|Summary:||ASTERISK-03589: [request] Brazilian Caller ID Detection - DTMF without polarity reversal|
|Date Opened:||2005-02-25 13:19:38.000-0600||Date Closed:||2011-06-07 14:05:14|
|Description:||On Brazil the caller id comes before the first ring, without any notice, no polarity reversal, absolute nothing... just a quick burst and the DTMF tones.|
Soren tried to detect the first burst using the simple patch on this post:
that patch simulates a polarity reversal, and as noted on the post, generates some false positives.
****** ADDITIONAL INFORMATION ******
bug ASTERISK-5 and ASTERISK-1258 mentioned this problem, both of them are closed now and the problem isn't solved.
bug ASTERISK-1695 make a good aproach for UK BT lines (v23) but use the unwelcome but efficient history buffer
|Comments:||By: nick (nick) 2005-02-25 20:22:16.000-0600|
If you can find me this weekend on IRC (#asterisk-dev on irc.freenode.net) with a machine to test on, I've got an idea on this...
By: Clod Patry (junky) 2005-03-13 22:59:47.000-0600
Any updates here?
Your idea worked for that bug?
By: allgood (allgood) 2005-03-14 05:35:18.000-0600
the problem isn't the detection of the start, the best way is to fake a polarity reversal on the first high signal that cames through the line. It would be great if this resource can get to HEAD now.
Our problem is the detection of tones on dsp.c, it appears that brazilian caller id tones have a too low volume. I'll try to tweak dsp.c, with the help of someone, I hope. For now I'm learning it. Please keep this bug open.
By: Brian West (bkw918) 2005-03-14 23:12:42.000-0600
Please move this to the wiki's bounty page. The intrest in this bug isn't much so you'll have to find someone that might wanna code it for you. Then we can reopen the bug and add it once we have something thats workable.