I am working with ultrasound signals with frequency 100 kHz. Chirp signals with bandwidth 16 kHz are being modulated with 100 kHz and then transmitted via one ultrasonic sensor. With another ultrasonic sensor signals are being received, demodulated, low pass filtered and then matched filtered with transmitted signal. Can anyone back me up with the following reasoning: the peak of matched filter will be delayed by the time of flight between the sensors, time needed for ADC+DAC conversion and time spent for low-pass filtering and matched filtering? To remove the time offset due to ADC+DAC conversion and filtering I would try connecting output of DAC on input of ADC and find the peak of MF. Then for the measurements with the sensors I would subtract the previously mentioned offset time. Is that correct?

  • $\begingroup$ See the links I gave you on my last question that specifically detail the danger with this simplified approach of determining delay from a simple correlation. If there is dispersion or distortion in your channel the approach may not work well and won't be robust. (and added below as I think it answers this question) $\endgroup$ – Dan Boschen Jun 25 '20 at 2:31