Wireless 9-1-1: Phase 1
Implementing Wireless E9-1-1
in the Real World
by Paul D. Linnée
| Paul Linnée is a GIS/GPS and E9-1-1 systems consultant with Geo Comm Corp. in St. Cloud, MN. He has been a police officer, fire fighter and the manager of two E9-1-1 PSAPs (one very large and one medium sized). He is the immediate past president of NENA of Minnesota and is the author of "ICK! I Hate Technical Stuff," which tries to explain all sorts of (icky) technical stuff for public safety "lay" persons. |
| Sidebar: Why We Need Cellular ALI |
| Related Articles: Wireless 9-1-1: Phase 2 |
| Photographs: (select thumbnail to view detail image) |
 With the increased use of cellular telephones, 9-1-1 Centers such as this one in Knox County, Tennessee, are encountering difficulty locating wireless 9-1-1 callers, many of whom may not be able to identify their locations. With the FCC's Phase 1 mandate to identify cellular telephone numbers and cell site locations on 9-1-1 calls having passed in April 1st, a number of technologies have come into play to meet this challenge. Ken Kerr |
 Software programs such as this GeoLynx map display from GeoComm depicts the location of E9-1-1 calls. Landline calls are shown by red arrows. The red block denotes the cell site of the wireless 9-1-1 call, with the yellow shaded area depicting the radio frequency coverage area for that cell site. Rural cell sites like this use Omni-directional antennas whereas a more urban "sectorized" cell site would have coverage areas more like three 120° "pie slices" as opposed to the 360° coverage shown. GeoComm |
 This wireless E9-1-1 solution from XYPoint allows wireless carriers to deliver required location information without installing costly new software or hardware. 9-1-1 data delivery is achieved through existing SS7 connectivity like the North American Cellular Network. See: http://www.xypoint.com XYPoint |
 Dispatcher Shawn Wood in Arlington (TX). The Tarrant County 9-1-1 District was one of the first areas to implement a method of cellular 9-1-1 identification. Each cell site is given a telephone number and an address. That allowed wireless to be selectively routed and, when answered, ANI/ALI data would clearly state the address for the cell site It was an early start in addressing FCC Phase I challenges. courtesy Alisa Simmons/Tarrant County 9-1-1 District |
This article can be found on
page 58 of the May/June 1998
issue of 9-1-1 Magazine.
The problems associated with cellular 9-1-1 - such as locating the caller and routing the call to the appropriate answering point - are well documented. Jurisdictions throughout the US are facing the same problems. In even the most remote and rural areas, there is now almost ubiquitous cellular coverage, and there is all too often no rural addressing in place, which further compounds the problem.
In some ground-breaking work, the Tarrant County, Texas 9-1-1 District implemented a "kind of" fix in the late 1980's. They devised a system where each cell site is given a telephone number and an address. That allowed wireless to be selectively routed (a common E9-1-1 function which routes E9-1-1 calls to a pre-determined PSAP based on the address and phone number in service at that address) and, when answered, ANI/ALI data would clearly state the address for the cell site and what the phone number for the special 9-1-1 phone(s) for that Site was. (Not that the number was of any value, since it was not a number which could be called back). In areas where local (city and county) jurisdictions assumed responsibility for all public safety responses to freeways, this system held promise. However, it still lacked the ability to very closely pin-point the caller's location and to identify the caller's call back number.
As the need to have a better cellular E9-1-1 system began reaching the national press, 9-1-1 interest groups such as APCO and NENA geared up their rather impressive efforts to lobby the FCC, congress or anyone who could help to deal with this problem. For better or worse, the Feds had decreed when cellular service was initiated that providers would not be subject to regulation of the 50 State Utilities Commissions, but only to regulation by the FCC.
Finally, after repeated remonstrations and petitions to the FCC and congress, the FCC opened up a "docket" (94-102) in 1994 to take the matter under study, and, in late `96, published its Report and Order, which required several significant actions:
That by April 1, 1998, wireless telephone service providers must be prepared to send a call dialed to 9-1-1 to an E9-1-1 PSAP with the location of the cell site of origination of that call included in the data presented to the PSAP, and that these providers must be prepared to send that wireless 9-1-1 call to an E9-1-1 PSAP with the 10 digit call back number for the initiating wireless phone as a part of the call
(These requirements had two major caveats attached to them: First, that the E9-1-1 jurisdiction in question must affirmatively request such service, and second, that there must be a cost recovery mechanism in place so that the provider can recover the costs of providing such service(s).)
