I'm working on a project that requires me to wirelessly transmit analog information over a short range (few meters at maximum) to a receiving end from two transmitting ends. The tricky part is I can only use one receiver and I have to build these transmitter-receiver parts from scratch, meaning no microcontrollers or ready-made modules (like those 433MHz RF transceivers that I see commonly used).

I want to use IR or LASER LEDs for this purpose but am unsure how that would turn out. What kind of output would I observe from a photodiode if I were to shine two LASER beams on it simultaneously? For instance, if I were to drive the two LASER diodes with PWM and have a significant difference between their frequencies (5kHz and 100kHz for example), would I be able to filter the two transmitted signals using basic filters with corresponding center frequencies on the receiving end? I don't need extreme precision, as long as I can compare the amplitudes of my original signals on the receiving end I'm good.

I realize that with this kind of set-up the best method would be time division multiplexing but I'm unsure how to achieve that with analog circuitry.

Would you guys suggest any other method for this type of wireless communication? I don't think I have enough experience with electronics to build my own RF transceiver but if you have any tips on that it would also be appreciated. Thank you!

  • \$\begingroup\$ Can you tell us how quickly the analog signal changes. \$\endgroup\$
    – RoyC
    Dec 24, 2016 at 9:18
  • \$\begingroup\$ Thank you for the reply first of all. The analog signal is originally a constant DC level taken from a temperature sensor with 10mV/Celsius sensitivity. I was thinking of converting this to PWM. Since I have two of these DC levels, I thought I could have the two converted signals with different frequencies, so in summary I would have control over their frequency. (I hope I understood your question correctly and this reply answers it) \$\endgroup\$ Dec 24, 2016 at 10:22

2 Answers 2


I think you are almost there with the pwm idea. Keep the pwm modulation at the same frequency and then take the output of this at each transmitter and use it to modulate a much higher carrier frequency, different in each transmitter. Filter the two carriers at the receiver and you end up with two PWM modulated signals.

As long as you are not saturating the photodiode the received signal should be the sum of the two incoming signals.

  • \$\begingroup\$ Thats basically what happens in fibre comms. +1. I would prob lose the pwm in favour of AM though. \$\endgroup\$
    – Andy aka
    Dec 24, 2016 at 9:54
  • \$\begingroup\$ How would you calibrate the AM signal for different distances and variations in components etc? \$\endgroup\$
    – RoyC
    Dec 24, 2016 at 9:59
  • 1
    \$\begingroup\$ I'd send a reference signal at a different carrier frequency. I'm not keen on pwm because of overlapping sideband issues but it could be done that way. \$\endgroup\$
    – Andy aka
    Dec 24, 2016 at 10:02
  • \$\begingroup\$ I like that but we are piling on the complication for what I suspect is a college assignment. Note the much higher carrier f. \$\endgroup\$
    – RoyC
    Dec 24, 2016 at 10:17
  • \$\begingroup\$ You are right in assuming it is a college assignment :) So I have to admit I am favoring simplicity of implementation over precision and stability for this project. \$\endgroup\$ Dec 24, 2016 at 10:28

Well-separated FM, PPM, or PWM modulated carriers would be best. The bigger issue is maintaining calibration if you intend to use the analog voltage inputs to directly modulate the carriers. You can probably initially hand-adjust things to get reasonable accuracy for a while, but this won't last. There are many ways to improve the long-term accuracy:

  • Multiplexing reference values along with the measured signal.
  • Digitizing the measured signal and sending the data via FM or PPM/PWM
  • Using analog-to-digital conversion and digital frequency synthesis to generate the FM, and digital PLL demodulation at the receiver.

These methods will be complicated to build using discrete components, but fairly simple using (for example) an Arduino. With the Arduino (or similar) I would digitize the analog input and send the data via binary FM.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.