Wireless Location Uses Geometrical Transformation Method With Single Propagation Delay: Model and Detection Performance

Abstract

Inspired by recent promotions of wireless location services and applications, this paper develops a simple model for detecting a mobile caller location with low implementation complexity and promising accuracy. The model that is described in this paper employs the geometrical transformation method with single propagation delay measurement. Depending on the number of available surrounding base transceiver stations (BTSs), a set of the pseudodistances and pseudodistance differences is provided through the geometrical distances between the home and multiple foreign BTSs, which is called the geometrical transformation method. Accordingly, the mobile caller location can be detected using the circular location technique and the hyperbolic location technique. The major benefit of the proposed model is that during a call to detect the mobile caller's location, the use of geometrical transformation allows us to overcome the location handover problem. A forcing handover in a global system for mobile communications (GSM) network or a three-way soft handover in a universal mobile telecommunications system (UMTS) network, for example, occurs when detecting the location when a standard model that is based on the measurement of the propagation delay is utilized. Comparing the standard model with the proposed model, the former needs at least three propagation delays that were simultaneously measured from at least three BTSs, whereas the latter needs only a single propagation delay that was measured from the home BTS. Moreover, in the case where we do not have a priori information of a non-line-of-sight (NLOS) error, a deterministic parameter that is related to the propagation delay is defined. This parameter is shown as limited a priori information for calibrating the coarse resolution in the proposed location model, which is caused by the NLOS error. To have significant findings, we conduct a practical experiment in an urban environment based on a 1.8-GHz commercial cellular network. In the five-BTS case where the location error is less than 300 m, the probability of location error that the proposed model with a bit period of 1/2 achieved is greater than 0.85. Although the performance of the proposed location model does not satisfy the U.S. Federal Communications Commission Enhanced 911 (FCC E-911) requirements, it is a good solution for the deployment of wireless location services and applications.