As explained in part 1 of this series, whenever a mobile phone makes a call, the Network Operator can establish through which cell site the phone is communicating. Based on the location of the cell site, the Operator is able to estimate the location of the phone. In a particular network, the cell site has a unique identifier known as the Cell Identifier or Cell-ID. This method of location estimation, which uses Network measurements of the Cell-ID, is known as Network Cell-ID or NCID.
The most precise case is when a handset is in active communication with the network. In general the handset will connect with the nearest base station and therefore this information constrains the handset to the region served by that cell site. In some cases, for instance when the base station concerned supports a micro or pico cell with a very limited transmission range, the fact that the mobile is communicating with that base station provides a relatively precise constraint on the position of the handset, of the order of perhaps 50 to 100 meters. Of course with a sparse distribution of cell sites, for instance in rural areas, one site may cover many tens of square kilometers. As a result the positional information provided by the association between base station and mobile is much less precise.
For an omni-directional site, the best estimate of the handset position is the coordinates of the site. In cases where the serving base station is part of a multi-sector site, the coverage area of the particular base station is likely to be more restricted. In this case it is possible to refine the estimated handset position to the notional centroid of the coverage area of the relevant sector.
In most cases, Cell-ID measurements are used in isolation to derive a coarse point estimate of the mobile's position. However, in some cases, multiple Cell-ID measurements may be available. In such cases the measurements can be combined to obtain a more accurate fix.
When a handset is not in active communication, in so-called idle mode, it is not possible to identify the closest cell site. However, the network can constrain the estimated position to within a location or paging area. The network maintains this information in order to be able to focus its paging attempts when there is an incoming call destined for the handset. A location or paging area usually represents a small collection of cell sites. The uncertainty associated with the handset's position may be up to an order of magnitude greater than when the handset is in-call. In such cases, if a higher accuracy is required, the network may establish a brief connection with the mobile to identify the nearest base station for greater accuracy.
Source by Chris Drane