[linux-audio-dev] Latency and feedback problems: soundcard with live microphone pass-thru, optimal solution ?

fons adriaensen fons.adriaensen at skynet.be
Sat Oct 29 21:25:35 UTC 2005

On Sat, Oct 29, 2005 at 06:43:05PM +0100, James Courtier-Dutton wrote:

> Benno Senoner wrote:
> >As we know in analog PA gear you have the microphone feedback problem 
> >(usually it comes in form of high pitched whistle sounds).
> This problem is common as you know, and it has been solved a long time 
> ago. The solution is to do a slight frequency shift to the audio.
> Small frequency shifts are not noticeable to the listener, and that it 
> what is used at all those loud pop concerts.

Frequency shifting will give a few dB extra, not more. It's based on the
fact that a typical room response will have many very narrow peaks only
a few Hz or less apart, and shifting the frequencies will smooth out the
response as the sound circulates around the loop.

It's less useful for stage monitoring for two reasons: first, there is
usually a dominant direct path from the monitor to the artist and the mic
(otherwise the monitor isn't very useful), and the room response has less
impact on the loop gain, and second, the combination of the direct sound
and the shifted one can be very confusing for the artist.

> Another application of feedback suppression is satellites where the 
> forward uplink frequency is slightly different from the forward downlink 
> frequency.

Slightly different in this case usually means at least several tens of 
MHz, and often up and downlink will be in different microwave bands.
There'is only one application where up and downlink frequencies are 
linked to each other, and in that case they have a constant ratio, not
difference. This makes the downlink frequency coherent to an on-ground
reference, and thus allows for Doppler measurements.


More information about the Linux-audio-dev mailing list