|Issue:||Asia-Pacific II 2002|
|Topic:||Choosing the Right Technology for Location – Based Services|
Jason Angelides is a director at TruePosition in King of Prussia, PA. He has been with the company since 1998, and has been involved in the design and development of wireless products since 1994. Jason has a BS from Villanova University and a Master’s degree from the Illinois Institute of Technology.
Location – based services differ in performance and in their ability to support legacy mobiles and low – cost devices. As a general rule, U-TDOA, which calculates location based on any user’s phone signal, is best for high accuracy and availability in indoor, urban and suburban areas. E-OTD, which uses the base station’s signals, is for suburban applications requiring moderate accuracy. A-GPS, based on satellite signals, is accurate in suburban and rural environments and Cell ID works for low – accuracy services in cities.
Asia has long been a leader in GPS and other location – mobile services (LBS). Korea has a GPS – enabled network. “Real-time traffic information, time – to – destination estimates and travel directions” In Japan, KDDI provides a location-based store finder for McDonalds outlets. In Hong Kong, you can get a bilingual Chinese-English LBS service for motorists that provides real-time traffic information, time – to – destination estimates and travel directions. It will even tell you where you can park your car or find a nearby gas station. Researchers claim that location-enabled services will generate as much as $13 billion in revenues for today’s hard-pressed wireless voice services by 2005. For nearly half a decade, location-based services (LBS) have been regarded as a way for wireless operators to expand their service offerings, helping them to retain and attract customers while creating completely new sources of revenue. Although wireless operators have thus far been reluctant to commit to widespread LBS deployment, a recent IDC survey testing receptivity to emerging mobile phone services named emergency assistance, navigation and concierge services as the three most desired services among consumers, showing that interest in location is still very high. Members of the wireless community match this sentiment, where it is generally agreed that LBS will be a large part of a wireless operator’s future revenue stream. While there has been substantial discussion about the types of services that will be offered and how they will be used, there has been surprisingly little talk about the underlying location technologies that will enable these services. This is an important issue, as the variances in performance, cost and availability can directly impact the effectiveness of a given location service. In examining each of the location technologies available to the mass market it becomes clear that they do differ in their performance characteristics, suitability to certain environments and, consequently, their ability to support various location – based services. In the widely contrasting Asian land-scape, ranging between densely packed, obstructed urban areas and sparsely populated plains, the choice of the appropriate technology is critical if performance and profitability goals are to be met. Wireless Technologies Examined In the US, where the FCC’s E-911 mandate is driving development and adoption of high – accuracy location technologies for public safety use, three location technologies have emerged to help operators comply – Assisted Global Positioning System (A-GPS), Enhanced Observed Time Difference (E-OTD) and Uplink Time Difference of Arrival (U-TDOA). Each method utilizes the same mathematical triangulation principles to calculate locations by measuring the difference in time it takes for a signal to reach a set of known objects. Here in the European union, another technology called Cell ID has been introduced into the commercial market and is being used to support a variety of community services such as location – based dating and games. The primary difference between the first three technologies mentioned lies in the place the location is calculated and in the equipment that is needed to perform the calculation. With A-GPS, an end – user device, such as the mobile phone, is equipped with a specialized chipset and antenna that receives signals being sent from the U.S.’s GPS satellite constellation. The signals originate from the satellites at precisely the same time and propagate toward earth at the speed of light. When the end – user device receives the signal, it calculates its location based on the difference in time it takes for each of the satellites’ signals to reach the end – user’s special phone. E-OTD operates on the same principles, except instead of a signal originating from a satellite, signals are sent from cellular base stations and are then picked up by specialized mobile phones that receive the signals and calculate a location. In both A-GPS and E-OTD, the location is calculated in the phone. U-TDOA is the opposite of E-OTD and A-GPS in that the end – user mobile device is the source of the signal, and signal detection and calculation is done at equipment located in a wireless operator’s base station. With U-TDOA the mobile device does not need to be modified for the system to work, meaning it operates with all existing mobile phones and devices. Finally, Enhanced Cell ID utilizes a wireless network’s inerrant capability to recognize which sector of a given cell site a phone is communicating with and enhances the data to determine what general area of the cell sector the caller is in. Like U-TDOA, Cell ID does not require the use of a specialized phone to operate. Each location technology has unique strengths and weaknesses relating to enabling location-based services. From a pure performance perspective, location technology is best compared by examining the two core components of a wireless location service, its accuracy and yield. Yield refers to the location technology’s ability to obtain a location fix in all environments – indoor, urban, suburban and rural. This