On Mon, 28 Jul 2003 10:41:19 -0500, Keith Larson
<k-larson2@NOSPAM.ti.com> wrote:
>PS: Has anyone out there in comp.dsp land done any work on a time
>compression algorithm? It would definitely help with my busy schedule.
I've stopped work on that in order to make progress on the
space-expansion algorithm for my garage.
Eric Jacobsen
Minister of Algorithms, Intel Corp.
My opinions may not be Intel's opinions.
http://www.ericjacobsen.org
Reply by Keith Larson●July 28, 20032003-07-28
Hi All,
Yeh, the Sliding Goertzel (SG) is pretty cool, I have looked into it,
and I have working code. The SG has interesting benefits over the
Sliding DFT, and yet it also has some limitations.
The Sliding DFT is best for sample by sample algorithms where R/I data
is needed constantly (like a filter)
The Sliding Goetzel is best when you only need to sub-sample the filters
R/I information say every 10th sample. My opinion is that stopping the
SG process to 'mine' the filter data on every sample is somewhat costly
when compared to the SDFT and since most of what I do is using the SDFT
as a filter, I stick with the SDFT.
Otherwise, the SDFT and SG are equivelent and where the SG is best
suited, my advice is... use it.
I also found that a typical DSP has this thing called a multiplier
within it. And, not surprisingly, the MPY can be brought to bear
during adds, subtracts, reads and writes :-) This somewhat blunts the
SG's advantage over the SDFT, but the SG is definitely still a winner
when the 'data mining rate' is low. BTW, I might as well define
Data_Mine_Rate= Sample_Rate / Input_Samples_Per_Output_Samples
I would like to say that I have the Sliding Goertzel sanitized enough to
add it to the latest VC33/DSK3 software, but Ive been too busy with a
number of other priorioties. BTW, if anyone is interested, I even have
some PC side code that generates filter response curves for both Sliding
DFT and Sliding Goertzel. This later turned into a FIR filter generator
application for the new DSK.
If anyone is interested in the latest DSK software, it does contain a
fair bit of application material and write up regarding the Sliding DFT,
and a lot of other stuff. You can download the latest version from the
following Spectrum Digital (who is building the VC33 DSK for us) web link.
http://www.spectrumdigital.com/drivers/docstore/UniversityDSKs/TMS320VC33/
And, since I just mentioned the VC33 DSK, here is that link
http://focus.ti.com/general/docs/university/univ.tsp?templateId=5807&navigationId=10543&path=templatedata/cm/univgen/data/univ_pricing_ovw
Oh, Jerry... Next month I plan to be doing some planetary imaging of
Mars as it comes to opposition. If interested, some samples of my
plantary imaging work can be found at main link of my personal, non-TI
web page at
http://home.houston.rr.com/klarson
These planetary images are not taken using my C3x based CCD camera, but
rather a low light NTSC video camera. But never fear, a new C3x DSP
based camera is in the works... time permitting
Best regards,
Keith Larson
PS: Has anyone out there in comp.dsp land done any work on a time
compression algorithm? It would definitely help with my busy schedule.
+-----------------------------------------------+
|Keith Larson |
|Member Group Technical Staff |
|Texas Instruments Incorporated |
| |
| 281-274-3288 |
| k-larson2@ti.com |
| www.micro.ti.com/~klarson |
|-----------------------------------------------+
| TMS320C3x/C4x/VC33 Applications |
| |
| TMS320VC33 |
| The lowest cost and lowest power 500 �w/Mflop |
| floating point DSP on the planet! |
+-----------------------------------------------+
Jerry Avins wrote:
> Rick Lyons wrote:
>
> ...
>
>> Ya know, the notion of a sliding Goertzel filter was relayed to me in
>> a phone conversation with Keith Larson (Tex. Inst. DSP guru, and
>> frequent poster here.) I'm not sure where Keith encountered the
>> sliding Goertzel idea, but I'm willing to bet that it just 'came'
>> to him. He eats, sleeps, and breaths DSP.
