[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Ecrit] emergency call termination
This is a complex issue, made more complex by current deployments.
First of all, we need to distinguish between a call that is complete, where
a BYE ends the call, and a call that is not complete where a CANCEL ends the
call. -phonebcp discusses a completed call where BYE would be used to
terminate. It is currently silent on uncompleted calls.
Your example is an uncompleted call, but you can extend it to include the
case where the caller hangs up just as the call taker answers. Both of
these are considered "abandoned calls". In many jurisdictions, they respond
to an abandoned call. They don't know why the call is abandoned. So, most
PSAPs I've talked to want to force the call to complete, and have the call
taker hear (or read) that the problem is solved, and they don't need to send
a cop to check it out. This is one reason why they want this capability.
The basic CPH is for another purpose, which is pre-mature hang up. It
happens often enough that PSAPs care. The user, who is stressed, hangs up
before the call taker has enough information. They want the call to be
reconnected if the user does that and subsequently goes on hook. They want
to be able to ring and/or send howler tone to alert the user to get back on
the d&*m phone.
It used to be that 9-1-1 in the U.S. had these features. In Canada, they
still do. In the U.S. they got lost in the ISDN signaling change, much to
the regret of the PSAPs. ISDN (and ISUP) didn't retain the feature, and, I
am told, the PSAPs howled, but it didn't get fixed. Now, some of them are
used to not having it, but many wish they could get it back. The Canadians
have it still, and don't want to lose it.
It is clear however, that there are circumstances where CPH is not desired.
The best example is where you have a mass calling event and some calls are
answered on an IVR (or "IMR" -- multimedia). Clearly, you want the caller
to be able to hang up.
Now, Guy Caron has proposed several more features, one of which is the
ability to signal the hook state of the phone. If we had that, then the
PSAP could return BYE automatically upon receipt of an on-hook state. That
would allow the PSAP to control CPH without a negotiation of features at the
phone.
So, I propose that we retain the prohibition on BYE, and we add the proposed
hook state signaling (sending SDP with inactive). PSAPs who don't want CPH
should send BYE upon receipt of inactive. Note that -phonebcp says hold
should be disabled.
Brian
> -----Original Message-----
> From: ecrit-bounces at ietf.org [mailto:ecrit-bounces at ietf.org] On Behalf Of
> Ted Hardie
> Sent: Thursday, July 10, 2008 12:22 PM
> To: g.caron at bell.ca; ecrit at ietf.org
> Cc: nena-i2_5 at listserv.neustar.biz
> Subject: Re: [Ecrit] emergency call termination
>
> At 12:31 PM -0700 7/7/08, g.caron at bell.ca wrote:
> >Dear ECRIT,
> >
> >NENA wishes to provide comments regarding "emergency call termination"
> >behavior in -framework-05 and -phonebcp-04.
> >
> >The UA behavior currently defined disallows a UA engaged in an emergency
> >communication to release the call and states that only the PSAP should
> >be able to terminate an emergency call.
>
> Sorry for the late comment on this, but having thought about this a bit,
> I agree that this should not be mandatory. If a caller realizes after
> starting the call that it is not an emergency (the person choking has
> their airway clear; the parent realizes a child has dialed 911, whatever),
> they should be able to terminate the call.
>
> >NENA wishes to express that this behavior, while desired or mandated in
> >certain jurisdictions, is undesirable in others. For example, the
> >majority of the U.S. jurisdictions do not use this feature and prefer to
> >allow the caller to release an emergency call.
> >
> >It is not NENA's intent to revert the current behavior in such a way
> >that the caller has always the ability to release an emergency call.
> >NENA would like however that this feature be made optional and
> >controlled at the PSAP end. This would allow a roaming device to adjust
> >its behavior according to the rules of the jurisdiction it is currently
> >in.
>
> Though I am no expert in this, it's my understanding that this is
> not behavior supported by the current cellular networks, so I
> believe that it would be both simpler and more appropriate to
> define user terminated calls as permitted behavior.
> regards,
> Ted
>
>
>
>
>
> >Thanks,
> >
> >
> >Guy Caron
> >On behalf of NENA
> >_______________________________________________
> >Ecrit mailing list
> >Ecrit at ietf.org
> >https://www.ietf.org/mailman/listinfo/ecrit
>
> _______________________________________________
> Ecrit mailing list
> Ecrit at ietf.org
> https://www.ietf.org/mailman/listinfo/ecrit
_______________________________________________
Ecrit mailing list
Ecrit at ietf.org
https://www.ietf.org/mailman/listinfo/ecrit