Re: [Ecrit] Cell-id in HELD and other parts of the emergency servicesframework? (Was: WGLC on draft-ietf-ecrit-phonebcp-04 (cell-id))

["Thomson, Martin" <[e-mail]>]

The mobile host could use HELD to get location.  If you consider the
fact that the host may be unaware of the specifics of its connection to
the Internet, it employs the usual methods for determining its location.

As to determining location based on IP in a cellular network, this
question has already been asked ;)  The first step is to determine
MSISDN/IMSI from the network infrastructure.  Apparently this is
expected of SUPL location platforms [1], but the method for solving the
problem is unspecified.  The options that we discovered included AAA
servers (which at the time involved proprietary methods), the GGSN
(another proprietary interface required), or the DHCP server (apparently
there are "conventions" used for filling out certain DHCP headers, and
you can use DHCP lease query now).  I'm not aware of any standard
methods, although I've not been following 3GPP that closely of late.
Alternatively, a LIS could sweep the whole thing under the carpet and
provide an IP address to a GMLC over MLP.  Of course, the GMLC would
have to use the same methods.

If the host is aware of the nature of its network access, it could
provide this information in a HELD request.  I've written an extension
to HELD that enables the use of device-provided information [2].  Cell
identifiers would be a reasonable addition to that document (right now,
I can't remember why I didn't add them in the first place), or a similar
document.  It would not be unreasonable for the LIS to have access to a
cell location database.

Cheers,
Martin

[1]
http://www.openmobilealliance.org/Technical/release_program/supl_v1_0.as
px
[2]
http://tools.ietf.org/html/draft-thomson-geopriv-held-measurements-01

