"Ian McBride" <ianmcbride2003@yahoo.com> wrote in message
news:bfagg9$1sv$1@slb3.atl.mindspring.net...
> "Jim Thomas" <jthomas@bittware.com> wrote in message
> news:3F1800AA.C0893999@bittware.com...
> >
> > This might help:
> > http://www.dspguru.com/info/faqs/mrfaq.htm
>
> Great!  Just shove zeros in there and run it through an FIR..  ..WOW..
>
> OK.  Another question.  Lets say I have FSK data at 1200 bits per second.
I
> translate 0's to -5000Hz and 1's to +5000Hz and compose a "baseband" of 1
> sample per bit.  Can I zero stuff this with 79 zeroes, run it through an
FIR
> filter to get my filtered FSK baseband?  This seems confusing.  I'm
> surprised that all of those zeroes wouldn't give you DC output with minor
> "bumps" showing up 1200 times a second.
>
> Thanks!
>

A little thought will show you that this won't work. Let's say your data is
all ones, which should produce a 5 kHz tone in the baseband. You propose
only one sample every 1/1200 sec, but the Nyquist rate is Fs > 10 kHz, which
must be satisfied at every point in your system. You can't interpolate your
way out of it.

John

"Fred Marshall" <fmarshallx@remove_the_x.acm.org> wrote in message
news:rSzTa.3405$Jk5.2403537@feed2.centurytel.net...

> Ian,
>
> The answer is "it depends".  If you can meet all the specs with one filter
> then why not? - subject to breaking the filter up into stages, etc.
>
> This implies that you want to upsample/interpolate prior to
modulating.....
> I don't understand what your implementation block diagram looks like so
it's
> hard to comment.  But it seems a little strange.

This is a personal experiment -- I am carrying on a digital base transmitter
project that my company dropped, for my own education.  So, it my indeed be
strange.

A block diagram of sorts.. The transmitter DSP accepts data for
transmission -- either 8ksample/s PCM voice, or digital data in the form of
4800 baud 1's and 0's.  The PCM voice is converted into FM deviations, and
the data is converted into FSK deviations.    The voice data is interpolated
to 96ksamples per second, by zero-stuffing and polyphase filtering, and the
FSK data is "interpolated" by repeating each sample 19 times to create a
perfect "square wave" of  4800 baud data at 96ksamples per second.  In
either mode, the resulting 96ksamples per second is filtered (premodulation
filter) and then converted into a step angle for a quadrature NCO also
running at 96000 iterations per second which produces an I/Q "baseband"
representation of the FM signal.  The 96,000 samples per second of I and Q
FM are then fed to a graychip part that further interpolates them to
60,000,000 samples per second and mixes them with another NCO to create a
21.4mhz IF of the FM signal.  This connects to an A/D converter, to RF
circuitry, to a PA, to an antenna.

My understanding is that the premodulation filter for data is a 12th-order
Bessel with a 3.9khz cutoff, while the voice is a Butterworth with a 3.9khz
cutoff.  I was assuming that the interpolated data would need to be run
through an IIR filter before it is handed to the NCO for conversation into
step angles.  Also, the output of the NCO will be interpolated and filtered
big time by the graychip part, which needs to be programmed with its own FIR
filter coefficients.

Some of this was put together by the consultants before we backed off of the
project.  I am taking it at face value that the design makes sense.  I am
continuing it because (1) I believe we will eventually do it, and (2) I want
to learn a bit more on how these things work.  If it is restarted, I would
like to again be the one managing it, but I need to know more about the
subject.

Right now I am right after the interpolation, but just prior to the
premodulation filter, wondering if the two can be combined.  However, I am
happy to hear "big picture" comments from anyone reading through my above
ramblings!  If this is all missing the point on some level, then I'd like to
know that too.

Thanks to everyone for all the help.

"Ian McBride" <ianmcbride2003@yahoo.com> wrote in message
news:bfm041$f31$1@slb6.atl.mindspring.net...
>
> "Jon Harris" <jon_harrisTIGER@hotmail.com> wrote in message
> news:3f1aeb4d$1_1@newsfeed...
> > In fact, "connecting the dots" is usually a "bad idea" for two reasons:
> >
> > 1) If the in-between data is all zeros, the FIR filter can be optimized
to
> > execute _much_ faster.  The zero valued samples can simply be skipped
> since
> > they contribute nothing to the result.
> >
> > 2) Linear interpolation is a low-pass filtering operation, and not a
very
> > good one at that.  It introduces some roll-off in the pass-band which is
a
> > problem in some applications.  True, it does give you some additional
> > stop-band attenuation, but you are almost always better off to skip the
> > linear interpolation and use the time saved to do a better (longer)
single
> > FIR.
>
> OK.  Now, I am wondering if the "zero-stuffing/polyphase FIR"
interpolation
> method can also serve as a premodulation filter.  I.e., if my
pre-modulation
> filter needs to be an nth-order Butterworth or Bessel with a certain
cutoff,
> can the FIR filter smoothing out all of those zeros also take care of this
> too, or is this usually another IIR stage operating on the interpolated
> data?
>

Ian,

The answer is "it depends".  If you can meet all the specs with one filter
then why not? - subject to breaking the filter up into stages, etc.

