Meeting the wireless 911 mandate

Wireless phone use has increased dramatically in the past couple of years. A significant number of new users are looking to cellular for increased security away from home. In fact, according to Cellular Telecommunications Industry Association statistics, some 41 million U.S. wireless communication subscribers are making more than 18 million 911 and other emergency calls each year.

Unlike traditional wireline 911, wireless 911 does not provide emergency personnel with either a caller's location or identification-such as a call-back number. This lack of information can delay emergency response and also significantly increase the workload for emergency personnel.

National and state emergency services organizations, the Federal Communications Commission and industry organizations have come together to address this problem and provide a wireless 911 service that is at least equivalent to wireline enhanced 911 service. The FCC recently announced a report and order for 911 that defines wireless 911 requirements, some of which must be implemented within 18 months.

The challenge for service providers will be to deploy an architecture that meets the FCC requirements and provides a smooth integration of wireless and wireline systems while protecting the emergency services' substantial investment in enhanced 911 network and customer premises equipment. It is possible to use off-switch home location register (HLR) and service control point (SCP) technologies to meet and exceed initial requirements, while at the same time provide a logical migration to more sophisticated location technologies as they become available.

Enhanced 911 Available to 85% of the U. S. population, enhanced 911 is a tremendously successful wireline communications service. Its success is due to its ability to provide consumers and 911 public safety answering positions (PSAPs) with an important set of features that are critical in an emergency situation:

Calling party number-through automatic number identification (ANI) Calling party location-through automatic location identification (ALI) Selective routing to provide automatic routing to the correct primary PSAP Selective transfer to automatically re-route the call to another PSAP-such as from police to fire Features such as support for telecommunication devices for the hearing impaired.

Some of the major enhanced 911 features depend upon the pre-provisioning of data to network elements (Figure 1). This architectural limitation is part of the challenge for wireless 911. Another challenge in implementing a wireless 911 service is that the signaling interface to the 911 selective routing function is through centralized automatic message accounting (CAMA) and is usually limited to delivering 1+7 digits.

The ball was set in motion in June 1994 when the emergency service organizations and PCIA issued an emergency access position paper that was filed with the FCC as part of the personal communication services proceedings. This position paper helped spur the FCC to action with its November 1994 notice of proposed rulemaking (docket 94-102).

The FCC recently issued a report and order for docket 94-102 that defines wireless 911 requirements and mandates implementation of Phase I within 18 months and Phase II within five years. The primary difference between Phase I and Phase II is the degree of caller location accuracy. Phase I requires accuracy to the cell/sector level, while Phase II requires accuracy within 125 meters 67% of the time.

Concurrent with the report and order, the FCC announced a further notice of proposed rulemaking. In that notice, the FCC indicated it would consider additional requirements for wireless 911 such as even more accurate location information (within 40 feet, including altitude); time constraints on providing location; and updated location information during the call.

It seems that the FCC is diligently and aggressively protecting the country's 911 service and plans to continue this in the foreseeable future.

In the early versions of cellular, interoperability between vendor equipment was unheard of. As the industry continued its incredible growth, roaming between systems began to take on a sizable portion of the traffic. With it came three issues: validation, handoff between border systems to assure call continuity, and call delivery.

These issues led to the formation of a standards body for creating an interoperability protocol, which would later be known as IS-41. The IS-41 protocol is used with analog cellular networks and with digital wireless networks supporting time division multiple access and code division multiple access.

Instead of IS-41, networks supporting the GSM Pan-European digital wireless standard use the mobile application part (MAP) protocol. Because GSM had no legacy issues to deal with, it started as a far more sophisticated standard than IS-41. However, the two are slowly converging-more so in functionality than in physical realization.

These standards have defined a reference architecture for the elements of a wireless network. Note that this is a functional rather than a physical model, implying that multiple functions could be combined on the same platform. Key mobility interfaces of this architecture have been or are in the process of being standardized, paving the way for off-switch, or adjunct, solutions in wireless.

As networks have matured, the focus has extended beyond simply the issues associated with mobility management and has moved into the realm of services and service creation. This is the purview of the wireless intelligent network (WIN) and its GSM counterpart, customized applications for mobile network enhanced logic (CAMEL).

