Sect_6.2.5

6.2.5 Measurement Biases and Errors

RECEIVER CLOCK UNCERTAINTIES

GPS receivers are equipped with quartz crystal oscillators, which have the advantages that they are small, consume little power and are relatively inexpensive. In addition they have good short-term frequency (or time-keeping) stability (section 1.3.2). However, some receivers are equipped with I/O ports to permit the connection of an external frequency standard such as a cesium, rubidium and even a hydrogen maser, for specialist applications.

Although the time scale defined by individual receiver clocks have essentially arbitrary origins, they can be tied to a well established time scale, such as GPST, in a number of ways. Generally, the time origin of a GPS receiver is set automatically as soon as sufficient satellites are tracked to carry out a single point pseudo-range navigation solution, using the solution strategy described for "Receiver-Biased Range Positioning" in section 1.4.2. The subsequent time scale defined by the corrected receiver clock is then nominally that of GPST because:

The synchronisation at some epoch (that is, the process of defining the time origin) is susceptible to error. Generally, it can be carried out only at the 0.1 microsecond level under Selective Availability (and perhaps to the 0.01 microsecond level with SA off).
The stability of the time scale is directly related to the quality of oscillator used, and how often the current clock time is synchronised with GPST through the use of GPS pseudo-range observations. This "clock drift and reset" is illustrated in the Figure below.

Clock drift and periodic reset to "true" time.

RECEIVER CLOCK BIAS

MAGNITUDE:

The receiver clock is synchronised to GPST through the normal operation of code-correlating receivers to about 0.1 msec accuracy under SA. Therefore residual biases of the order of a dekametre (tens of metres) remain, and must be accounted for in some way.

OPTIONS:

Construct a range-like observable from which the receiver clock error has been eliminated -- DIFFERENCE between-satellites.

Model the satellite clock error as a "random process" -- ESTIMATE an additional parameter.

These are the same options available as in the case of the satellite clock bias.

STRATEGY:

Either option requires that two or more satellites are tracked simultaneously by the GPS receiver. Both options are very effective at COMPLETELY eliminating the effect of the receiver clock bias.

Back To Chapter 6 Contents / Next Topic / Previous Topic