AEC | Double Talk | ERLE | G.168 | LEC | NLMS

Hi,

I would like to know if it is normal to include a 2nd redundant echo canceler
algorithm for line echo cancellation.

I came across brief notes on the internet that suggested use of two echo
cancelers in line echo cancellation. The additional echo canceler is enabled in
case the primary echo canceler goes out of bounds (diverges). I could not find
much useful information on this subject.

1. Is this a general rule in LEC (including 2 cancelers)?
2. What kind of algorithms are preferred or generally used in the 2nd redundant
echo canceler?

Any pointers would be of great help. Thanks.

Regards,
Deepak

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of echocancel -- send a blank email to echocancel-subscribe@yahoogroups.com )

Hi,
>
>I would like to know if it is normal to include a 2nd redundant echo
canceler algorithm for line echo cancellation.
>
>I came across brief notes on the internet that suggested use of two echo
cancelers in line echo cancellation. The additional echo canceler is enabled in
case the primary echo canceler goes out of bounds (diverges). I could not find
much useful information on this subject.
>
>1. Is this a general rule in LEC (including 2 cancelers)?
>2. What kind of algorithms are preferred or generally used in the 2nd
redundant echo canceler?
>
>Any pointers would be of great help. Thanks.

I'm not sure there are any general rules in LEC. One common strategy is
certainly to maintain two or more sets of FIR coefficients. If the working set
seems to be drifting badly off tune, the canceller reverts to a "known
good" set. The details of how this works is left as an exercise for the
reader.

Steve

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of echocancel -- send a blank email to echocancel-subscribe@yahoogroups.com )

Hi Deepak,=20
=A0
(1) There is no rule in echo cancellation to use two echo cancellers.=20
(2) Some authors suggest to use additional filters or best converged filter=
 coefficest set=20
to give better echo cancellation in case of echo path variations or=A0in ca=
se of=A0doble talk.=A0=A0=A0=A0=A0 Some authors call this additional filter=
s as "Auxiliary filters". It may use same algorithm as the primary
echo can=
celler or a simplified one. Your logic should automatically=A0 turn on/off =
of the primary/auxiliary echo canellers depend on the situation. For refere=
nce please refer: =A0"Siho Kim, Kennsung=A0=A0Bae, 2002 IEEE paper ....
Tit=
le: =A0"Applying an auxiliary fitler in ....=A0=A0=A0".
=A0
Regards,
Sameer

--- On Tue, 4/21/09, Deepak Garise <g...@yahoo.com> wrote:
From: Deepak Garise <g...@yahoo.com>
Subject: [echocancel] Redundant or 2nd echo canceler in Line Echo Cancellat=
ion
To: e...@yahoogroups.com
Date: Tuesday, April 21, 2009, 11:56 PM

Hi,

I would like to know if it is normal to include a 2nd redundant echo cancel=
er algorithm for line echo cancellation.

I came across brief notes on the internet that suggested use of two echo ca=
ncelers in line echo cancellation. The additional echo canceler is enabled =
in case the primary echo canceler goes out of bounds (diverges). I could no=
t find much useful information on this subject.

1. Is this a general rule in LEC (including 2 cancelers)?
2. What kind of algorithms are preferred or generally used in the 2nd redun=
dant echo canceler?

Any pointers would be of great help. Thanks.

Regards,
Deepak

=20=20=20=20=20=20

_____________________________________

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of echocancel -- send a blank email to echocancel-subscribe@yahoogroups.com )