The University of Melbourne  

FREQUENTLY ASKED QUESTIONS

451-337 Satellite Positioning and Geodesy


Students with questions on practical work are asked in the first instance to email their question rather than coming to see the lecturer personally. A written response will be provided within 1 working day of the question being posed. Both the question and the answer will be published (anonymously) on this page.
 
General questions
Assignment 1
Assignment 2

Assignment 3

Assignment 3
Assignment 5
Back to 451-337 home page
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 


Question Answer
GENERAL QUESTIONS
Are there any texts or reference books that you would recommend to go with the lecture notes for Satellite Positioning and Geodesy? Although there are many excellent GPS texts and web pages available, there is no one ideal text that covers the syllabus for the entire subject. If you want to see some books that I suggest as useful, there is a list in the front of the class notes. If you haven't got a hard copy of the notes from the Book Room, you can access a digital version from the subject web-page.
   
QUESTIONS ABOUT ASSIGNMENT 1
Getting familiar with coordinates
What format should be used in writing-up Assignment 1? This prac requires a series of calculations involving geodetic coordinates to be carried out. Present your results and any pertinent comments that may help to explain what has been done and what conclusions could or should be drawn. There is no need to explicitly follow a scientific report format. By the same token, a page of coordinates, without explanatory comments, is not acceptable.
   
QUESTIONS ABOUT ASSIGNMENT 2
Applications of GPS
No questions asked as yet!  
   
QUESTIONS ABOUT ASSIGNMENT 3

GPS navigation solution

What value should we use for the initial receiver clock error in Assignment 3 ? In fact, you can use any value you like. The further away your initial value is from the true value, the longer it will take for the solution to converge. My advice is to use an initial value of 0 seconds. This is because we know the clock error will be small (some milliseconds). By choosing zero as the initial value, convergence should occur quickly.
In calculating the PDOP and GDOP for Assignment 3, do we have to transform the variance matrix from the geocentric cartesian frame into the local (east, north up) reference frame? No you don't. Both PDOP and GDOP involve all three components of position. Thus if you add the three variances expressed in terms of geocentric cartesian coordinates or the local system, you will get the same answer. For example:
For Assignment 3, what is an acceptable level for the PDOP and GDOP values? There is no definitively correct answer to this question. DOP values of less than 5 are generally acceptable. Certainly PDOP or GDOP values greater than 8 are cause for concern.
   
QUESTIONS ABOUT ASSIGNMENT 4
The future of GNSS
No questions asked as yet!  
 

 

QUESTIONS ABOUT ASSIGNMENT 5
GPS control survey
For Assignment 5, what datum should be used when adjusting the GPS network in DNA?

The project specification requires GDA94 coordinates to be supplied for BULL, TRACK and LAWN. The easiest way to achieve this is to adjust the network within the GDA94 reference frame.

For Assignment 5, what process should be used for scaling the variance matrix and why is this done?

Scaling the variance matrix is done when you are convinced that the a priori stochastic model is inappropriate - nearly always the case with GPS observations. The sequence to follow is :

  • Perform the minimum constraint adjustment
  • Eliminate any poor quality baselines (think about how you will test this??)
  • If Chi-squared is too big (again a statistical test) after bad baselines are removed, you must conclude that the a priori variances are too small, therefore you can legitimately scale the variance matrix by sigma zero^2
  • Re-run the adjustment
  • The Chi-squared value will equal the degrees of freedom (as expected) and sigma zero will = 1
  • Only do this for the minimum constraint adjustment, not the constrained adjustment (why??)
 
 
 
 
 
 
 
 
 
 
 
 
 

Created : 27 June, 2005
Last modified : 26 June, 2006 10:30
Maintained by : Philip Collier
Contact : p.collier@unimelb.edu.au