Abstract
Recent developments in advanced GPS receiver designs require lower-level
observables than what are currently modeled by most commercial GPS
software simulation packages. Kalman filtering based joint signal
tracking algorithms, a subset of Direct Correlator Output Processing
(DCOP) technologies, promise more robust tracking performance by
jointly processing the demodulated I and Q signal components of each
signal, thereby exploiting the geometrical correlations of the tracking
loop errors and platform dynamics. Ultra-tightly integrated GPS/INS
systems also require these I and Q signal components, which are scarcely
available in today's commercial GPS simulation software packages.
Of those commercial packages that do generate these measurements,
few, if any, are driven by a receiver trajectory, satellite ephemeris,
and other environmental parameters. All the models discussed in this
paper are driven by the above mentioned truth environment. This is
important for integration into inertial measurement simulators when
evaluating GPS/INS coupled systems. A high-fidelity approach to generating
these I and Q measurements was accomplished by designing a GPS signal
generator at the output of the analog-to-digital (A/D) converter
and a realistic GPS DSP Receiver model to process the GPS signal
generator output. A faster analytical method was developed that generates
the same I and Q measurements through a mathematical model of the
tracking process. Each I and Q simulator (fed by the same trajectory,
satellite position, and other environmental parameters) generated
independent I and Q signal components which were compared for a first-level
validity check. The analytical I and Q model followed the higher
resolution truth model very well, especially at low dynamics and
over short time periods. At higher dynamics, the increased fidelity
of the signal and receiver simulators become apparent, and the I
and Q analytical approximations exhibited less accuracy. Both the
high fidelity and the analytical models provide the ability to design
and test new signal tracking architectures, with access to the demodulated
but unprocessed I and Q signal components. The analytical I and Q
model operates as a self-contained I and Q signal generator, driven
by the same truth environment components used to simulate an INS.
It provides the typically unavailable I and Q signal components of
a GPS signal, and has been used in the modeling and simulation of
ultra-tightly integrated GPS/INS systems.
Users
Please
log in to take part in the discussion (add own reviews or comments).