GNSS 2003, Graz, Austria, 22-25 April 2003 1/18
Hardware Simulator Models and Methodologies
for Controlled Indoor Performance Assessment of
High Sensitivity AGPS Receivers
Gérard Lachapelle
Department of Geomatics Engineering
University of Calgary
2500 University Drive NW
Calgary, Alberta T2N 1N4, Canada
Lachapelle@geomatics.ucalgary.ca
Co-Authors
Elizabeth Cannon, Richard Klukas, Sanjeet Singh, Robert Watson University of Calgary
Peter Boulton, Arnie Read, Ken Jones, Spirent Communications, UK
1 Abstract
High sensitivity GPS-enabled mobile telephones and personal digital assistants are being
introduced for emergency call location and location-based commercial services. Simulation of
weak GPS signals received in various environments has been proposed as a cost-efficient
method of verifying the positioning performance of such mobile telephones. Recent results (e.g.
Boulton et al 2002) have demonstrated both the success and limitations of weak signal and
multipath models implemented in an advanced GPS hardware simulator. In particular, the lack of
variability of multipath delay as a function of time is suspected to be a major reason for the
discrepancy between simulator and field test results. This paper discusses how these model
limitations have been recently addressed and demonstrates the improvement in the correlation
between the simulator results and field test results. A high sensitivity GPS receiver type is used
to collect data under two scenarios, namely one in an outdoor environment near a building and
another inside a residence. This data is then used as a baseline to attempt to generate signals in
the simulator that results in measurements with similar stochastic characteristics. An analysis of
the results in terms of fade and position demonstrates the viability of this approach to test
receivers under controlled environments.
2 Introduction
Assisted-GPS technology is used to acquire and track weak signals in partly or totally obscured
locations, such as in urban canyons and indoors. External assistance consists of ephemeris and
Doppler information used during the acquisition phase. Since the largest user group by far is
cellular phone users, the external information coming from a GPS reference receiver is usually
provided to the receiver via the cellular phone network. Once signal acquisition has occurred,
successful weak signal tracking is achieved through the use of longer signal integration time
periods (e.g. Garin et al 1999). A receiver capable of tracking weak signals is referred to as a high
sensitivity GPS (HSGPS) receiver herein. In order to test the HS capability of an AGPS receiver,
signals can first be acquired under normal LOS conditions in the field or on a simulator and then
the receive can be moved indoor in the case of field testing or the signals modified in the case of
a simulator.
GPS hardware simulators can accurately replicate signals-in-space through modeling of the
satellite constellation motion and that of the user. In a live field environment, the GPS receiver,
typically embedded in a cellular phone handset, is subject to the effects of signal degradation
caused by path loss [the attenuation of the line-of-sight (LOS) signal as it propagates from
satellite to user and, in particular, multipath or reflected signals. For a laboratory test to serve
satisfactorily as an alternative to field tests in weak GPS signal environments, the modeling of this