uses measurements from a reference receiver established at a known location,

along with differencing algorithms, to remove common satellite and signal propaga  

tion errors from the PVT solutions of other (mobile) receivers operating in the

vicinity of the reference station. The residual errors that remain uncorrected are

due to multipath and noise in the receivers. DGPS techniques can be applied to the

real time PVT solution or to recorded measurement data.  Real time DGPS requires

a data link  pass the reference measurements to the mobile receiver(s). DGPS

techniques can be applied to nondifferential receivers if the raw measurement data

and navigation message are accessible.  There are two primary variations of the

differential techniques, one based on ranging code measurements and the other

based on carrier phase measurements.

Ranging code DGPS (RCD) techniques can be applied to receivers with any of the

tracking architectures described in the previous paragraphs.  For RCD,

measurements from the reference receiver are used at the receiver site to calculate

corrections, which are then broadcast to the mobile receivers.  The mobile

receivers incorporate the corrections into their PVT solution, thereby removing the

common errors and improving accuracy.

The reference receiver can develop corrections for the position solution or

individual satellite ranging signals.  If the corrections are provided for the position

solution, the correction is simply the difference between the measured PVT solution

and the "true" solution consisting of the surveyed location, zero velocity, and

precise or smoothed time. However in this case, the reference and user receivers

must either use the same satellites to calculate the same solution, or PVT

corrections for each possible combination of satellites must be broadcast.  It is

usually more efficient and flexible to broadcast corrections based on individual

satellite ranging errors, thereby allowing the user receiver to select the corrections

that are applicable to the particular set of satellites that it is tracking.  Real time

RCD is capable of producing accuracies on the order of 1 metre.

Carrier phase DGPS (CPD) systems essentially calculate the difference between

the reference location and the user location using the difference between the

carrier phases measured at the reference receiver and the user receiver.  In real 

time systems, carrier phase data from the reference receiver is broadcast to the

mobile receivers.  The mobile receivers use double differencing techniques to

remove the satellite and receiver clock biases, then use the phase differences to

determine the position of the mobile receiver with respect to the reference receiver

location.

Determining the initial phase offset (cycles plus fractional phase) between the

reference station and the mobile receiver has traditionally been a process that

required several minutes.   Therefore, it is important to maintain phase lock on the

carrier signals to maintain a continuous flow of position data and avoid

reinitialization.  Consequently, CPD systems have  traditionally used continuous

tracking receivers.  Receivers which gather measurements from more than four

satellites are common since they add robustness in the event of loss of lock on one

satellite and since additional satellites can reduce initialization time.  The CPD

techniques were originally developed for surveying applications where real time

data was not essential.  However, near real time and real ti me techniques are

under development with the goal of supporting applications such as precision 

approach for

2 4