Differential GPS - Understanding GPS: Principles and Applications - page 422

mitigated due to the rapid changing path length difference between the direct and
reflected signals, which tends to average the multipath error.
The use of dual-frequency measurements can be very important for an IGPS FRS
in some applications since it allows for the mitigation of ionospheric errors, particu-
larly for longer baselines. Once the ambiguities are resolved, the system could revert
back to a single-frequency system. Because of the shorter wavelength of the L1 sig-
nal, multipath error would be reduced by approximately a factor of 4.5.
SA was turned off in May 2000. The following discussion has been retained for
a historical perspective. The effect of SA is negligible for most differential systems
because the ground and airborne GPS receiver measurements are usually synchro-
nized with GPS time, some to within 2 ms. A constant SA velocity error is removed
through a simple linear extrapolation. SA acceleration error growth is limited due to
the 2-ms synchronization. Typically, SA acceleration is on the order of a few mm/s 2 .
After 1 second, the unknown SA acceleration cannot introduce more than 1 cm of
error (1/2 at 2 ). Even in the worst imaginable case of an SA acceleration of 100 m/s 2
(10.2g), the SA error growth during the 2-ms interval would only amount to 0.2
mm. Tracking loop error would likely be a much larger concern under these circum-
stances. It is noted, however, that the airborne position can only be calculated to the
centimeter-level accuracy after the measurements from the reference receiver are
received. This can introduce data latencies of 1-2 seconds. During this time, the air-
craft position must be propagated to maintain the desired flight path. During peri-
ods of time with normal SA levels, the airborne integrated carrier phase
measurements, corrected for SA velocity error, can be used to propagate the aircraft
position with centimeter-level accuracy. During times of exceptionally large SA
accelerations, the aircraft inertial velocities could be used to propagate the aircraft
position.
8.4.3.4 Carrier Cycle Slips
The carrier-phase observable must be tracked continuously by the receiver, or the
agreement between the fixed and floating baseline solutions will diverge rapidly.
Loss of signal can occur due to the setting of a satellite, excessive maneuvering of
the user (a large bank angle in the case of an airborne user during approach or take-
off), or an obstructed view of the sky in the direction of the satellite. In any event, a
loss of signal continuity, no matter how brief, results in an unknown signal loss or
gain of carrier cycles when the signal is reacquired by the receiver. In a kinematic
environment, detection of cycle slips is vital, since allowing corrupted carrier-phase
measurements to propagate forward usually causes immediate loss of the fixed
solution. As such, identification of the cycle slip becomes paramount, and, rather
than attempt repair, the offending SV is “ignored” for a predetermined number of
epochs with the assumption that the signal will quickly return to normal. At the
conclusion of this time-out period, the data from the offending SV is once again
accepted, and the carrier-cycle integer ambiguity resolution process restarts for the
SV. In the interim, if a minimum of four SVs (not including the offending SV)
had their ambiguities resolved at the time of cycle-slip detection, the fixed baseline
solution is maintained. Otherwise, at best only a floating baseline solution can be
provided.