[Elecraft] Auto Notch Revisited

[Sverre Holm (LA3ZA)]

In the context of the LMS filter for FIR notch filtering and noise
cancellation, one common way to find the reference is by a simple delay of
the input signal. If the purpose is to separate voice from a sinusoid, then
one utilizes the fact that a sinusoid has a long correlation function in
time compared to voice. If the delay is longer than usual correlation time
for voice, one will get a reference for the desired signal - at least in the
correlation domain which is what counts here.

For a noise reduction filter, the delay is often shorter, as noise has short
correlation time compared to a voiced signal.



David Woolley (E.L) wrote
> The problem with it in the context of this thread is that it needs a value
> for d(n). 
> Although d(n) is confusingly defined in the actual article, I believe it 
> represents the "correct" (desired) output of the filter.





-----
Sverre, LA3ZA

K2 #2198, K3 #3391,
LA3ZA Blog: http://la3za.blogspot.com,
LA3ZA Unofficial Guide to K2 modifications: http://la3za.blogspot.com/p/la3za-unofficial-guide-to-elecraft-k2.html
--
View this message in context: http://elecraft.365791.n2.nabble.com/Auto-Notch-Revisited-tp7591810p7591896.html
Sent from the Elecraft mailing list archive at Nabble.com.