Location-Based Services Challenge Deployment

Market forecasts for LBS range from $10 billion to $100 billion by the end of this decade. So what are LBS, who wants them, and what are the barriers to their successful rollout?

LBS can be split into two key segments. The first segment, most visible in North America, is the capability to provide enhanced emergency services to mobile wireless users. Surveys of U.S. mobile wireless users after Sept. 11, 2001, show that enhanced safety and security have become the most desired services. In the E911 Phase II rules, the FCC has defined a health and safety service motivated by the capability to accurately locate a mobile user in an emergency situation, such as a car accident or major natural or man-made disaster.

Deployment of E911 Phase II solutions requires challenging new technologies. Complex coordination among public safety agencies, wireless carriers, technology vendors, equipment manufacturers, and local exchange carriers also is necessary.

With these challenges in mind, the FCC defined a four-year schedule for the Phase II solution. The rollout was scheduled to begin Oct. 1, 2001, and is intended to be completed by Dec. 31, 2005.

Unable to comply with the mandate to pinpoint mobile users in an emergency to within 50 to 300 meters in the re-quired time frame, the U.S. carriers requested Phase II waivers from the FCC. Per the waivers, 95% of all handsets in service must be location capable by Dec. 31, 2005.

In Europe, project LOCUS was initiated to advise the EU on implementing E112 emergency calling services. The initial E112 implementation phase will begin in early 2003. However, the project’s report recommended that consideration of providing the location of emergency callers be deferred to 2005. This recommendation was based on a number of technical, regulatory, and business concerns.

In addition to the need to comply with government mandates, carriers want to fill a growing demand for the other segment of LBS: value-added location-based commercial services. In Japan, KDDI launched its EZ Navigation service in December 2001 and currently has more than 70 services available, including friend finder, mapping, personal navigation, stolen-property recovery, restaurant guides, train schedules, town guides, and weather information. In the second quarter of 2002, KDDI announced that it had sold more than one million handsets since service began.

KT Freetel and SK Telecom are rolling out similar services in Korea during 2002. Wireless carriers globally are keen to exploit the additional revenue opportunities from location-specific advertising and other service offerings.

Technology Choices

Both mobile- and network-based technologies can be used to implement LBSs. However, the FCC’s accuracy requirements for E911 mobile-based solutions are more stringent than those for network-based. CDMA carriers such as Verizon and Spirent PCS have chosen mobile-based techniques such as A-GPS and AFLT to determine a mobile’s position in the network.

In the case of A-GPS, the CDMA network assists the mobile by sending information about the position of the GPS satellites over the CDMA link. The mobile then can quickly make GPS pseudorange measurements and report them back to the CDMA network.

To supply the measurements required for AFLT, a mobile makes pilot phase measurements on base stations and reports them back to the network using IS-801-1 signaling. A-GPS excels in suburban and rural environments having low base-station density, while AFLT is more suited to urban environments with high base-station density and in-building applications.

By using A-GPS and AFLT hybrid techniques, the mobile benefits from their dissimilar strengths. A commercial example of a hybrid A-GPS/AFLT solution is QUALCOMM’s gpsOne®.

For U.S. GSM networks, the choice of the major carriers has been E-OTD. E-OTD systems operate by placing receivers with accurate timing sources, called LMUs, at multiple sites geographically dispersed over a wide area. The time differences of arrival of signals from three or more base stations at an E-OTD-enabled mobile and an LMU are used to calculate the mobile’s position. The principle of E-OTD can be seen in Figure 1.

Mobile-based measurements are reported back to a location processor in the network that calculates the location of the mobile and sends it to the appropriate client. In CDMA networks, this element is a PDE and in GSM networks an SMLC. The architecture of a typical E911 solution implementation for a CDMA network is shown in Figure 2.

Location Test Standards

In the United States, the time lines associated with the E911 Phase II mandate are driving the accelerated rollout of new location technologies. The carriers’ waiver agreements with the FCC call for a percentage of all new handset activations to be location-capable within challenging time frames. To reduce the risk associated with this rapid deployment, carriers and mobile manufacturers are combining lab-based performance tests and field-based tests to evaluate a mobile’s capability to perform accurate position measurements and report them back correctly to the serving network.

For E911, the FCC has elected not to establish a type-approval approach to certifying the performance of Phase II solutions. Instead, it published a set of guidelines in the OET Bulletin No. 71. This bulletin lays out a test methodology that includes empirical testing methods, including the random selection of sample locations, use of prediction models, and a statistical approach to demonstrating compliance from the resulting accuracy measurements.

To show compliance, a given set of accuracy measurements must meet the FCC’s location error thresholds. For handset-based solutions, this is 50 meters for 67% and 150 meters for 95% of the measurements.

The LIF is an industry body dedicated to advancing LBS globally. Among its wide range of activities is ITG. This group is formulating a lab-based approach to satisfying the requirements of the FCC’s OET71 for the dominant air-interface technologies. The ITG hopes to gain support for its test methodology from manufacturers, carriers, and ultimately the FCC.

