Sect_2.4.1

2.4.1 How Good is GPS?

FACTORS INFLUENCING GPS ACCURACY

Although GPS can claim to excel with regard to a number of performance measures, the most important for most users is usually accuracy. The main factors influencing GPS positioning accuracy are:

Measurement errors and biases.

Absolute or differential positioning mode.

Satellite-receiver geometry.

Processing algorithms, operational mode and other enhancements.

A clear distinction must also be made between precision, accuracy and reliability of positioning results.

Back To Chapter 2 Contents / Next Topic / Previous Topic

Measurement Errors and Biases

All GPS measurements: pseudo-range, carrier phase or Doppler frequency, are affected by biases and errors (section 6.2.1). Their combined magnitudes will affect the accuracy of the positioning results. Biases may be defined as being those systematic errors that cause the true measurements to be different from observed measurements by a "constant, predictable or systematic amount", such as, for example, all distances being measured too short, or too long. Biases must somehow be accounted for in the measurement model used for data processing if high accuracy is sought. There are several sources of biases with varying characteristics, such as magnitude, periodicity, satellite or receiver dependency, etc. Biases may have physical bases, such as the atmosphere effects on signal propagation, but may also enter at the data processing stage through imperfect knowledge of constants, for example any "fixed" parameters such as the satellite orbit, station coordinates, velocity of light, etc. A useful way of considering biases is as errors which are correlated in space or time Residual biases may therefore arise from incorrect or incomplete observation modelling and hence they will be treated as random errors.

Absolute or Differential Positioning Mode

There are two GPS positioning modes which are fundamental to considerations of (a) error/bias propagation into (and hence accuracy of) GPS results, and (b) the datum to which the GPS results relate. The first is absolute or point positioning, with respect to a well-defined coordinate axis system. The coordinate system generally associated with GPS positioning is the earth-centred WGS84 Cartesian reference system. This coordinate system is realised via the coordinates of the monitor stations (of the Control Segment), and subsequently transferred to users via the (changing) coordinates of the GPS satellites. As the satellite coordinates are essential for the computation of user position, any error in these values, as well as the presence of other biases, will directly affect the quality of the position determination.

The most basic correlation between accuracy and positioning mode must also take into account the observation type used in the GPS positioning application. Hence accuracy versus positioning mode is a complex mosiac.

Higher accuracies are possible if the relative position of two GPS receivers, simultaneously tracking the same satellites, is derived. Because many errors will affect the absolute position of two or more GPS users to almost the same extent, these errors largely cancel when differential or relative positioning is carried out. There are different implementations of the differential positioning procedures, but all share the characteristic that the position of the GPS receiver of interest is derived relative to another fixed, or reference, receiver whose absolute coordinates in the satellite datum are assumed to be known. One of these implementations, based on combining the data from the two receivers before processing, is the standard mode for GPS surveying. GPS surveying is therefore essentially concerned with the measurement of the baseline components between simultaneously observing receivers. (The effect of satellite-receiver geometry in differential positioning is more complex than in the case for point positioning.)

Satellite-Receiver Geometry

The satellite-receiver geometry will also play an important role in error/bias propagation into GPS positioning results (section1.4.9, LANGLEY, 1991c, and LANGLEY, 1999).

Processing Algorithms, Operational Mode and Other Enhancements

Finally, GPS accuracy is also dependent on a host of other operational, algorithmic and other factors:

Whether the user is moving or stationary. Clearly repeat observations at a stationary station would permit an improvement in precision due to the effect of averaging over time. A moving GPS receiver does not offer this possibility.
Whether the results are required in real-time, or if post-processing of the data is possible. Real-time positioning requires a "robust" but less precise technique to be used. The luxury of post-processing the data permits more sophisticated modelling and processing of GPS data to minimise the magnitude of residual biases and errors.
The level of measurement noise has a considerable influence on the precision attainable with GPS. Low measurement noise would be expected to result in comparatively high accuracy. Hence carrier phase measurements are the basis for high accuracy techniques, while pseudo-range measurements are used for low accuracy applications.
The degree of redundancy in the measurements. For example, the number of tracked satellites (dependent upon the elevation cutoff angle, the number of receiver tracking channels, satellites apart from GPS such as GLONASS and pseudolites, etc.), the number of observations (dual-frequency carrier phase, dual-frequency pseudo-range data).
The algorithm type may also impact on GPS accuracy. For example, "exotic" data combinations are possible (carrier phase plus pseudo-range), Kalman filter solution algorithms, more sophisticated phase processing algorithms.
Techniques of data enhancements and aiding may be employed. For example, the use of carrier phase smoothed pseudo-range data, external data such as from Inertial Navigation Systems (and other such devices), additional constraints, etc.

Back To Chapter 2 Contents / Next Topic / Previous Topic