Each of these service strategies builds upon IS-41 and MAP, introducing appropriate new elements.

The fundamental focus is the creation of a means of interactive call control between switching entities and adjuncts. WIN and CAMEL both define a series of call triggers, or points of interaction. The triggers occur at predefined points in a call flow, allowing service alteration at these points. The triggers themselves may be routed to a number of adjunct platforms that instruct the switch on how to complete the call, thereby offering a specific service. It is these triggers that offer multivendor service solutions for carriers going forward.

On or off-switch? Although Phase I of wireless 911 could be implemented as an embedded switch function, wireless 911 Phase II may require an off-switch solution using WIN and CAMEL call triggers. An off-switch solution provides a low-cost, standards-based service that will grow in capability as 911 services evolve.

The fundamental operational needs to support Phase I and II 911 include:

A mapping and geolocation database for PSAP call routing Geographical boundary provisioning Out-of-band signaling for real-time location update A large processing capacity for location calculation Interactive call control A network-centric solution, shared across switching platforms High reliability and manageability.

The intensive processing demands, as well as the need for large database support and management, are best served by an adjunct computer-based solution. Additionally, location technology offers myriad vertical service opportunities beyond wireless 911, which could help share the anticipated Phase II cost burden.

Potential services include location-sensitive billing, fleet management and a mobile Yellow Pages application that would enable a mobile user to obtain directions to a commercial establishment, tailored to the user's current location. Ideally, a wireless 911 implementation should support these other service opportunities in a timely fashion.

The architecture for an adjunct-based wireless 911 service takes advantage of existing triggers to introduce special processing into the 911 call (Figure 2). The architecture also provides seamless integration with the existing enhanced 911 system by providing data to the PSAP via the 911 ALI database.

The mobile switching center (MSC) forwards the 911 call to the wireless 911 SCP as it would in a basic number translation scenario. The SCP performs number translation, either using a simple cell/sector-to-PSAP lookup for Phase I or perhaps a more complex latitude and longitude to PSAP geographic translation for Phase II. The wireless 911 SCP can send additional information about the call, such as a callback number or detailed location information, to the ALI database. The ALI database stores this dynamic call information in preparation for the data request by the PSAP.

The HLR represents the only off-switch element that has been deployed in any significant quantity in wireless networks to date. Its primary use has been as a centralized subscriber database. For carriers that have deployed HLRs, the wireless 911 solution should run as an additional application on the HLR, leveraging the existing investment.

For carriers that have not invested in an HLR, a small adjunct, or SCP, could be offered. This would use the same hardware platform as the HLR only on a smaller scale, thus allowing a migration path to an HLR if desired. This strategy mitigates the cost and operations, administration and maintenance effect for carriers deploying their initial adjunct, while providing maximum leverage for carriers that have already deployed adjuncts in the form of HLRs.

Maintaining flexibility The adjunct architecture has a number of important benefits for wireless carriers, wireline carriers and the PSAP community. The architecture can be implemented with minimal changes to existing networks: Only the wireless 911 SCP and the ALI systems need to change. At the same time, the architecture provides for many flexible deployment options. The SCP could be hosted on an off-switch HLR platform or as a stand-alone SCP.

The SCP could be provided as a common service to multiple wireless networks. The interfaces between the MSC, SCP and ALI system could be standards-based or proprietary. The signaling interface between the MSC and the selective router could be the existing CAMA signaling, feature group D signaling or ISDN user part. The selective router and the ALI systems could be combined into a single system.

There are few standards for existing enhanced 911 services. In the past, wireline equipment vendors defined de facto enhanced 911 standards simply by building the product. Recently, local telephone companies have worked together with the industry to build on these standards. The Telecommunications Industry Association's TR45.2 committee has formed an ad hoc emergency services committee to define standards as they work to meet the FCC ruling.

An off-switch architecture offers a wireless 911 solution that is switch vendor-independent and protects the existing PSAP investment. An adjunct solution meets Phase I requirements and can provide an easy migration to Phase II and further requirements as they may develop.

Daniel Tell is a Senior Member of Technical Staff at Motorola Cellular Infrastructure Group, Arlington Heights, Ill. David Hose is President of SignalSoft Corp., Boulder, Colo.