[dsp] PWM/phase relationship

[Scott Miller]

> Hmm, why did you chose to do it that way?
> 
> If the resolution of your calculation is greater than those 4 
> bits, there
> are a few tricks you can do to expand it to have those 8 
> samples be a bit
> more meaningful.

The limitation is really how fast the processor can get data ready.  It's
got 192 bytes of RAM, which isn't enough to buffer a whole message.  It
winds up doing HDLC bit stuffing and FCS calculation on the fly.  And
really, it seems to be more than adequate.  I've never had any trouble with
proper demodulation.

> The idea is to send PWM encoded voice--so as not to have to 
> modulate the
> intensity of the laser--which takes some clever work and 
> doesn't buy us
> much.  On and off are about all I can trust it to do safely. 
> :)  I don't

I haven't messed with laser communications since I was a kid, and we didn't
have diodes, just He-Ne tubes.  Which I suppose were easy enough to do AM
with, but I don't think you'd want to pulse them at any significant rate.
For long range it was easier just to shine it through a fan to modulate it.

> Hmm, that may work as you'll have plenty of gain from 
> directivity--I assume
> you'll keep a telescope on it as the 'receive antenna'.  You 

Yep, that's the idea.

> a yellow filter over the lens to act as a 'blue block' to 
> filter out any
> sky noise--which should mostly be blue. 

Or just fly it at night.  =]

> 'face to face', so it's not really an issue.  Expecially since they're

Warning: Do not look into laser with remaining eye.   =]

Scott
N1VG