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]]
> Sent: Friday, 7 March 2008 9:48 PM
> To: Thomson, Martin; [e-mail]; [e-mail];
> [e-mail]; [e-mail]; [e-mail]
> Subject: RE: [Ecrit] Cell-id in HELD and other parts of the emergency
> servicesframework? (Was: WGLC on draft-ietf-ecrit-phonebcp-04 (cell-
> id))
> 
> Hi Martin,
> 
> Martin Thomson wrote:
> >One of the main reasons for having a LIS is to provide the
> >translation from network-specific information (cell-ID) into
> >something more global (lat/long, civic).  This means that
> >services can be implemented to use access-agnostic information.
> 
> How do you think this should be done in practice? Would the mobile
host
> use HELD to get its location or reference to it, as usual? How does
the
> mapping from the host's current IP address to one of the location
> formats happen in practice? I suppose the mobile operator would have
> AAA
> server of some sort which can keep track of which IP address is given
> to
> which user, but I don't know the rest of the backend infra well enough
> to understand how the LIS obtains the information it needs to respond
> to
> a HELD query.
> 
> The reason I'm asking is that I'm trying to understand where the Cell-
> id
> would fit wrt. HELD and LIS. In practice, would it be useful for the
> HELD client to tell the LIS what it's Cell-id is directly, or do you
> expect the backend infra to do all the necessary correlations starting
> from the IP address?
> 
> Markus
> 
> 
> >-----Original Message-----
> >From: ext Thomson, Martin [mailto:[e-mail]]
> >Sent: 07 March, 2008 01:17
> >To: Isomaki Markus (Nokia-OCTO/Espoo);
> >[e-mail]; [e-mail]; [e-mail];
> >Bajko Gabor (Nokia-OCTO/MtView); [e-mail]
> >Subject: RE: [Ecrit] Cell-id in HELD and other parts of the
> >emergency servicesframework? (Was: WGLC on
> >draft-ietf-ecrit-phonebcp-04 (cell-id))
> >
> >Extend away, but try to understand where the information is
> >going to be consumed.
> >
> >I am not comfortable with a cell identifier escaping the
> >access network context.  In fact, I'd argue that a cell
> >identifier is not a form of location, but a parameter that can
> >be used to derive location information when combined with
> >contextual information.
> >
> >The carrier that puts the cell in place knows where the cell
> >is, but it is unreasonable to expect others to know this.
> >(3GPP & OMA evidently
> >agree: see Lr interface and the concepts of V-GMLC and V-SLP)
> >
> >To extend the analogy, would you also mandate that LoST
> >servers also have a database of WiFi access points, WiMAX base
> >stations, cable network head-ends, and so forth?
> >
> >One of the main reasons for having a LIS is to provide the
> >translation from network-specific information (cell-ID) into
> >something more global (lat/long, civic).  This means that
> >services can be implemented to use access-agnostic information.
> >
> >I have no objection to someone describing its use, but to
> >expect that a LoST server also have a database that maps
> >cell-ID to location is unreasonable (such a database is
> >necessary for dispatch, even if a routing database also
> >exists).  The two accepted forms for location are geodetic and
> >civic and these should be the only mandatory forms.
> >
> >Cheers,
> >Martin
> >
> > (A question from the side-lines: for UMTS would you use the
> >UTRAN cell identifier, the SAI, or allow for both?)
> >
> >> -----Original Message-----
> >> From: [e-mail] [mailto:[e-mail]]
> >On Behalf
> >> Of [e-mail]
> >> Sent: Friday, 7 March 2008 1:07 AM
> >> To: [e-mail]; [e-mail]; [e-mail];
> >> [e-mail]; [e-mail]
> >> Subject: [Ecrit] Cell-id in HELD and other parts of the emergency
> >> servicesframework? (Was: WGLC on draft-ietf-ecrit-phonebcp-04
(cell-
> >> id))
> >>
> >> Hi,
> >>
> >> Talking about the different scenarios how Cell-id could be
> >used in the
> >> IP emergency service framework, we also still have the
> "conventional"
> >> way of first mapping it into geo or civic location.
> >>
> >> The phonebcp draft currently says:
> >>
> >>    Other forms of
> >>    location representation must be mapped into either a geo or
civic
> >> for
> >>    use in emergency calls.
> >>
> >> So, how do we expect this to happen for Cell-id? One possibility I
> >> would see is that we could extend HELD so that the HELD
> >request could
> >> carry Cell-id, and the HELD server could use the Cell-id in
addition
> >> to IP address to determine the location of the mobile host. (This
is
> >> what Google does already in their map application, so actually the
> >> server doing the mapping does not even have to be associated
> >with the
> >> access network, although such association would probably make the
> >> mapping
> >more
> >> reliable.)
> >>
> >> Would this seem like a sensible extension to HELD?
> >>
> >> Given how ubiquitous Cell-id already is in the cellular wireless
> >> infrastructure, I think we should actually develop protocols that
> >allow
> >> it to be used in at least these ways:
> >>
> >> a) Use it as an input in LoST for mapping to PSAP URI.
