[Laser] AW: coherence and scintillation

[F1AVYopto at aol.com]

James
You are  right on many points.
LEDs always give a lower scintillation level:
Like  with stars the scintillation spectrum keeps 90% of its energy in the 2 
to 90 Hz  range gap.
The bigger the optical devices diameter for the TX and the RX the  lower the 
scintillation amplitude.
The frequency peak of this scintillation  spectrum goes slowly down with the 
optical diameter.
I have some curves about  this phenomenon for stars.
LEDs are not perfect monochromatic devices and  their frequency spectrums are 
often wider than 20 nm.
Cancellation  probability is very low with this wide spectrum.
A semiconducteur laser has  about 2 nm of frequency spectrum with a very thin 
main peak.
Cancellations  probability becomes great near the focusing point.
The erratic cancellation  and additions on the photo detector give an extra 
noise with lot of "clicks"  that give a scintillation spectrum well over the 
KHz.
>In theory, twinkle  should increase with longer path through the atmosphere,
>and if there is  a local concentration of turbulence, the effect should be
>greater as you  get farther away from the concentration. ( You should get
>different  results if you pass a light beam through 10 Km of turbulent air
>then 10  Km of calm air when compared to calm air first then turbulent. )
>On the  other hand, coherence is degraded the further a light beam passes
>through  the air.  This degradation results in cancelation.  But it  would
>seem that you should get less effective cancellation as the  coherence is
>degraded.
I did some experiment with LEDs and laser  diodes in the 70 to 80 Km range.
With a laser the scintillation stays very  strong at all ranges and begins at 
the first Km!
I am not certain the level  increases strongly with the range.
Over 70 km the signal seems even less  noisy that to 20 km.
With a LED the scintillation amplitudes increases slowly  with the range.
Above 60 Km it becomes very deep but in calm air the signal  never goes to 0.
>Anyway, if there are two mechanisms, it does not seem to  help solve the
>problem of atmospheric noise on a light communication  channel.  The
>Australian experiments may be better for using LEDs  instead of lasers.  I
>started using LEDs instead of lasers for my  demonstrations, but that was
>because kids kept finding ways to look down  the light beam.  LEDs were
>easier to AM modulate and had all the  range that I needed, indoors and out.
May be it depends the modulation  choice.
For direct audio voice modulation LEDs seems better with a good  linearity 
and a low scintillation capability.
For high-speed data link the  lasers allow a very high power in the beam and 
the communication recovery  protocols can reduce the scintillation and the 
cancellation effects.
For a  not line of sight laser communication via a bouncing media it seems to 
me the  laser is the best because it allows a small illuminated area at a 
long  range that can be fully intercepted by the RX optic.
>From the TX to the  target the radar equation with its 1/D² law not apply and 
the losses are very  low into this beam. (Only the air absorption losses)
The 1/D² law applies  only from the target to the RX with a lambertian 
dispersion.
With a LED it is  often impossible to produce a beam smaller than the target 
and so the 1/D² law  applies with a very big loss on the path.
73 Yves F1AVY