This implies that you want to upsample/interpolate prior to modulating.....
I don't understand what your implementation block diagram looks like so it's
hard to comment.  But it seems a little strange.

Fred

"Jon Harris" <jon_harrisTIGER@hotmail.com> wrote in message
news:3f1aeb4d$1_1@newsfeed...
> In fact, "connecting the dots" is usually a "bad idea" for two reasons:
>
> 1) If the in-between data is all zeros, the FIR filter can be optimized to
> execute _much_ faster.  The zero valued samples can simply be skipped
since
> they contribute nothing to the result.
>
> 2) Linear interpolation is a low-pass filtering operation, and not a very
> good one at that.  It introduces some roll-off in the pass-band which is a
> problem in some applications.  True, it does give you some additional
> stop-band attenuation, but you are almost always better off to skip the
> linear interpolation and use the time saved to do a better (longer) single
> FIR.

OK.  Now, I am wondering if the "zero-stuffing/polyphase FIR" interpolation
method can also serve as a premodulation filter.  I.e., if my pre-modulation
filter needs to be an nth-order Butterworth or Bessel with a certain cutoff,
can the FIR filter smoothing out all of those zeros also take care of this
too, or is this usually another IIR stage operating on the interpolated
data?

Thanks for all of the help!

"Jon Harris" <jon_harrisTIGER@hotmail.com> wrote in message
news:3f1c2fc1$1_2@newsfeed...
> "Fred Marshall" <fmarshallx@remove_the_x.acm.org> wrote in message
> news:YOeTa.3375$Jk5.2341110@feed2.centurytel.net...
> >
> > "Rick Lyons" <ricklyon@REMOVE.onemain.com> wrote in message
> > news:3f1d70f5.21806828@news.earthlink.net...
> > > On Mon, 21 Jul 2003 12:24:50 -0700, "Jon Harris"
> > > <jon_harrisTIGER@hotmail.com> wrote:
> > >
> > >   (snipped)
> > > >>
> > > >> Well, the most common method of interpolation is to
> > > >> stuff the 11 zeros in between your original samples
> > > >> and then lowpass filter the new sequence.  (As far as I know,
> > > >> there's no benefit in performing the 'linear interpolation'
> > > >> of putting the 11 dots on a straight line.)
> > > >
> > > >In fact, "connecting the dots" is usually a "bad idea" for two
reasons:
> > > >
> > > >1) If the in-between data is all zeros, the FIR filter can be
optimized
> > to
> > > >execute _much_ faster.  The zero valued samples can simply be skipped
> > since
> > > >they contribute nothing to the result.
> > >
> > > Hi Jon,
> > >
> > >
> > >   VERY GOOD POINT !!!
> > >
> > >
> > > [-Rick-]
> >
> > This is just another way of saying what a time-domain "polyphse" filter
is
> > doing, right?
> > I thought your book made it much clearer than most Rick.
> >
> > Fred
>
> Yes.  In my way of looking at things, the polyphase filter thing is just a
> mathematical formalization of some simple optimizations that would be
pretty
> obvious to most people if when implementing interpolation.

Jon,

Yes and it has such a high falutin' name too!  Who might imagine that
something with a name like that is just another way of saying: "now, common
sense about efficiency might suggest ...... (something like what you said)"
Worth mentioning so folks can make the connection - my purpose.

Fred

"Fred Marshall" <fmarshallx@remove_the_x.acm.org> wrote in message
news:YOeTa.3375$Jk5.2341110@feed2.centurytel.net...
>
> "Rick Lyons" <ricklyon@REMOVE.onemain.com> wrote in message
> news:3f1d70f5.21806828@news.earthlink.net...
> > On Mon, 21 Jul 2003 12:24:50 -0700, "Jon Harris"
> > <jon_harrisTIGER@hotmail.com> wrote:
> >
> >   (snipped)
> > >>
> > >> Well, the most common method of interpolation is to
> > >> stuff the 11 zeros in between your original samples
> > >> and then lowpass filter the new sequence.  (As far as I know,
> > >> there's no benefit in performing the 'linear interpolation'
> > >> of putting the 11 dots on a straight line.)
> > >
> > >In fact, "connecting the dots" is usually a "bad idea" for two reasons:
> > >
> > >1) If the in-between data is all zeros, the FIR filter can be optimized
> to
> > >execute _much_ faster.  The zero valued samples can simply be skipped
> since
> > >they contribute nothing to the result.
> >
> > Hi Jon,
> >
> >
> >   VERY GOOD POINT !!!
> >
> >
> > [-Rick-]
>
> This is just another way of saying what a time-domain "polyphse" filter is
> doing, right?
> I thought your book made it much clearer than most Rick.
>
> Fred

Yes.  In my way of looking at things, the polyphase filter thing is just a
mathematical formalization of some simple optimizations that would be pretty
obvious to most people if when implementing interpolation.

