I'm looking at building a cost-effective active noise control system.

Basically, the system comprises two microphones - one reference and one error. A DSP unit samples and phase flips the reference microphone and plays the flipped signal against the input for destructive interference.

The error microphone then feeds-forward the resulting signal, which should be 0 (but is never quite). How might one develop a proportional control algorithm for the DSP unit to adapt the reverse signal from the error?

  • $\begingroup$ active noise cancellation is kind of a big topic :) So, the problem is that proportionality in your control algorithm might simply not cut it, starting with the problem that the path from the noise source and the cancellation signal speaker will not be equal in general, thus necessitating the need to first delay before doing the phase flip. Now, that delay will in practice be frequency-selective, because audio things typically don't have constant group delay. $\endgroup$ – Marcus Müller Jun 6 '18 at 12:15
  • $\begingroup$ Is that assuming the noise source is some oscillatory repeating signal, the speaker will have to wait one cycle to play the sample at which the microphone picked up? @MarcusMüller $\endgroup$ – Thomas Shemeld Jun 6 '18 at 12:20
  • $\begingroup$ Ah, periodicity of noise is a nice restriction to start with! You'd still need a way to estimate (and correct) the time difference of arrival from the noise and cancellation signal at the point(s) you want the signals to cancel each other $\endgroup$ – Marcus Müller Jun 6 '18 at 12:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.