[Asterisk-Users] RE:911, networks of * servers, etc. (was: VOIP Dialtone?)

Tom Zimnicki asterisk at zimnicki.org
Thu Aug 21 21:50:58 MST 2003


>
> This is excellent data; thank you.  I'll review before harrassing the
> E911 folks at VON.
>
> However, this only seems to solve problems for local PSAP
> connectivity to your * server.  Or am I mistaken?
>

You are correct. My need is for a low-cost solution for bridging a nex-gen
service with a traditional local requirement.

But there are two parts to the issue:

Asterisk to Network Interface & Asterisk to User Interface

While the *NI may be particular to my environment (and that being particular
to Canada at the moment. Quick question - Anyone aware of other
jurisdictions where FG-C is used instead of FG-D or CAMA for 911?) the *UI
would be useful to all.

> I think a larger and simpler (read: less hardware and no "custom"
> circuits) system needs to be developed to hand information to the
> PSAP.  I know that none of my customers would ever, EVER consider
> purchasing a dedicated circuit into the 911 system until they were
> forced to by some sort of laws.  With VoIP providers all over the
> nation, I think this is an unworkable solution.  I may be whistling
> in the wind, but that's the reality of the situation.
>

The thing here is if you are trying to do genuine local dialtone
replacement, I think the regulatory agencies would not see much of a
difference between VoIP & POTS with regards to 911.

If what you are doing is second line, foreign exchange service, etc...then
they would be more apt to treat it like they treat traditional PBX's

I think the answer to VoIP 911 will ultimately be based on what happens with
E911 for Wireless (substitute wireless location schemes for finding handsets
with manual user intervention or some sort of simple location protocol built
into the phone or soft client or GPS/Triangulation for Wi-Fi).

In the meantime though, I don't think developing 911 or E911 for *NI would
hurt (and the E part would get more interest for getting SS7 working with *)
and getting the *UI part going will need to be done eventually anyway, so
why not start looking at it now when the standards are still being developed
and * is still growing up.


