Re: [BLACKBOX] Another LibVector question

From: [at]} <Wojtek>
Date: Fri, 27 Mar 2009 20:04:36 -0400

----boundary-LibPST-iamunique-1614340453_-_-
Content-type: text/plain

Robert:

   here is another question concerning LibVectors.Conv.

This procedure extends the original data and returns
a vector whose length is a sum of A+B, where A and B are vectors
being convolved. It seems extremely cumbersome to me.
Now I have to relate the index of a pattern found
in the convolved vector with the one in the original
vector.

As you know one of main applications of the convolution
is finding patterns. (Well, in this case the convolution
is called "correlation", but I am a practical person,
so I am not splitting the hair over those names).

Now I found the pattern by looking at maximum of the
correlation, but the maximum is shifted with respect
to original data. It is confusing at best.

Any comments why it should be so?

IGOR is quietly extending the data in order to provide
the "warmup region" for the FIR, but then it discards the
extended pieces before returning the correlation result
to the user.

I know you will not change LibVectors after they are out
for so many years.... but it is confusing to find those warmup
regions in the result...

W.

----
To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy----boundary-LibPST-iamunique-1614340453_-_-
Content-type: application/rtf
Content-transfer-encoding: base64
Content-Disposition: attachment; filename="rtf-body.rtf"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----boundary-LibPST-iamunique-1614340453_-_---

Received on Sat Mar 28 2009 - 01:04:36 UTC