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

Hi all, Tones really screw up the adaption of an EC. To get through the G.168 tests there seems to be no alternative but to detect that strong narrowband energy is present in the channel, and stop adaption while it is there. There are a number of obvious strategies for doing this, but they all seem to require substantial compute resources. I'm sure I am missing a simple solution which is adequate for the job. Google is a rich source of information about most aspects of EC design, but on this point I couldn't find anything (apart from lots of papers saying a narrowband detector is necessary). Can anyone comment on the techniques generally employed in real world G.168 compliant cancellers to deal with the strong narrowband energy problem? Regards, 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 )