4.4.1 Testing GPS Surveying Systems:
The dilemma facing individuals and government Certifying Agencies is deciding
on just what aspect of the GPS technology to actually test, and then to
decide on how to go about doing it. Consider the following possibilities:
- Testing of the GPS "system" as a whole -- primarily the
Space and Control Segments (section 2.2.2
and section 2.2.3).
- Testing a class or brand of GPS instrument, involving the combination
of hardware and software -- the User Segment (section
2.2.4).
- Testing a particular GPS instrument unit -- on a regular basis,
or in response to a suspected problem.
- Testing the skill of personnel -- in field and data processing
procedures.
In this section several testing strategies are mentioned. The most common
type of test is based on the results of the baseline
solution, a product of the processing of datasets explicitly collected
for the purpose of testing specified hardware and software. The results
are compared against "ground truth" data, either provided by previous
GPS surveys (using a method which is of a higher accuracy than that being
tested) or conventional survey techniques. It is less common to test
the raw data quality.
However, before proceeding to consider some testing strategies, certain
difficulties shared by some or all of the testing procedures should be mentioned:
- GPS errors/biases are a function of time
-- how can a test then be truly "conclusive" if it is carried
out only once?
- Some GPS errors/biases are a function of geographic
location -- how can a test be considered "conclusive"
if it is carried out in only one location?
- The propagation of most GPS errors/biases into the baseline solution
is a complex combination of factors, such
as time, location, baseline length and satellite geometry -- should
all baseline lengths be sampled, all geometries, etc.?
- The quality of GPS baselines is not just a function of errors and biases,
but also the length of observation session and the type
of carrier phase solution -- what operational procedures for
data collection and data processing should be insisted upon?
- The quality of GPS baselines is influenced by the data editing and
pre-processing procedures which are used -- should automatic processing
procedures be insisted upon?
- What should be the outcome of the testing -- a lifetime "stamp-of-approval",
or an annual "certification"?
Three test strategies can be identified under this category:
Calibration Testing
Certification Testing
Investigative Testing
Back To Chapter 4 Contents
/ Next Topic / Previous Topic
© Chris Rizos, SNAP-UNSW, 1999