E911 Emergency Network Myths

Sattel Technologies

A Leader in Aerospace Technology

The Myths of E911 Cellular Location Services

People who know the limitations of the E911 network as an emergency service
should be reluctant to rely upon it for protecting human life.

Since 1995 the FCC has promoted cell phones as the primary means for locating persons in emergencies. However, adding location services to cell phones has been a daunting task for the cellular industry. After a decade of development and long delays, results have been unsatisfactory. Some fundamental technical reasons for this are summarized below.

GPS satellites work very poorly or not at all in the places where most people spend most of their time; indoors, and outdoors under trees or in urban canyons. GPS requires a clear view of the sky to lock onto and track satellites.
Cellular tower triangulation has disadvantages because: 1) cellular towers are designed to cover small areas for maximum frequency re-use, and 2) cellular signals penetrate man-made structures relatively poorly. Thus it is difficult to receive signals from multiple cellular towers at any given indoor location.
GPS and cellular signals at frequencies in the range of 800 to 1,900 MHz are weak to begin with, then are attenuated by passing through man-made objects. Also, at these frequencies signals reflect off surfaces and split, arriving via multiple indirect paths called “multi-path” reception that distorts location measurements.

There also are important non-technical problems with using cell phones for emergency locating services, including:

Operating a cell phone may be too difficult for children and infirm adults, especially when under stress, and is impossible for incapacitated persons.
Cell phones must be large enough to contain a display and keypad, and so are not concealable. In a kidnapping, a cell phone would be confiscated.

Location information is received by network operators, making it accessible to everyone including the government and stalkers.

Mobile telephone networks are perceived to have "wide-area" coverage. Developers and the FCC have been attempting to exploit cellular networks for other applications such as location measurement. Some cellular networks have started to offer the locations of their devices as part of an effort to expand beyond basic telephony services. Separately, the Global Positioning System (GPS) was designed for maximum geographic coverage area. At first glance, it appears logical to combine GPS with cellular networks.

Given these facts, why are cellular telephones with GPS plagued with problems when used as locating devices? The answer is that neither cellular phone location nor GPS provides: 1) adequate coverage measured by percent of time that location measurements are available in people’s daily lives, and 2) a low barrier to entry for users and developers.

GPS works "world-wide", can be added to existing devices (at significant cost), and provides accuracy of 10 meters or better. But what does "world-wide" mean? The fact is that GPS fails to provide the coverage needed for emergency location services because GPS receivers, although capable of operating everywhere on the earth's surface, have poor coverage measured by the percent of time they work where people spend most of their time.

GPS requires a clear view of the sky: thus it does not work indoors and performs poorly in “urban canyons” formed by buildings. This limited availability severely restricts GPS' applications. GPS has been useful for navigation and search and rescue applications which are primarily outdoor activities. Most persons, however, spend most of their day and night indoors.

E911 Phase II legislation in the United States requires cell phone companies to be able to locate handsets within 150 meters by Dec 31, 2005. Cellular networks track their devices using tower triangulation and assisted GPS techniques that employ network servers to "assist" the GPS chips placed in the devices. AT&T Wireless’ Friend Finder location service is an example: it works where normal cell phones do, but it does not provide a low barrier to entry. The reason is that in cellular provider-based location the information is owned by the service provider rather than the client. Users have to use programming interfaces and endure pay-per-use costs imposed on location information. Also, the user has no choice but to trust the cell phone provider with his or her location information since their location is tracked by the provider whether or not the user is making use of location services. Finally, provider-driven location works only on cell phones and not other devices like notebooks and PDAs.

Independent tests have shown the inadequacies of cellular/GPS for emergency location measurement.

New technology has been shown to be superior in terms of usability in peoples' daily lives: that is, in places where most people spend most of their time. The patent-pending SafeLink system recently announced uses existing local transmitters as position references instead of weak cellular tower and GPS signals. Such transmitters are ubiquitous in all metropolitan areas and are designed for excellent building penetration. This technology is independent of the cellular networks, avoids per-use charges, and is inexpensive to implement.

To quantify the coverage and measurement accuracy of beacon-based location systems, and to compare them with GPS coverage, a test was conducted in Washington State in October 2004 by an independent research group. Users were outfitted with small devices which they carried during a typical day capable of monitoring a representative sample of local transmitters: GPS, GSM and 802.11. Second, both 802.11 transmitter density and corresponding location measurement accuracy were measured in an urban, a residential and a suburban area. Both the user-time experiment and the density experiment showed nearly ubiquitous GSM and 802.11 coverage, the density of which correlated with population density. Accuracy results showed that with sufficient coverage density, 802.11 beacons alone can provide median accuracy of around 20 meters while GSM beacons alone provide accuracy of 100-200 meters.