capability is important for many of the location – based services, especially those that require high reliability such as emergency services and personal security, where anything less than near perfect yield could result in catastrophe. In rural environments, all of the location technologies mentioned will show nearly perfect accuracy. However, as population density increases along with physical obstructions such as buildings and bridges, A-GPS begins to suffer. This is due to the distance the satellites are from earth. By the time the signal reaches earth it is extremely weak, and even though improvements have been made to overcome this issue, in many situations A-GPS systems simply cannot calculate the distance of a location. This is not an issue with the other location technologies as their close proximity to the signal source allows locations to be processed in all environments. Accuracy refers to the radius in which a location technology can pinpoint the location of a wireless phone. With A-GPS, typical accuracies are in the 50-metre range and in unobstructed environments accuracies can drop down into the 10 – 20 metre range. Similarly, U-TDOA accuracies range from 50 – 60 metres and drop to the 20 – 30 metre range in areas of reduced obstruction. E-OTD, once considered a direct match to GPS in terms of accuracy, has proven to be much less successful as witnessed by the results from the recent trials in the U.S. With E-OTD typical accuracies in suburban and rural environments are in the 75 – 125 metre range. These numbers increase in indoor and urban situations because the signal processing power of the mobile unit is not powerful enough to detect highly attenuated signals from distant or blocked cell sites. Cell ID is the least accurate of all of the location technologies. With Cell ID accuracy is a function of how closely spaced cell towers are. As a result it is essentially useless in non-urban environments where cell sites may be many kilometres apart resulting in accuracies of several hundreds or even thousands of metres. In urban centres where cell sites are much closer together, accuracies can reach the 150m range but typically fall between 250 and 500 metres, making Cell ID only suitable for only a small percentage of location services. While accuracy and yield figures are helpful in establishing the differences between the technologies, the commercial applicability of the technologies is best seen when put in the context of the performance requirements of leading location – based services. The chart below juxtaposes the capabilities of each of the location technologies with estimated performance requirements of major location – based services. As shown in the chart, the technologies vary drastically in their ability to support the leading location – based applications. Beyond accuracy and yield, there are two other factors that affect the location technology: the addressable market a location technology covers, and the ability for a given location technology to support device – based applications. Both A-GPS and E-OTD require special handsets to calculate a location, limiting the addressable market for the provision of location services to those who are willing to replace their phone. Network-based technologies such as U-TDOA and Cell ID locate all existing and future phones on the operators’ network. This is a serious issue when you consider that only a small percentage of location – specialized E-OTD and A-GPS phones may actually be in service within the next two to three years. Assuming a penetration rate of 20% per year, this still means that, by 2005, little more than a half of an operator’s market would be covered. Finally, many high – value applications such as child or pet locators, or asset tracking, require a small location device to operate. All of the technologies listed can support device – based applications, however they differ in the level of specialization their respective location devices require. Because U-TDOA and Cell ID can calculate locations without any enhancement to the mobile phone, location devices need only consist of little more than a stripped – down wireless phone. Both A-GPS and E-OTD require additional components in the phone or device to increase the level of specialization and concurrently the overall cost of the device. Conclusion The underlying technologies supporting location – based services differ in the two critical performance measurements of accuracy and yield as well as their ability to support legacy mobiles and low – cost devices. For wireless operators and location – based service providers careful consideration should be given to selecting the most appropriate location technology for a given set of services. As a general rule, U-TDOA is best suited for applications that require high accuracy and availability in indoor, urban and suburban areas, E-OTD is suited for suburban applications requiring moderate accuracy, A-GPS is suited for applications requiring high accuracy in suburban and rural environments and Cell ID is best for low – accuracy services in dense urban areas. Beyond performance, the ability of U-TDOA to combine good performance with the ability to locate all phones makes it an appealing choice for operators looking to provide services to large numbers of their subscriber base. As location services become more widespread, the underlying positioning technology will be transparent to the end user, who will care only about availability, reliability, performance and cost. Because there will most likely be multiple location technologies available to operators, a hybrid approach may be the key to future success, complete market coverage and cost – effective location support. For the short term however, U-TDOA offers the location performance to enable high – value applications in a package that makes location available to all wireless subscribers today. As other location technologies become more prevalent, they can be combined with U-TDOA to offer subscribers the widest variety of services in the broadest customer base.