>>
>> See Ya, [-Rick-]
>>
>
> Keith doesn't sleep. :^) At night, he takes breathtaking astronomical
photographs. (Digitally processed, of course.) You can see some at
> http://home.houston.rr.com/klarson/PICTURES.HTM
>
> Jerry
>> photographs. (Digitally processed, of course.) You can see some at
http://home.houston.rr.com/klarson/PICTURES.HTM
>
> Jerry
Reply by Jerry Avins●July 28, 20032003-07-28
Rick Lyons wrote:
>
...
>
> Ya know, the notion of a sliding Goertzel filter was relayed to
> me in a phone conversation with Keith Larson (Tex. Inst.
> DSP guru, and frequent poster here.) I'm not sure where
> Keith encountered the sliding Goertzel idea, but I'm willing to bet
> that it just 'came' to him. He eats, sleeps, and breaths DSP.
>
> See Ya,
> [-Rick-]
Keith doesn't sleep. :^) At night, he takes breathtaking astronomical
photographs. (Digitally processed, of course.) You can see some at
http://home.houston.rr.com/klarson/PICTURES.HTM
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Steve Underwood●July 27, 20032003-07-27
Rick Lyons wrote:
> Ya know, the notion of a sliding Goertzel filter was relayed to
> me in a phone conversation with Keith Larson (Tex. Inst.
> DSP guru, and frequent poster here.) I'm not sure where
> Keith encountered the sliding Goertzel idea, but I'm willing to bet
> that it just 'came' to him. He eats, sleeps, and
> breaths DSP.
I think you will find it goes back a long way.
Regards,
Steve
Reply by Rick Lyons●July 27, 20032003-07-27
On Tue, 22 Jul 2003 12:50:00 -0500, Dennis@NoSpam.com wrote:
>Besides Eric Jacobsen and Richard Lyons excellent article in the IEEE Signal
>Processing Magazine March 2003 issue entitled "The Sliding DFT" here are two
>Sliding Goertzel papers. (Unfortunately they are not free. Hope your Library
>has them or the pdf download will cost you 30 bucks!)
>
>(1) Joe F. Chicharo, Mehdi T. Kilani, "A Sliding Goertzel Algorithm" , Signal
>Processing, Vol 52, 1996 p283-297, Publisher=Elsevier Science B.V.
>
>(2) Joe F. Chicharo, Mehdi T. Kilani, "Arbitrary multifrequency estimation and
>the relationship between the LMS and Goertzel algorithms.", International J.
>Electronics, Vol 83, No.3 1997 p287-298
>
>Dennis
>
Hi Dennis, thanks for the references. I'll take a look at them.
Ya know, the notion of a sliding Goertzel filter was relayed to
me in a phone conversation with Keith Larson (Tex. Inst.
DSP guru, and frequent poster here.) I'm not sure where
Keith encountered the sliding Goertzel idea, but I'm willing to bet
that it just 'came' to him. He eats, sleeps, and
breaths DSP.
See Ya,
[-Rick-]
Reply by ●July 22, 20032003-07-22
Besides Eric Jacobsen and Richard Lyons excellent article in the IEEE Signal
Processing Magazine March 2003 issue entitled "The Sliding DFT" here are two
Sliding Goertzel papers. (Unfortunately they are not free. Hope your Library
has them or the pdf download will cost you 30 bucks!)
(1) Joe F. Chicharo, Mehdi T. Kilani, "A Sliding Goertzel Algorithm" , Signal
Processing, Vol 52, 1996 p283-297, Publisher=Elsevier Science B.V.
(2) Joe F. Chicharo, Mehdi T. Kilani, "Arbitrary multifrequency estimation and
the relationship between the LMS and Goertzel algorithms.", International J.
Electronics, Vol 83, No.3 1997 p287-298
Dennis