Hi George,
My back-of-envelope says about 260 Hz bandwidth.
As a sanity check, I opened up Frerking ("Digital Signal Processing
in Communication Systems").
He characterises the bandwidth in terms of the ratio of the tone spacing
divided by data rate. His plot show examples for this ratio h, of .5
(very
narrow shift), .64 and 1.5. The h we use for our RTTY is 170/45 or
3.8, which is way wider than all of his examples.
Anyway, for his wide shift FSK case (h=1.5), his diagram of the
spectrum definitely dies off at the first zero of the baseband signal
(at least in the linear plot that he used :-).
If you use the first zero of a random baseband 45 baud signal, you
end up with the 260 Hz which I has guessed.
OK, perhaps it is worthwhile cranking out the actual spectrum of the
Baudot stuff that we use. Now... I need to find that FFT library...
give
me a little time (I also have to figure out how to get the plots onto a
web page).
73
Chen, AA6TY
|