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

Hello!
Some of the authors say that to cope with echo in VoIP-enabled network, an echo canceller with
the increased number of taps is needed. They deem so because of the fact that a packet-switched
network incurs significant delay. However, as far as I am aware, an echo canceller is placed as
close to the source of echo as possible. Echo is a leak and is only possible in analog
networks, that is, in PSTN. An echo canceller must be placed at the point closest to the PSTN.
But then, why does the IP delay matter?.. The echo signal is delayed no more than before (from
the echo canceller's perspective). Am I right thinking so? Can it be that echo cancellers are
placed not only near the echo sources?
Thanks!
Stas

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