[Laser] forward error correction and time interleaving

[TWOSIG at aol.com]

I was reading about MT63 to see if it could be adapted for light  
communication.  The idea is to send 64 simultaneous tones, evenly spaced,  and each BPSK 
modulated.  The data is encoded so that even if 16 of the  tones are lost, the 
data can still be recovered.  Add to that, interleave  the data over a time 
interval.  I follow some of it, but not as much as I  would like.
 
The consept is to spread the data over the frequency and time domains so  
that impulse noise and frequency fading do not result in loss of data.  Are  
these the sort of problems that need to be addressed in light  communication?  
Lightning is a good example of impulse noise.  Is  there a light communication 
difficulty that would relate to frequency  fading?
 
I did not understand why, but statements were made to suggest that MT63 is  
not appropriate for weak signal work.
 
I do recognise that MT63 is a system that depends on linear delivery.   The 
distortion of the signal tones that would be expected from modulating a  laser 
or gas discharge tube directly, would result in lots of noise within the  
spectrum.  Could this be addressed by selecting tone frequencies that would  be 
less affected by the non-linear distortion products?
 
One concept that I had was to pulse width modulate the composite tone  
signal.  The nonlinear issues then are pushed outside the band of interest  for 
decoding the signals.
 
Perhaps there is another way to adapt the concepts of data interleaving and  
forward error correction to problems relating to light communication.
 
 
James
N5GUI