That by October, 2001, the providers must be prepared to provide information to the E9-1-1 PSAP (in a format not specified and by a technology not identified) that identifies the actual, physical location of said wireless E9-1-1 caller to within 125 meters of accuracy, 67% of the time. (With the same "cost recovery", etc. caveats)
The Order also specified (which was recently reaffirmed) that 9-1-1 calls must be processed even from un-activated wireless phones and that calls from TDD/TTY devices working through wireless phones must also be processed. (This part has recently been amended to provide another year for compliance by digital wireless providers).
As it turns out, NENA, APCO and everyone else who cares succeeded, kind
of. They got the FCC to take action (albeit somewhat after the fact). Now,
the 9-1-1 public safety community (not to mention the wireless community)
must face up to the challenges of putting this "solution" in place.
One observer was recently overheard repeating the old axiom, Be careful
what you ask for.......you might just get it!"
Phase I Implementation
It has been my observation that wholesale implementation of this solution throughout the USA is also taking a "phased approach." By this, I mean that jurisdictions recognize that the technology required to determine the physical location of a wireless caller has yet to be firmly agreed to by the industry, not to mention who will pay to deploy such technologies.
With this recognition, and not knowing in what format and via which means
this caller location will be presented to the E9-1-1 network and its PSAPs,
the 9-1-1 community seems generally to be choosing to implement relatively
simple "Phase I" solutions beginning in 1998. These solutions
come in two basic flavors, as follows:
"Feature Group D"
This approach uses generally existing technologies and capabilities within
wireless "central offices" (MTSO) and networks as well as within
E9-1-1 networks and PSAPs. An MTSO determines which of the many cell sites
it serves received the 9-1-1 call, assigns a "Pseudo ANI" (P/ANI)
to that call, and sends that call over Feature Group D CAMA trunks to an
E9-1-1 Tandem Router serving the area within which the initiating cell site
is located. Then, depending on whether or not the E9-1-1 Tandem Router in
question can or cannot process more than 8 ANI digits (most cannot, they
only deal with the 7 digit ANI number, plus an added 8th digit that specifies
which of up to 4 area codes should be inserted in front of the 7 digit number)
the call back number is included in the initial data stream going to the
PSAP (In cases where the 9-1-1 Tandem can process more than 8 digits) or
it is "stripped off" from the initial Feature Group D data stream
and sent to the 9-1-1 ALI database for transitory insertion into the ALI
record for the cell site location through which the 9-1-1 call was initially
processed. In either case, the cell site of origination is derived by an
ALI "lookup" which uses the P/ANI as the key to find that cell
site's location from the ALI database. (In unaddressed areas, or areas where
cell sites are not located at an addressed location, that text-only data
will be expressed in latitude and longitude coordinates.)
Signal Control Point (SCP )
This competing "Phase 1" approach is somewhat similar, except that it employs a partial SS7 solution involving a Signal Control Point (SCP) for determination of and passage of the calling party's number (CPN) to the E9-1-1 network and PSAP.
There has been some hot debate around the county about which of these
two approaches makes most sense. Those who favor SCP refer to Feature Group
D as "throw away technology" and claim that their approach provides
a "migration path" for the yet-to-be defined caller location data
called for in Phase 2. Those who favor the Feature Group D approach state
that they don't mind paying for a "throw away technology" today
since they don't know what Phase 2 will look like. I recently heard one
9-1-1 expert whose agency will be the one paying for the selected approach
in his state say, "I don't buy 35 MM cameras anymore. If I want to
take a picture, I buy a single use camera, and throw it away when I am done,
rather than investing in some multi thousand dollar camera today that may
be obsolete in a few years." It has also been my observation that the
per-system installation and per-subscriber recurring costs for the Feature
Group D model are quite a bit less than those for the SCP.
Cost Recovery
One of the stickiest issues in all of this has been the not insignificant matter of the "cost recovery mechanism" that needs to be in place before the providers can be required to do any of this in a given jurisdiction. Without going into too much detail here, it needs to be understood that the cost recovery mechanisms used to pay for regular wired 9-1-1 systems are about as many as there are stated that have such mechanisms. However, they generally fall into one of two models. The first model is the "Statewide" model, which has the state collecting a uniform "9-1-1 surcharge fee" on all phone lines in that state. The state then either pays for all or part of the 9-1-1 service (network and equipment) or funnels the state-collected funds back to the cities and/or counties which operate the 9-1-1 systems, or a combination of both. The other popular method is what I call the "County/District Model" wherein the state passes a law permitting counties or special "9-1-1 districts" to impose a surcharge or local tax on phone lines or other activities, with the proceeds funding implementation and on-going costs of 9-1-1 in that area.
