[Asterisk-Users] BAD/GOOD Echo Cancel

Krystian Filiks krystian.filiks at kfiliks.com
Mon Feb 6 07:44:48 MST 2006


Doug Lytle wrote:

> Krystian Filiks wrote:
>
>> Did you test the echo delay?
>> will 64ms be suffitient?
>>
>> You can easily test the delay by recording the transmit and receive 
>> path to a sound file and using some sound editing software see how 
>> big the delay is.
>> That is how I did it when I worked for a Telco in Switzerland on 
>> theis TDM switch they had a big Echo problem on their SS7 links and 
>> PRI's
>>
>> Krystian
>>
>
> No, but the reports for the last 1 1/2 hours of use is very 
> encouraging.  No echo so far.  The echo was on the near side 
> (Employees heard their own voices).
>
> Doug
>
> _______________________________________________

If they could hear their own voices than I would not invest in echo 
cancelling and for this is the far end responcible so I would take 
contact with the service suppliers and ask them if echo canceling is 
included.

As far as I know (from my SS7 training) the common practise (default 
function) of echo cancelers on PRI/SS7 networks is that the you cancel 
echo for the farend and the farend for you. If you want to cancel your 
own echo then you have to turn the canceller back to front and if you 
want to cancel both ends then you have to combine 2 cancellers turned 
with the back to eachother.

We had a problem that our supplier refuzed to give us cancellins because 
it was against their policy (the route was below 1000km) so we ended up 
with setting up 2 cancellers turned with the backs to eachother and 
cancelled echo both for the far and nearend.

Echo cancelling is somewat exiting but expencive.

What do the stats from the cancellers show is there any echo that is 
beeing cancelled?

Krystian




More information about the asterisk-users mailing list