TZ

> JT
>
>
>
> At 7:15 PM -0400 8/21/03, Tom Zimnicki wrote:
> >
> >Which brings me to an application at hand:
> >
> >We currently have an * box connected to a SIP Media Gateway which is
> >connected to the PSTN via SS7.
> >
> >We have MF FG-C 911 trunks connected to a DMS and can bring an MF T-1
into
> >the * box if we buy a T400P card.
> >
> >
> >The question is how do you support ringback with SIP devices? (i.e. the
PSAP
> >operator keeps the 'trunk' siezed and forces ringing on the SIP phone
that
> >is on-hook. It should also generate a 'howler' tone on an off-hook set)
> >
> >(There was a list for SIP 911 issues, but it appears to be dormant:
> >http://lists.cs.columbia.edu/mailman/listinfo/911 )
> >
> >There was a reference to using re-invites and suspend/resume via RFC 3064
> >and that T1.628 - 2000 might have been looking into it. Does anyone have
a
> >current copy they could check to see if this issue has been addressed?
> >
> >(Also GR-529
>
>http://telecom-info.telcordia.com/site-cgi/ido/docs.cgi?ID=159801680D000801
&KEYWORDS=529&TITLE=&DOCUMENT=&DATE=&CLASS=&COUNT=1000
> >
> >&
> >
> >ANSI T1.628a-2001. http://www-comm.itsi.disa.mil/t1/628.html will
probably
> >be of interest.)
> >
> >Does anyone here have the interest or experience to try and make this
happen
> >on * ?
> >
> >(And although the world is moving to E911 and there are folks working on
SS7
> >for *, is anyone interested in getting the Digium cards and * to
intereface
> >with the Emergency Service Feature Group C trunks that a Nortel 911-DMS
> >Tandem uses? )
> >
> >Regards,
> >
> >Tom Zimnicki
> >
> >----- Original Message -----
> >From: "Michael Kane" <mkane at to-talk.com>
> >To: <asterisk-users at lists.digium.com>
> >Sent: August 21, 2003 3:25 PM
> >Subject: Re: [Asterisk-Users] RE:911, networks of * servers, etc. (was:
VOIP
> >Dialtone?)
> >
> >
> >>  There are TDM interfaces higher end PBX's use to interconnect to the
> >PS/ALI.
> >>  I beleive it's a CAMA trunk that signals using MF.
> >>
> >>
> >>
> >>  ----- Original Message -----
> >>  From: "John Todd" <jtodd at loligo.com>
> >>  To: <asterisk-users at lists.digium.com>
> >>  Sent: Thursday, August 21, 2003 3:00 PM
> >>  Subject: [Asterisk-Users] RE:911, networks of * servers, etc. (was:
VOIP
> >>  Dialtone?)
> >>
> >>
> >>  >
> >>  > Yes, I'm familiar with the E911 platforms and their requirements to
> >>  > some degree.  The trick is that the people running Asterisk PBX
> >>  > systems have no visibility into SS7, and that is an unreasonable
> >>  > expectation, so some other out-of-band method for moving caller
> >>  > location to the PSAP is required.
> >>  >
> >>  > As far as geographic location tracking is concerned: that is the
> >>  > user's problem.  If they don't have the correct information in their
> >>  > device, then they're SOL.  There is _no way_ to develop lat/lon/alt
> >>  > coordinates from an IP address, despite what any .com
> >>  > flash-in-the-pan company says they can do with their clever
> >>  > databases.  Thus, the PBX/switch provider will have to enforce their
> >  > > own database of device-to-geographic-coordinates.  (As mentioned,
> >  > > maybe a SIP header is a reasonable thing to use for the UA to relay
> >  > > this data to the proxy.)  I am not concerned so much about the
> >>  > ability of the devices to send their data to the proxy: I am VERY
> >>  > concerned about how the proxy then looks up the appropriate PSAP,
and
> >>  > then relays the data for the call to that PSAP.
> >>  >
> >>  > JT
> >>  >
> >>  >
> >>  >
> >>  >
> >>  > >911 through the phone system is tricky business. e911 which is the
> >  > > >automated process of handing the address to the 911 center uses
the SS7
> >>  > >database to do it's work (the database is created when the LEC runs
> >>  > >physical lines to locations not by people filling anything out).
Cell
> >>  phone
> >>  > >service providers have the simuliar problems as VoIP service
providers
> >>  are
> >>  > >facing are realizing with call forwarding and call following it
will
> >get
> >>  > >worse.. Congress has mandated that the cell phone industry make it
> >>  possible
> >>  > >to track a cell phone users within 300yards via cell sites and
> >>  > >triangulation. By 2005 every cell phone will be required to have a
GPS
> >>  and
> >>  > >send GPS information to the 911 system when they call 911. If you
want
> >>  more
> >>  > >information on e911 try http://www.fcc.gov/911/enhanced/ . As the
cell
> >>  > >phone industry grows there will be a need for a national 911 call
> >routing
> >>  > >center. I bet it won't be free.
> >>  > >
> >>  > >
> >>  > >Original Message:
> >>  > >-----------------
> >>  > >From: John Todd jtodd at loligo.com
> >>  > >Date: Thu, 21 Aug 2003 01:32:24 -0700
> >>  > >To: asterisk-users at lists.digium.com
> >>  > >Subject: [Asterisk-Users] 911, networks of * servers, etc. (was:
VOIP
> >>  > >Dialtone?)
> >>  > >
> >>  > >
> >>  > >
> >>  > >OK, that "VOIP dialtone?" thread was getting really out of hand, so
> >>  > >I'll condense my answers into one big ugly message:
> >>  > >
> >>  > >
> >>  > >1) 911 service.  Yes, that is one of three reasons to keep your
PSTN
> >>  > >line.  The other two reasons are:   Inbound calls from local
callers
> >>  > >still should work on a POTS line, for now.  You can't find VOIP
> >>  > >providers in most area codes, so you'll most likely need to have a
> >>  > >"local" number that finds it's way to you for "local" tasks.
> >>  > >Secondly, the Internet is not as reliable as the phone system.
> >>  > >Sorry, folks, it just works that way right now despite what your
> >>  > >network engineer might tell you.  That's not to say it's
unreliable,
> >>  > >but those last two nines are very expensive... Besides, any good
> >>  > >network engineer will tell you that you should have multiple paths
> >>  > >for your IP connectivity.  With few exceptions, most homes do not
> >>  > >have multipath connectivity.  (note: businesses may in fact have
> >>  > >better uptime on their IP network than their phone network, if they
> >>  > >have competent engineers and a reasonable budget.)
> >>  > >
> >>  > >1.5) There are reasonable technical solutions to this problem, but
> >>  > >for the life of me I can't figure out why the 911 centers haven't
> >>  > >gotten their act together and solved this.  There are two halves to
> >>  > >this problem: "What PSAP do I call? (and what phone number)"  and
> >>  > >"How do I get my location data to the PSAP once I call them?"
> >>  > >C'mon, this is not difficult.  The first question can be answered
> >>  > >trivially: there _must_ be a database of address-to-PSAP mappings.
> >>  > >Any PBX administrator (or SIP phone owner, for that matter) should
be
> >>  > >able to figure out their address.  Methods for associating the PSAP
> >>  > >number with the phone are numerous, and trivially implemented - if
> >>  > >people don't keep their address information updated, they're SOL
> >>  > >(though you can remind them in an automated fashion to keep it
> >>  > >updated - just forbid them from using the service unless they
verify
> >>  > >the address every month or so.)
> >>  > >
> >>  > >The second question is more difficult, but certainly possible.
There
> >>  > >may be kludge ways of doing it, and there should be more elegant
ways
> >>  > >of doing it.  A SIP header with lat/lon/alt data that gets sent
from
> >>  > >the UA only on 911 (or other programmable string) calls might be
> >>  > >reasonably elegant... maybe.  But that only gets the data to the
SIP
> >>  > >proxy.  That doesn't solve the issue of how you get that data from
> >>  > >the SIP proxy to the PSAP, which at some point will be almost
> >>  > >certainly through a PSTN connection... ADSI FSK, maybe?  Ugly, and
> >>  > >PSAPs would not want to invest in equipment.  A national caller-id
to
> >>  > >location clearinghouse in which your proxy could participate (any
911
> >>  > >calls would create a temporary mapping)?  Maybe, but probably not.
> >>  > >Non-standard, and I doubt PSAP operators would want another tool,
> >  > > >even if it is web-based and so easy a monkey could use it.  I
don't
> >>  > >know.  I guess I'll grill the PSAP people at the panel next month
at
> >>  > >VON.  :-)
> >>  > >
> >>  > [snip]
> >>  > _______________________________________________
> >>  > Asterisk-Users mailing list
> >>  > Asterisk-Users at lists.digium.com
> >>  > http://lists.digium.com/mailman/listinfo/asterisk-users
> >>  >
> >>
> >>  _______________________________________________
> >>  Asterisk-Users mailing list
> >>  Asterisk-Users at lists.digium.com
> >>  http://lists.digium.com/mailman/listinfo/asterisk-users
> >>
> >
> >_______________________________________________
> >Asterisk-Users mailing list
> >Asterisk-Users at lists.digium.com
> >http://lists.digium.com/mailman/listinfo/asterisk-users
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users at lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
>




More information about the asterisk-users mailing list