"Rick Lyons" <ricklyon@REMOVE.onemain.com> wrote in message
news:3f1d70f5.21806828@news.earthlink.net...
> On Mon, 21 Jul 2003 12:24:50 -0700, "Jon Harris"
> <jon_harrisTIGER@hotmail.com> wrote:
>
>   (snipped)
> >>
> >> Well, the most common method of interpolation is to
> >> stuff the 11 zeros in between your original samples
> >> and then lowpass filter the new sequence.  (As far as I know,
> >> there's no benefit in performing the 'linear interpolation'
> >> of putting the 11 dots on a straight line.)
> >
> >In fact, "connecting the dots" is usually a "bad idea" for two reasons:
> >
> >1) If the in-between data is all zeros, the FIR filter can be optimized
to
> >execute _much_ faster.  The zero valued samples can simply be skipped
since
> >they contribute nothing to the result.
>
> Hi Jon,
>
>
>   VERY GOOD POINT !!!
>
>
> [-Rick-]

This is just another way of saying what a time-domain "polyphse" filter is
doing, right?
I thought your book made it much clearer than most Rick.

Fred

On Mon, 21 Jul 2003 12:24:50 -0700, "Jon Harris"
<jon_harrisTIGER@hotmail.com> wrote:

  (snipped)
>>
>> Well, the most common method of interpolation is to
>> stuff the 11 zeros in between your original samples
>> and then lowpass filter the new sequence.  (As far as I know,
>> there's no benefit in performing the 'linear interpolation'
>> of putting the 11 dots on a straight line.)
>
>In fact, "connecting the dots" is usually a "bad idea" for two reasons:
>
>1) If the in-between data is all zeros, the FIR filter can be optimized to
>execute _much_ faster.  The zero valued samples can simply be skipped since
>they contribute nothing to the result.

Hi Jon,


  VERY GOOD POINT !!!


[-Rick-]
>

"Rick Lyons" <ricklyon@REMOVE.onemain.com> wrote in message
news:3f1804b6.17234828@news.earthlink.net...
> On Fri, 18 Jul 2003 08:09:16 -0400, "Ian McBride"
> <ianmcbride2003@yahoo.com> wrote:
>
> Hello Ian,
>
> >I am trying to figure out interpolation (I have given up on decimation
for
> >now).
>
> Humm, I've always found decimation easier to understand than
> interpolation.  Anyway ...
>
> >My scenario is that I have 8 ksamples per second audio (from a phone
line),
> >and I want to interpolate it to 96 ksamples per second.  I know my audio
is
> >band limited within 0-4kHz, but other than that it could be anything --
> >human voice, modem tones, etc.
> >
> >I understand the most basic concepts.  I could use linear interpolation,
> >where I connect the 8kHz dots with lines and then add 11 evenly spaced
dots
> >along these lines.  I suppose I could do this and then run the result
> >through a LPF filter and it would improve it.  But I suspect these
methods
> >are hardly state of the art.
>
> Well, the most common method of interpolation is to
> stuff the 11 zeros in between your original samples
> and then lowpass filter the new sequence.  (As far as I know,
> there's no benefit in performing the 'linear interpolation'
> of putting the 11 dots on a straight line.)

In fact, "connecting the dots" is usually a "bad idea" for two reasons:

1) If the in-between data is all zeros, the FIR filter can be optimized to
execute _much_ faster.  The zero valued samples can simply be skipped since
they contribute nothing to the result.

2) Linear interpolation is a low-pass filtering operation, and not a very
good one at that.  It introduces some roll-off in the pass-band which is a
problem in some applications.  True, it does give you some additional
stop-band attenuation, but you are almost always better off to skip the
linear interpolation and use the time saved to do a better (longer) single
FIR.

"Ian McBride" <ianmcbride2003@yahoo.com> wrote in message
news:bfagg9$1sv$1@slb3.atl.mindspring.net...
> "Jim Thomas" <jthomas@bittware.com> wrote in message
> news:3F1800AA.C0893999@bittware.com...
> >
> > This might help:
> > http://www.dspguru.com/info/faqs/mrfaq.htm
>
> Great!  Just shove zeros in there and run it through an FIR..  ..WOW..
>
> OK.  Another question.  Lets say I have FSK data at 1200 bits per second.
I
> translate 0's to -5000Hz and 1's to +5000Hz and compose a "baseband" of 1
> sample per bit.  Can I zero stuff this with 79 zeroes, run it through an
FIR
> filter to get my filtered FSK baseband?  This seems confusing.  I'm
> surprised that all of those zeroes wouldn't give you DC output with minor
> "bumps" showing up 1200 times a second.
>

Let's see if I understand:
FSK
+/-5,000Hz
Modulation of one bit per symbol: 0=-5,000Hz, 1=+5,000Hz
Bit/symbol rate of 1200 bits per second
You didn't say what the carrier frequency is.
You didn't say what the sample rate is.

I guess you consider the FSK baseband to be -1="0" and +1="1", something
like that?

Assign a carrier frequency of 50,000Hz.  Then the data has this as a center
frequency and the bandwidth will be something greater than 10kHz.

It appears what you would want to do is decimate, not interpolate and in the
process you would need to demodulate - which might be part of decimation.  I
think there's a nonlinear process needed here.

Fred