In the Washington State test the researchers defined the “coverage” of location measurement technology as the percentage of time that the technology can produce a new estimate of the user’s position. If a GPS device, for example, has a satellite lock for 15 minutes then loses its lock for the next 30 minutes and gets it back for 15 minutes, the coverage for that period of time would be 50% with an average gap of 30 minutes. Note that the definition of coverage is based on the user’s time, not on geographic area.

For the experiment, measuring beacon density and its effect on accuracy, the researchers gathered GPS, 802.11 and GSM signal data from three areas:

Urban – a mix of commercial and residential urban high-rises
Residential – a medium-density residential neighborhood
Suburban – a sparse suburb of single-family homes

For each area, the researchers drove around for sixty minutes with a laptop, a GPS unit, and a Nokia 6600. 802.11 scans were performed four times per second, and both GPS and GSM readings were taken approximately once per second. At all times the researchers tried to navigate within areas in which GPS lock would not be lost as GPS provided the reference location to be used to estimate beacon positions and measurement accuracy. The results of the user-time coverage experiment are shown in the following table:

User-Time Coverage of Measured Beacons
Test Subject	GPS		GSM		802.11
	Coverage	Avg. Gap	Coverage	Avg. Gap	Coverage	Avg. Gap
Person A	12.8%	68 min	100%	-	87.7%	1.6 min
Person B	0.6%	78 min	98.7%	2 min	95.8%	1 min
Person C	0%	171 min	100%	-	100%	-
Average	4.5%	105 min	99.6%	1 min	94.5%	1.3 min

The coverage of GPS matched expectations. It has poor user-time coverage and long gaps because satellites are cut off indoors, in vehicles, in urban canyons and under cover where many people spend the majority of their time. However, the test subjects saw nearly seamless GSM coverage. Even in places where the signal level was too low to make an actual phone call, for example in an elevator or basement, it was still often possible to see GSM beacons. The measured 802.11 coverage was slightly lower than GSM but with similar gap sizes.

The Washington State tests did not include measurement of broadcasts, paging services or other signals because these are well known to have excellent coverage within their service areas, indoors and out. These high-power transmissions can be received in some of the most difficult locations. With well over 20,000 such transmitters licensed in the United States alone, their density in every significantly populated area is high. Including them in the long list of available beacons for location measurement further increases beacon density and expected location measurement accuracy.

The test data supports the claim that beacon-based location technology such as SafeLink has the potential to provide user-time coverage which significantly exceeds that of GPS based on just GSM and 802.11 beacons alone.

SafeLink will work far better than cellular/GPS in the places
where most people spend most of their time on a daily basis.

The location technology used by SafeLink is independent of cellular networks and GPS. SafeLink instead uses existing local transmitters at known locations, which are ubiquitous in urban and suburban areas worldwide. Commercial broadcasts, paging systems, trunk radio systems, cellular towers, 802.11 transmitters and others are all examples of potential location measurement reference signals. The coverage and accuracy of SafeLink's technology depend on the number and mix of such signals in the local area. Obviously there is correlation of transmitter density to population density, and location measurement therefore works better in urban centers than in less populated or rural areas. This is the opposite of GPS/cellular, which does not work indoors or in urban canyons.

Location measurement technologies on which human life and safety depend require near-100% availability of location information in people’s daily lives, just as cellular networks are held to a 100% availability standard. SafeLink's technology is capable of achieving this level of performance as it:

maximizes coverage in terms of where people spend their time on a daily basis
offers a low barrier to entry for users and application developers
is protective of users' privacy

The test referenced above confirmed that GPS, often thought of as a pervasive location technology, in fact lacks availability in people’s daily lives since people are frequently indoors, in vehicles or under cover. On the other hand, signal sources such as 802.11 and GSM, high-power paging and commercial broadcast services, are almost always available where people routinely spend their time, indoors and out.

Specifically, for 802.11 beacons alone, the test showed that if density is high enough for devices to see at least 4 or 5 beacons during a 10 second window, beacon-based locator technology can achieve median accuracy of around 20 meters. This accuracy is slightly lower than GPS but, unlike GPS, beacon-based location covers nearly 100% of users’ daily lives.

Of course, more beacons are available when one takes into account high-power wide-area-coverage broadcast transmitters, of which there are more than 20,000 licensed in the United States alone. These transmitters operate at high power, cover large metropolitan areas, and are designed to serve the (primarily indoors) areas where most people spend most of their time.