The major problem with this cost recovery issue is occurring where the "County/District Model" is in place. Around a dozen states have enacted "Wireless E9-1-1 cost recovery mechanisms" in the past year, and almost all of them have been states which have the "Statewide Model" in place. Wireless carriers are adamant that they can't or don't want to have to deal with surcharges or fees all over the map, depending to which address or zip code the bill gets sent. On the other hand, Counties and 9-1-1 Districts for which these local fees are their primary (and sometimes only) funding lifeblood, don't want to lose these fees to their State nor have to rely on their State to funnel them back to the counties or 9-1-1 districts. It does appear as if some resolution to this issue is developing, even in states where the "County/District Model" is in place. Specifically, I have seen cases wherein the local entities are agreeing to a uniform statewide fee being imposed on wireless phones and the proceeds either being remitted to the local jurisdiction based on where the subscriber lives and the local jurisdiction being responsible to contract for the Wireless E9-1-1 service, or the statewide uniform fee being remitted to the State and the State assuming the costs of contracting for the Wireless E9-1-1 service, on behalf of the Counties/Districts.
Most areas seem to be implementing Phase I with Feature Group D. By doing this, they are able to fully utilize existing E9-1-1 network, router and PSAP equipment and components. The wireless E9-1-1 calls come to the PSAP over the same dedicated E9-1-1 trunks that wired calls come in on now. They are selectively routed to the "appropriate PSAP" (more on this later) using the P/ANI assigned to the cell site based on its address or location. And they are answered on existing PSAP telephone equipment with the text only data displaying on existing PSAP ANI/ALI display equipment.
There appear to be two major issues in this Phase I implementation. They are:
- The determination of which PSAP is "most appropriate" for
E9-1-1 calls from a given cell site.
- The problem in requiring a 9-1-1 dispatcher to figure out where a cell
site is located and what land mass is covered by its radio signal base
on the text only location data that will be delivered.
As it relates to issue #1, the problems are several:
Which E9-1-1 PSAP jurisdiction(s) are covered by a given cell site's RF pattern?
If multiple PSAP jurisdictions (two cities, for example), to which of the two should the 9-1-1 calls be routed? What if funding from whatever "surcharge" is in place is dependent on 9-1-1 call volumes? What if city vs. county or vs. state police "politics" enter the picture?
How can one get good cell site RF coverage maps to make these decisions from wireless providers who often treat this as "privileged and confidential" data and don't want to share it with anyone, much less a public agency which may be required to "make public" any data it receives or possesses?
What needs to be in place in the area's E9-1-1 network to permit wireless E9-1-1 calls received at one PSAP (the one decided to be "most appropriate" will not necessarily always be the "correct" PSAP) to be transferred to another E9-1-1 PSAP and to have the P/ANI, cell site location and 10 digit CPN transferred with the call? What if there are multiple and "incompatible" E9-1-1 network Tandem Routers in place within a region, each serving some of the E9-1-1 PSAPs in that region? What if there are multiple and generally incompatible E9-1-1 ALI data providers within a region?
The answers to these issues are as many and varied as the places where they are being asked. All of them are present, to some degree, in all E9-1-1 areas where Phase I implementation is underway.
It is my final opinion that regardless of what political or technological solutions are found to these issues, several lasting facts remain:
That people want and will buy wireless phones.
That the use of wireless phones (including the new 2 GHz "PCS" phones) will grow nearly exponentially.
That the occurrence of wireless 9-1-1 calls will grow almost as fast.
That we will see, with increasing frequency, cases where people no longer have wired phones or phone lines in their residences and all calls they will make will be wireless, including 9-1-1 calls.
That such a development will threaten E9-1-1 with address based "selective routing" as we know it today, and that selective routing decisions will someday have to be made on the basis of the caller's "X and Y" location and not an address.
That we will need to implement intelligent GIS mapping systems to depict the location of incoming wireless E9-1-1 calls for dispatchers, since many will not come with an address, per se.
That such systems as CAD will soon have to be capable of referencing incident locations by X/Y (and probably "Z" to reflect verticality, as in "the 14th floor of a building") instead of or in addition to conventional addresses.
We did ask for this "improvement," didn't we? Really now, didn't we? Well, I guess we got it and now we have to deal with it.
