wrote:
> 
> >
> >
> >Eric Jacobsen wrote:
> >
> >> >If the process requires an AA filter to reduce bandwidth they
> >> >"necessarily" will.
> >>
> >> Back to my previous example with a solitary tone in a unity gain
> >> passband.   If the decimation rate is integer and the samples are not
> >> shifted in time, what will be different about the output samples as
> >> compared to the corresponding input samples?
> >
> >That is why the word "requires" is in the sentence. Your example does not
> >require reducing bandwidth therefore it has already been excluded from
> >consideration.
> 
> It seems that you're saying that the case to be considered is one
> where energy is removed by the antialiasing filter.

	I wasn't the one who said it, but yes,  that was the case that was
offered for consideration. For some reason considering that case has
proven to be extremely difficult for some folks.


>  If that's the
> case, then I'd go back to my previous point that I don't think anybody
> here has said that an output waveform should look like the input
> waveform if energy has been removed.   Clearly that's the normal
> function of a filter.   That's a bit orthogonal, I think, to the
> definition of "interpolation", though, unless you want the definition
> of "interpolation" to be conditional on whether or not energy was
> removed during the process.   Is that what you're suggesting?

I wasn't the one who made the suggestion originally. But understanding the
simple concept that was suggested is no where near as difficult as you are
making it out to be. 


> 
> >> >> And since the topic is also terminology "anti aliasing" filters are
> >> >> usually analog.
> >> >
> >> > If your point is digital AA filters do not exist - then why not just say
> >> >that?
> >>
> >> I wasn't making a point, I was asking a question, which you snipped:
> >>
> >> " A decimating filter may need to be frequency
> >> selective in order to prevent aliasing in the output, but is that
> >> really an anti-aliasing filter or just a properly designed decimating
> >> (or downsampling) filter?"
> >
> >I didn't snip that question - I answered it. I responded: "prevent
> >aliasing" = "anti-aliasing".
> >
> >       You are asking if a filter designed to prevent aliasing is really an
> >anti-alias filter. How much of an answer do you want to a question like
> >that?
> 
> It's the same sort of question of whether or not an upsampling filter
> should be called an interpolator, and you've been happy to participate
> in that discussion.   As I've already said, a common definition of
> "anti-alias filter" is the analog filter preceding an ADC, and for a
> lot of folks that's the only way the term is used.  Using it
> generically to describe a frequency-selective decimating filter could
> cause the same confusion if the majority of relevant people thought of
> the analog filter when you meant the digital.  Do you not think that's
> worth clearing up?   Or do you think that the terminology confusion is
> okay?

The confusion you refer to, as far as I can tell, is not something that is
common to the world at large. it only exists in this thread. The use of
anti-alias filters for downsampling is not, as far as I know, something
that causes confusion.



> 
> >> Since a fair portion of the discussion is about terminology, and a few
> >> folks have been using "anti-aliasing" together with decimating
> >> filters, I thought it was a pertinent question.  Since "anti-aliasing"
> >> is usually an analog filter function preceding an ADC in my
> >> experience, I'm curious how many people also apply the term to
> >> decimating filters?   It seems unusual to me.
> 
> >
> >Well the point wasn't to spawn another tangential discussion about what
> >name you give something. The point being made was why the word
> >interpolation isn't used when the process involves reducing the frequency
> >content.
> >
> >-jim
> 
> And I think the word "interpolation" as commonly understood can
> certainly be applied to frequency-selective filters, whether they're
> upsampling or downsampling or preserving the sample rate.   If you're
> saying that "interpolation" should only be conditionally applied based
> on whether or not frequency selectivity is implemented I'd be
> interested in any arguments for why that should be the case.
> 
> After all, linear interpolation adds distortion to a signal, so a
> sample rate change (or even just time shifter)  that uses linear
> interpolation would change the frequency content of the signal, so is
> it not interpolation when used in that manner?


 The process called interpolation is simply to find new points on a
continuous curve that passes thru the input samples. The original question
was why the word "interpolation" is used by some to mean upsampling but
not downsampling. A simple explanation was offered. The explanation sounds
perfectly reasonable to me, but confusing to you. And as near as I can
tell there is nothing that can be done about that.

-jim


----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.pronews.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= - Total Privacy via Encryption =---