> >> b) Use it as an input in HELD for mapping to geo/civic location.
> >> c) Carry it within the signaling protocol for various potential
> >usages.
> >> I think this is already taken care by what is defined in RFC 3455,
> >> although I suppose it would be more inline with the rest of
> >> ECRIT/GEOPRIV work if it was actually carried within PIDF-LO in
> >> accordance to the SIP location conveyance.
> >>
> >> Markus
> >>
> >>
> >> >-----Original Message-----
> >> >From: [e-mail] [mailto:[e-mail]] On
> >> >Behalf Of ext Hannes Tschofenig
> >> >Sent: 06 March, 2008 09:26
> >> >To: Brian Rosen; [e-mail]; Bajko Gabor (Nokia-OCTO/MtView);
> >> >[e-mail]
> >> >Subject: Re: [Ecrit] WGLC on draft-ietf-ecrit-phonebcp-04
(cell-id)
> >> >
> >> >We also discussed this topic in a face-to-face meeting between the
> >> >3GPP and the IETF in July 2006. I recall that we were not
> >so negative
> >> >about the idea of using Cell-IDs.
> >> >
> >> >Furthermore, there are two separate things to consider:
> >> >
> >> >* Cell-ID usage for LoST to obtain a PSAP URI
> >> >* Cell-ID information as information for dispatch.
> >> >
> >> >Gabor is not proposing the second approach.
> >> >
> >> >
> >> >> The basic agreement we have made, in NENA, is that the operator
> >will
> >> >> agree with the PSAP on a location for the cellid.  This
> >could be a
> >> >> point or a polygon.  That location will be reported, in
> >> >PIDF-LO form, from the LIS.
> >> >> There was a discussion about actually allowing a cell-id as an
> >> >> alternate form of location representation, but it was decided
not
> >to
> >> >> do that.  A LIS could, of course, construct a location
> >reference of
> >> >> the form sip:<cell-id>@opco.net, and send that.  When
> >> >dereferenced, it
> >> >> would return the agreed-upon location.
> >> >>
> >> >> Brian
> >> >>
> >> >> > -----Original Message-----
> >> >> > From: [e-mail] [mailto:[e-mail]]
On
> >> >> > Behalf
> >> >> Of
> >> >> > James M. Polk
> >> >> > Sent: Wednesday, March 05, 2008 2:29 PM
> >> >> > To: [e-mail]; [e-mail];
> >> >[e-mail]
> >> >> > Subject: Re: [Ecrit] WGLC on draft-ietf-ecrit-phonebcp-04
> (cell-
> >> id)
> >> >> >
> >> >> > At 01:04 PM 3/5/2008, [e-mail] wrote:
> >> >> > >Cell-id is an identifier of a geographical area which is
> >> >mapped to
> >> >> > >psap already in today's databases. If you'll make an IP based
> >> >> > >emergency call on cellular, that will use now and for
> >> >many years to
> >> >> > >come the cell-id
> >> >> to
> >> >> > >select the psap where to direct your call. GAbor
> >> >> >
> >> >> > let's discuss this topic as its own thread.
> >> >> >
> >> >> > All (not just some or one) of the following questions
> >need to be
> >> >> > answered positively and definitively in order for your Cell-id
> >> >> > scenario to work in IP.
> >> >> >
> >> >> > Where is the Cell-id to PSAP URI mapping done?
> >> >> >
> >> >> > What database has this information?
> >> >> >
> >> >> > What IP protocol is used to resolve this mapping of Cell-id to
> >> >> > (appropriate) PSAP?
> >> >> >
> >> >> > Where is this IP-based server located?
> >> >> >
> >> >> > How can LoST be used to discover this?
> >> >> >
> >> >> > What LCP will give this Cell-id to a phone to include in a
> >> >SIP message?
> >> >> >
> >> >> > What SIP header is this information placed in?
> >> >> >
> >> >> > What form does this Cell-id have in the SIP header from
> >> >the previous
> >> >> > question?
> >> >> >
> >> >> > What IETF document(s) specifies this header and Cell-id form?
> >> >> >
> >> >> >   James
> >> >> >
> >> >> >
> >> >> > _______________________________________________
> >> >> > Ecrit mailing list
> >> >> > [e-mail]
> >> >> > https://www.ietf.org/mailman/listinfo/ecrit
> >> >_______________________________________________
> >> >Ecrit mailing list
> >> >[e-mail]
> >> >https://www.ietf.org/mailman/listinfo/ecrit
> >> >
> >> _______________________________________________
> >> Ecrit mailing list
> >> [e-mail]
> >> https://www.ietf.org/mailman/listinfo/ecrit
> >
> >---------------------------------------------------------------
> >---------------------------------
> >This message is for the designated recipient only and may
> >contain privileged, proprietary, or otherwise private information.
> >If you have received it in error, please notify the sender
> >immediately and delete the original.  Any unauthorized use of
> >this email is prohibited.
> >---------------------------------------------------------------
> >---------------------------------
> >[mf2]
> >
> >

------------------------------------------------------------------------------------------------
This message is for the designated recipient only and may
contain privileged, proprietary, or otherwise private information.  
If you have received it in error, please notify the sender
immediately and delete the original.  Any unauthorized use of
this email is prohibited.
------------------------------------------------------------------------------------------------
[mf2]

_______________________________________________
Ecrit mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/ecrit