[hpsdr] Morse transcription project

[Frank Brickle]

On Thu, Dec 2, 2010 at 3:25 PM, Chris Albertson
<albertson.chris at gmail.com> wrote:

> ...
> Why not do the work entirely in the frequency domain?    My theory for
> doing this was that when an operator presses his key down he is in
> effect creating a bit of a spike in the spectra of the band.  I did
> not see much reason for ever taking the signals down to base band

One big reason for going to baseband is that you can then also work on
downsampled data.

There are also certain kinds of things you might like to be able to
do, which sort of need to be done in the time domain -- computing LPC
coefficients, for example, or taking the wavelet transform.

Also, you're up against Heisenberg uncertainty, which might or might
not work at odds with the other kinds of analysis you're bringing to
bear -- for example, probability models of delays between successive
events in a causal vs. random signal.

In short, probably better to be prepared to deal with signal on both
sides (time and freq).

73
Frank
AB2KT

-- 
Pourtant, j'étais fort mauvais poète.
Je ne savais pas aller jusqu'au bout.
-- Blaise Cendrars, "Prose du Transsibérien et de la petite Jehanne de France"

 1291565303.0