Differential GPS - Understanding GPS: Principles and Applications - page 421

8.4.3.1 Stand-Alone Ambiguity Resolution
Using the approach outlined in Section 8.4.1, approximately 2 to 5 minutes are
required to resolve the carrier-phase integer-cycle ambiguities. The time required
depends on a sufficiency of satellites—six or more are generally needed. Good sat-
ellite geometry is also beneficial, as is motion of the airborne platform (though not
necessarily required). The latter supplements the normal motion of the GPS con-
stellation and reduces carrier-phase and code multipath. Motion of the constella-
tion is vital to the resolution of the carrier-cycle ambiguities. As the various
candidate ambiguity sets are identified and evaluated over time, only one set can
persist given the dynamics of the constellation and, to a lesser degree, the added
motion of the platform. Simply stated, without motion, the technique presented
would not work.

8.4.3.2 Pseudolite Ambiguity Resolution
During the initial feasibility studies for the FAA LAAS, integrity beacons (a form of
pseudolite) were used for rapid carrier-cycle integer-ambiguity resolution. These
devices are low-power transmitters, two of which are placed within several miles of
a runway threshold along the nominal approach path. These transmitters typically
operate at L1 and are modulated with an unused PRN code such that they are not
mistaken for an SV. The several minutes of time required to resolve the carrier-cycle
ambiguities as described earlier are reduced to seconds with this method due to the
rapid change in geometry as the aircraft passes through the signal “bubble” created
above the pseudolites. A second GPS antenna mounted on the belly of the aircraft is
used to acquire the pseudolite signals. The presence of the two pseudolites also
reduces the requirement of visible SVs to four and ensures that as the aircraft exits
the bubble, the carrier-phase integer-cycle ambiguities are resolved. Centimeter-
level positioning accuracy is thus ensured from this point to touchdown and rollout.
Both the real-time cycle ambiguity resolution and the centimeter-level positioning
accuracy have been demonstrated in flight testing with transport category aircraft
[22, 23]. More recently, pseudolites have been investigated as a means of improving
local GPS satellite availability [36].

8.4.3.3 Accuracy
Once the carrier-phase ambiguities are resolved, the accuracy of the DD measure-
ment is determined by the carrier-phase measurement. In this case, multipath is the
dominant error source. If the reflected signal is weaker than the direct signal, the
phase measurement can be in error by up to 0.25 . If the reflected signal is stronger

than the direct signal, cycle slips are likely to occur. Typical wide-lane DD measure-
ment errors are on the order of 2 to 10 cm (2

λ

). Due to geometry, vertical position-
ing errors are between 1.5 to 2 times the DD measurement error, resulting in up to
20 cm (2

σ

) vertical positioning errors. Horizontal positioning errors are generally
less than 20 cm (2

σ

). If both the ground and the airborne antennas are placed in a
rich multipath environment, vertical positioning error further degrades to approxi-
mately 40 cm. However, as soon as the aircraft is in motion, airborne multipath is

σ