The TIA recently published the first minimum performance test standard for location-capable CDMA mobiles. The TIA/EIA-916 specification for lab-based evaluation originally was developed by 3GPP2. It comprises a series of tests intended to ensure that mobiles meet at least a minimum performance threshold for A-GPS, AFLT, and hybrid technologies.

Table 1 shows a list of the minimum performance tests in TIA/EIA-916. The standard is intended to test only the position location functionality of a mobile station. It does not address interoperability with a network PDE or with other services such as voice or data.

* Hybrid tests defined only for mobiles that self-determine location

Conformance tests for location services in GSM/EDGE networks are published in 3GPP’s TS 51.010-1 standard. Test cases are defined for signaling conformance with GPS, A-GPS, and E-OTD technologies. Recently added test cases address E-OTD-capable mobile timing measurement accuracy performance.

Location Test Methodologies

To ensure compliance with the FCC’s OET71 guidelines, carriers have defined field test methodologies that characterize a mobile’s performance in representative environments such as indoors in a shopping mall or outdoors in an urban canyon. Location measurement data has been collected to perform the necessary statistical compliance analysis. However, these extensive field-test campaigns are expensive and time-consuming.

To accelerate location-capable handset time-to-market, mobile-device manufacturers need test equipment that can be configured to meet the requirements of both minimum performance testing and system performance testing required by the FCC. Such equipment can help anticipate mobile-related issues that may impact the results of field tests. This type of test system has the potential to reduce the time required to collect and analyze results in field test campaigns.

For location-capable CDMA mobile testing, the required test system must implement all of the key satellite and terrestrial network elements including the GPS constellation, multicell CDMA network, and PDE. Since 3G networks will exist concurrently with their 2G predecessors, the test system must support both IS-95 (2G) and CDMA2000 protocols.

Such a system needs a GPS environment simulation capability to accommodate eight or more GPS satellites with programmable impairments such as multipath and atmospheric effects. A real-time CDMA network emulation capability also is required, ideally simulating up to six independently programmable base stations for performance testing under handoff conditions.

The network emulator must support the IS-801-1 signaling required to exchange location messages with the mobile under test. The capability to subject the CDMA signals from the base stations to impairments such as multipath fading, noise, and interference also is highly desirable, enabling simulation of typical field environments such as urban, suburban, and rural terrain.

The majority of first-generation CDMA location-capable mobiles do not determine their own location. Instead, they pass GPS and CDMA measurements back to the PDE where the location calculation is performed.

For meaningful system-level testing of handsets, any PDE used in lab-based testing must represent those deployed in commercial CDMA networks. Licensing the algorithms from a widely deployed commercial PDE is the ideal. This capability would allow representative position fixes to be calculated and the FCC’s statistical analysis to be carried out on the results. The combination of the commercial PDE and the capability to simulate representative field scenarios in the lab would allow handset interoperability and performance issues to be identified and addressed before expensive field trials begin.

To realize the demanding measurement accuracy required to characterize mobile performance, the emulated CDMA and GPS signals must be synchronized to within nanoseconds. A test platform for location-capable mobiles should be fully integrated, with all its components accurately synchronized to ensure accurate and repeatable test results.

To maximize test resources and enable a large number of test scenarios to be performed in the minimum time, test-executive software should automate all aspects of test-campaign configuration, synchronization, execution, and results logging. Canned test suites should be supplied for both minimum performance and system performance testing, eliminating the need for a user to interpret test specifications and waste valuable lab time programming test parameters.

Including a DM in the test system, especially one capable of simultaneous control and performance monitoring of the mobile device under test, would enable closed-loop testing, minimizing the need for manual user intervention. A DM tool of this type recently was released. It integrates position location field test support with full IS-801-1 message logging and decoding as well as location charting and the statistical error analysis according to the FCC’s OET71 compliance methodology for use in E911 field trials. An example of this type of system is shown in Figure 3.

Rapid, Reliable Deployment

To accelerate the reliable deployment of LBS, whether commercial or driven by government mandate, mobile manufacturers and service providers must thoroughly evaluate the new wave of mobile devices equipped with location capabilities. Given the conflicting requirements of running a complex and increasing number of mobile test cases in the shortest possible time, manufacturers and service providers require integrated systems equipped with full test automation to maximize valuable resources, minimize time to market, and meet challenging government mandates.

About the Author

Nigel Wright joined Spirent in 1998 to run its GPS testing business in the United Kingdom and today is director of position location test solutions for the Spirent Wireless business unit. Previously, he was sales director at Telecom Analysis Systems and ran a telecommunications training business. Mr. Wright received a B.S.E.E. from the University of Brighton, England. Spirent Communications, 541 Industrial Way W., Eatontown, NJ 07724, 732-544-8700, e-mail: [email protected]

Return to EE Home Page

Published by EE-Evaluation Engineering
All contents © 2002 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.

October 2002