[Laser] AM versus PAM

[TWOSIG at aol.com]

Yves
 
Thank you for responding to my questions.  
 
I have considered a system using frequency modulation of a pulse  stream.  It 
would seem to have a lot of potential, but I have not  found a practical 
circuit for me to build - transmit or receive.  The  suggestion of 10 KHz pulse 
rate seems to require very sharp filtering  of the voice input, so I would 
probably try 20 - 30 KHz.  I will keep it in  mind for the future.
 
I am not familiar with a boxcar averager.  Can you describe what it  does and 
what sort of circuit is needed?
 
I understand ( at least a little bit ) that the more bandwidth that my  
receiver has, the more noise I will get in.  I do not see how that will  help me 
choose the modulation technique ( given choice of AM, PAM, or PWM )  for a set 
power output at the transmitter.  I am using a Ramsey receiver  for now ( 
basically a phototransistor followed by a 300 to 6000 Hz bandpass  filter driving 
an IC audio amp.  I doubt that I can create anything  significantly more 
sophisticated.  I might change the  filter frequencies, or sharpness, or use a more 
sensitive detector,  but it will still be a very broad bandwidth detector 
followed by audio  filtering.  In that, it would seem the noise bandwidth is 
completely  independent of the modulation method that I choose to use on the  
transmitter.
 
The system does work well with both PWM and AM, but I currently have no way  
to compare the PWM system that uses a laser pointer with the AM system that 
uses  and LED.  I was thinking about modifying the AM to operate either PAM or 
AM  at similar power.  Today I had another idea that might also provide PWM  
with the same basic circuit and try to match the power.  With the same  audio 
drive and light emitter current ( for now LED, but it should also work  with a 
modified laser pointer that I have, though AM may be limited linearity )  I 
should be able to compare the three modulation schemes.  Should I expect  them to 
perform the same?
 
 
I had thought that I could use a simple addition the the receive circuit to  
improve the performance of PAM over AM.  As long a the received pulses are  
stronger than the ambient light noise, stretching the pulses into the "off" time 
 should improve the performance so the audio filter "sees" more power.  What  
I was thinking is to use a peak detector with a bleed off resistor.  By  
capturing the peak input the pulse will stretch into the "off" time. If the peak  
detector stays up, we lose information on the next pulse.  However, letting  
the output of the detector bleed off, with a time constant long compared to the 
 pulse width, but short compared to the audio modulation, the distortion 
should  be acceptable.
 
Thank you again.
 
James
N5GUI
 
In a message dated 2/21/2006 11:29:36 AM Central Standard Time,  
F1AVYopto at aol.com writes:
With LED or lasers the pulse modulations give the  same problems.
One  can easily pulse a LED at low duty cycle and 1 to 10  watts peak power 
can be  output.
But ....if the pulses have T  duration, the bandwidth must to be 1/2T  to get 
the full pulse  amplitude in a linear mode RX.
If we increase the  bandwidth, we  increase the noise.
A threshold detector can detect the pulse  and a  boxcar averager can 
increase 
the S/N from a pulse constant rate but the  AM  modulation from pulses is not 
extract with these devices.
To use  the real  energy from a short pulse we have to use a none linear  
quantum detection with  APD in Geiger mode operation.
The system  works like photons detector and can  be triggered at the photons  
peaks.
The pulse rate could be modulated with a  FM sub carrier or  with a numerical 
coding frame.
The PWM (pulsed width  modulation)  with a LED is a good system to get a good 
modulation linearity. It   needs a low pass filter to reject the commutation 
frequency.
For the  noise  rejection in your experiment, it seems to me the best will be 
to  use a 10 KHz FM  sub carrier to modulate your LED diode and to use an IC  
demodulator in the  RX.
All the parasitic AM noise received will be  rejected.
Yves  F1AVY
http://pageperso.aol.fr/yvesf1avy/