The Role of the Device in the Location Architecture

The device fills a specialized role in the location architecture. As previously described, it is both the subject of location data and a client for that data. In this chapter, the device can take on some of the roles of both an ALE and LIS. The advantages of device-based measurements and location are significant, but it is also important to consider how this affects the overall architecture. This section describes how the device can add value to the location architecture by working with an existing infrastructure.

Device Measurements

A device that provides location measurements effectively becomes an ALE. However, to characterize the device as an ALE is not sufficient-the device does not share a number of the key characteristics of the ALE. The LIS-ALE relationship is static. In comparison, the device is not a network element that is under the control of the access network provider. Devices come and go. A device is not tied to a particular access network. Therefore, no preexisting relationship between the device owner and access network provider can be assumed.

The static LIS-ALE relationship also has the aspect that, by design, the LIS should understand the measurements provided by the network ALEs. The LIS is directly coupled to the access network it is serving and, by definition, can be expected to understand the technology used to provide that access. On the other hand, a device visiting an access network may have some set of location measurement or determination capabilities that are not understood by the LIS or may not even be supportable in the specific access network serviced by that LIS. Thus, in order to properly recognize and take advantage of device measurement capabilities, some mechanism for the device and LIS to be able to exchange and match location determination capabilities is required.

The relationship between the LIS and a device differs from the LIS-ALE relationship. However, HELD already provides a means to establish a relationship between the LIS and the device, including the discovery procedure, which must occur before communications are established. It makes sense to use HELD, an already established means of communication, for device-based location measurements, or at the least, to bootstrap the process. The extension options for HELD are examined in more detail later in this chapter.

Autonomous Location Determination

Some devices will be able to determine their own location without access network assistance. If a device is able to determine its own location and fill the roles that the LIS would otherwise provide, it can operate completely autonomously. Location determination is the primary function, but the device can also provide an external subscription interface with a location URI and privacy management. That is, the device and the applications on it that use location information always have the option of working independently of the LIS.

For a range of devices, a location determination capacity is all that they can provide. The other features that are normally provided by the LIS may not be feasible. The full range of features that the LIS can provide could introduce unmanageable complexity in device implementation; these features could also stretch limited network and computing resources available to the device. The decision on whether or not to use the LIS depends on the requirements of the application or the user of location information. In particular, providing a verifiable signature for location information that identifies a recognized provider is generally out of the scope of an individual device's capabilities.

Later in this chapter, several configurations are shown where a device is able to delegate responsibility for certain functions to the LIS.

Preventing Location Fraud

Location fraud, or providing incorrect location information for personal gain, turns into a real problem when devices become part of the location determination process. This is of special interest to the LIS since the LIS becomes the nominal source of the information and may represent the party responsible for the veracity of the information. This is true even when the LIS does not digitally sign location information.

The LIS should not use information provided by the device unless it can have some way of ensuring that it is correct. When location is provided by a device using autonomous location determination, the LIS must verify that this information is correct.

A basic way of verifying location information is to perform a simple containment check. The LIS determines a location estimate using trusted sources of measurements, which it then compares to the estimate provided by the device. If the device-provided location falls within the region of uncertainty determined by the trusted method, or it lies close enough, the LIS can use the device-provided information. The precise method of validation of asserted location information is a matter of specific policy at the LIS-the validation decision needs to balance the needs of applications for precision against the imperative that the LIS remain a trustable source of accurate data.

Location measurements from the device provide a means for a device to improve the quality of a location estimate. This presents another opportunity for providing fraudulent information to the LIS. Measurements can be dynamic or hard to obtain outside of a particular network; therefore, tricking the LIS into providing an incorrect location is more difficult to manage using measurements. However, it is equally difficult for the LIS to ensure that measurements received from a device are valid. Therefore, if the LIS cannot check the measurements, it should check any location estimate derived from those measurements. This check occurs in the same way that it would check location information that is sourced entirely from the device.

These checks ensure that a LIS can be trusted to provide accurate location information, while benefiting from the improved precision that can be gained by including the device in the calculation process. Location information that is provided by the LIS can be used with fewer concerns about fraud as long as the validation policies implemented in the LIS correctly balance the application's security, accuracy, and reliability needs against their precision and utility requirements.

Architectural Goals

It is important that the protocols and architecture facilitate choice. Devices that support any set of features should be able to operate independently where possible and then call on the LIS to provide those functions they do not support, or they cannot support, natively. If a device is capable of providing measurements, then the LIS should be able to use that information. But if a device can determine its own location, the LIS needs to provide supporting functions, like digital signatures, location URIs, or even GPS assistance data, as appropriate.