Re: [BLACKBOX] Another LibVector question

From: [at]} <Wojtek>
Date: Sat, 28 Mar 2009 08:21:30 -0400

----boundary-LibPST-iamunique-1303109147_-_-
Content-type: text/plain

Robert:

   thanks a lot! I will read the bitmap docu and I will make a note
in a comment in the program. I will also update Lib. For now it
just works because I intentionally defined all my FIRs to be the same
length, even though some of my filters could be shorter. But I wanted
to avoid latency mismatch among the filters. If they are all the same,
then any latency or offset is the same too and the indices in the
*processed* waveforms can find one another. That's what I need...

I have been doing quite a different type of research lately and
now I have to get up to speed on this one asap. Hence I am rushing
to the result and skipping the theories.

My other research is summarized here. Please everyone take a look:

     http://www.picasaweb.com/WojtekSkulski

Cheers!

Wojtek


On Sat, 28 Mar 2009, Robert wrote:

> Wojtek Skulski wrote:
>> It is truly confusing. The correlator adds its latency to the correlated
>> waveform. The latency is equal to 1/2 the FIR length. But the Conv function
>> adds the correlator length to the correlated wave.
>>
>> All the indices are now mismatched with respect to the original
>> feature in the ADC waveform.
>
>> It is confusing at best. Any comments why it should be so?
>
>
> Read the Docu extract I sent as a bitmap (*carefully*)!
>
> (The extract was for Complex, for Real just ignore the Conjugations.)
>
> The Docu explains how much the 'Correlation' output is offset, and why.
>
> The 'Convolution' output is not offset.
>
> Since, in your case, both functions return the same numbers these comments
> may seem contradictory, but they are different functions with different
> meanings and different purposes. Confusion, I contend, is generated by
> mentally mixing them.
>
>
> If you want to eliminate the 'Filter settling times' then do:
>
> z := Vec.Conv (x, h);
> z := Vec.Take (LEN (x), Vec.Loose (ASH (LEN (h), -1), z));
>
> This will line up z with x if the length of the filter h is odd. If it is
> even, bad luck!, but you will be within half a sample.
>
>
> Sometimes one wants the settling time data, sometimes one does not. If the
> library function removed it you would not be able put it back when you wanted
> to. This way you have the option.
>
> Cheers
>
> Robert
>
>
> ----
> To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to
> LISTSERV{([at]})nowhere.xy
>
>

----
To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy----boundary-LibPST-iamunique-1303109147_-_-
Content-type: application/rtf
Content-transfer-encoding: base64
Content-Disposition: attachment; filename="rtf-body.rtf"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----boundary-LibPST-iamunique-1303109147_-_---

Received on Sat Mar 28 2009 - 13:21:30 UTC