簡介


Cellular networks have quickly developed into an extensive wireless communication
infrastructure with almost worldwide coverage. Cellular service areas are divided into
cells and each of these cells has a base station (cellular tower) associated with it.
Techniques have been developed to track a mobile client when it is moving through the
network. These location management techniques rely on the two-way communication
between the mobile device and the network. When a user connects to the network, the
mobile device is allocated to the base station transmitting with the strongest field
strength. The most basic form of cellular positioning is to use the (known) location of this
base station. This method is known as cell identification (cell ID). Location accuracy
depends solely on the cell size, but this can be enhanced with support of other techniques.
For example, some cells are divided into different sectors by directional base station
antennas which can substantially reduce positional error. Further improvements can be
achieved by using the received signal strength, although signal strength can vary considerably
due to fading, topography, obstacles and other factors. Finally, cell ID techniques
can be improved by using the timing advance that is calculated by the base station. These
techniques are referred to as enhanced cell ID or E-CID.

When a mobile device is within range of multiple base stations, more complex
positioning algorithms can be employed, mostly relying on Time Difference of Arrival
(TDOA). For a review see Chapter 8 in Bensky (2008). Positional accuracy of these
techniques depends on the density of base stations and the reliability of time of arrival
measurements. The latter factor is in turn dependent on the bandwidth of the cellular
signal, making the Global System for Mobile Communications (GSM) signal potentially
more accurate than the Code Division Multiple Access (CDMA) signal. Angle of Arrival
(AOA) is another algorithm adopted by cellular service providers, but is not as widely
implemented. Fingerprinting techniques have also been developed for cellular positioning
similar to those discussed for WiFi positioning (e.g. Juurakko and Backman 2004, Chen
et al. 2006). Due to the effort involved in calibration, however, fingerprinting has not yet
been adopted by any of the major cellular service providers.

Whatever specific positioning algorithm is used, the positional accuracy of cellular
positioning depends greatly on the density of base stations. Horizontal error has therefore
been found to vary greatly across urban-rural gradients, with a median error in the
order of 50 to several hundred meters in urban areas and in the order or several hundred
meters to several kilometers in rural areas (e.g. Weiss 2003, Lin and Juang 2005, Mohr
et al. 2008). In one of the more comprehensive recent studies by Mohr et al. (2008) using
three different cellular operators in the U.K., the median error was 246 m in a dense
urban setting and 626 m in a rural setting.

Notwithstanding recent refinements in positioning algorithms, there are fundamental
limits to the accuracy that can be achieved with typical densities of cell towers
(Gustafsson and Gunnarsson 2005). This has been widely recognized by cellular providers
and hence the adoption of A-GPS as the technology of choice to meet the FCC
requirements for positioning. Nevertheless, cellular positioning remains of interest since
it may be available when other signals are not, as well as for older devices that are not
A-GPS and/or WiFi enabled. As a result, the refinement of algorithms for cellular
positioning remains an area of active research (e.g. Schwaighofer et al. 2003, Otsason
et al. 2005, Chen et al. 2006).

The cellular positioning implemented on Apple’s 3G iPhone is based on Google
Mobile Maps. Limited information is published on the performance of this particular
version of cellular positioning, but Google’s description suggests it is based on cell ID.

細部技術與運作



優點


缺點



參考資料


  1. Accuracy of iPhone locations-A comparison of assisted GPS WiFi and cellular positioning