[Asterisk-Users] DTMF Caller ID w/o polarity inversion

[Soren Rathje]

Renato Mintz wrote:
> Hi Folks,
>
> I've been looking around and found some references of some Caller ID
> patches (Mantis bug#9) for X100P and TDM400 for Netherlands, Sweden
> and UK. It's been quite hard to understand what has finally been
> incorporated to the distribution (if anything) or which patches must
> be applied in witch snapshot of the repository.
>
> I've tried some different approaches but nothing worked and my
> question finally is:
>
> Is there any implementation for X100P or TDM400 that supports DTMF
> caller ID WITHOUT the need of polarity inversion before the DTMF
> spill? Is anyone working on this? This is the way it works in Brazil
> and some other coutries...
>
> Thanks a lot,
>
> Renato

Renato,

Bug id=9 in the bugtracker is not currently inserted into CVS but can be
applied to it.

The additional patch for X100P that I put there (srathje) is a quick (ugly)
hack for the Danish CID system that uses the DTMF decoder made by egnarf.

In Denmark we have no warning (officially) before the CID is received but by
monitoring the line I found that a short burst or signal is received just
before the CID so I modified the wcfxo.c code to look for this.

There is a different approach in bug=1719 for monitoring UK BT CID as it
seems to be the same problem, no warning before the CID (V23 FSK) is
received. The method used here is a "history buffer" used to capture the CID
and decode when the first ring is detected.

I guess you will have to investigate both solutions and use what you can to
make a viable solution. If you are successfull in doing so please share with
others via the bugtracker.. :-)

Regards

Soren Rathje