Geo-Positioning, GPS, DGPS, and Positioning Accuracy

 Assigned Reading:

Please read Chapter 2 "Positioning Systems" in The Precision-Framing Guide for Agriculturists, by Deere & Company.

 On the Web http://www.trimble.com/gps/index.htm study the following:

How GPS works
Differential GPS
Putting GPS to work

Scan http://nespal.cpes.peachnet.edu/home/links/pa/gps_resources.asp
to see items that may interest you at that site (no assigned reading at this time).

Please read http://tycho.usno.navy.mil/gpsinfo.html and answer the following:
1.  The general population has access to Standard Positioning Service (SPS).  Give the expected accuracy specification with Selected Availability on for

a.      Horizontal____________

b.      vertical ______________

c.      time ________________

GPS Signal Characteristics
1.      The precision of a measurement is limited by the wavelength.  The shorter the wavelength, the more accurate the measurement.  If we use as a rule of thumb that measurements using with phase shift relationships, (http://www.sfu.ca/sonic-studio/Phase-Shift.html) we can measure to 1% of the wavelength, what are the missing wavelengths and expected limits of accuracy for each of the following.

Note frequency*wavelength=speed of light, MHz=106Hz or 106cycles/second.  The speed of light is 300,000,000 meters per second

 

Frequency band

frequency

Wavelength (meters)

Expected accuracy centimeters

L1

1575.42MHz

0.190

0.19

L2

1227.6MHz

0.244

____

C/A

1.023MHz

____

____

P

10.23MHz

____

____

This example explains how receivers based on the C/A frequency band only are less accurate that those that use L1 and C/A for measurement.  Systems that use C/A are sometimes called GIS receivers.  The best accuracy for this class is a little better than one meter.  Most of the units used for precision agriculture are of this class.  Those that use the carrier wave (L1) have the capability of being more accurate and are sometimes called surveying receivers with an accuracy of one centimeter.  The most expensive of this class use both L1 and L2.  This adds the ability to better account for errors caused by the atmosphere and increases the speed with which the unit can achieve centimeter accuracy.  These differences will be explored more later when you study the specifications of available equipment.

 2.      How long is the C/A code pattern (i.e. period). 

 Selective availability
3.      Selective availability reduction is accomplished by changing which two items?

a._______________

b._______________

 

GPS system segments
4.      How long is a sidereal day?  ____________________
http://pegasus.phast.umass.edu/a100/handouts/sidday.html

5.      How does this cause the satellites to appear 4 minutes earlier each day?



6.      Approximately how many times a day does a GPS satellite circle the earth.

 

 

 Clock accuracy
If a clock had a timing error of one microsecond, what would be the error in the distance measurement? (This exercise is to impress you on the quality of the clocks on the satellites)

7. What would be the error for 1 nanosecond clock error? _______________

 

 

Where are the satellites and how many are available at any given time?
The first thing to understand is how to change our time to Greenwich Mean Time.
http://www.greenwich2000.com/time/info/time.htm

 

Reading Greenwich Mean Time on this page will help you understand what time to enter to find where the satellites are in your area at this time.

To make some sense of these numbers look up your location on a globe. http://cliffie.nosc.mil/~NATLAS/index.html  and click world.   For example San Luis Obispo, CA has an approximate location of -120 degrees. (120 degrees west of Greenwich (0 degrees))  Since it takes 12 hours for the earth to rotate one half way (180 degrees) Greenwich is hours ahead of San Luis Obispo in time.

 

Then go to http://www.trimble.com/satview/index.htm and look at satellite locations for your area at the time you are observing the web page.   
Note that an azimuth is the clockwise angle from north.

 

Understanding GPS specifications
Read the definitions of 2DRMS, RMS and CEP at
http://www.redhorsetech.com/aggps/NFlinks.html and compare to page 24 and 25 of text.  Then use the following to relate it to statistics.
(Note if you like the presentation below, the total publication is for sale)
Go to the following web site: 
http://www.tellduxbury.com/seestatframeset.htm
click tour seeing statistics (twice)
click table of contents (on the side)
click Normal Distribution
click 7.0.1 and click through the pages to read what you need to understand 7.3.1
click 7.3.1 

Use the calculator to calculate the error circle as one changes the accuracy specification

 2DRMS
Middle  (on bottom)
mean =0
Standard deviation (Std Dev) = 50
Z score = 2
Press enter.
The program calculates that the range of error is +/- 100 meters.  Recall your answer to question 1. 

RMS
Z score = 1  Enter
Error circle = 50 meters  Percent of the time the answer is right 68.3%

 CEP
Probability (prob.)= .5 (Percent of the time the answer is right 50%)
Error circle = 34 meters

Thus a manufacturer can increase the apparent accuracy of his equipment (from 100 meters to 34 meters) by reducing the amount of time that the answer must be right.

8. Assume that the measurement is accurate within 30 meters 2DRMS with selective availability off.  (standard deviation = 15)
Selective Availability is discussed at 
http://www.csi-dgps.com/library/docs/articles/sa.htm  Note it is a controversial subject.

 

Find what the manufacturer's accuracy specification would be for RMS and CEP for the two blanks in the table.

Name of statistic

Probability

Fraction of time the answer is right

Error circle (meters) with selective availability on

Error circle (Meters) with selective availability off

2 s 2DRMS

.954

100

30

1 s RMS

.683

50

___

CEP

.5

34

___

 

Note that quoting an accuracy as CEP (wrong 50% of the time) and selective availability off are the easiest way for a manufacturer of inexpensive equipment to look better than expensive equipment to the uneducated eye.

 Equipment to be used in laboratories at Cal Poly

Collecting positions with the Trimble ProXL

Contouring a vineyard with the Trimble 4600LS.  Real time position corrections come from the base station located on the tripod in the right center of the picture.  The corrections are sent via the radio antenna at the right of the picture.

The Ph the field was measured using a grid pattern with the Trimble 4600LS.  The contours were drawn with Terramodel and added to ArcView with CivilView.  Terramodel and CivilView are products of Spectra Precision.  CivilView is listed on Spectra Precision web page under Spectra Precision Software Tools For ArcView   http://www.spectraprecisionatl.com/

The variable rate sprayer demonstration unit uses an Ashtech AgNavagator.  The real time differential correction is provided from a Coast Guard Beacon correction signal received on the second antenna.

 

Sources of Real Time GPS
Read http://www.navcen.uscg.mil/faq/dgpsfaq1.htm#How
go to http://www.navcen.uscg.mil/ADO/DgpsSelectStatus.asp
click on Coverage plot and select your area (pick one of the first five choices for an overview then pick your state for more complete information)

            Is your area covered? __________________

            What is the name of the site nearest you? ________________

For worldwide information see http://joe.mehaffey.com/dgps.htm
http://www.csi-dgps.com/library/docs/articles/dgps.htm
and select 
A WHITEPAPER ON DGPS (IN PDF) to see what is planed for the future for your area. (not required)

Reading the specifications of the equipment used in the laboratory at Cal Poly.
Read http://www.redhorsetech.com/aggps/NFdefinitions.html

Download the file agnav596.pdf.

 Go to http://www.trimble.com/ and click Mapping and GIS Systems
Click the work products in the second paragraph.
At the side bar click GPS Pathfinder Pro XR/XRS
Download the Product Datasheet: PDF

Click home on the side and Land Survey 
Click on "Trimble GPS Total Station® 4700"
On the side click on 4600LS
Download the Product Datasheet: PDF

 This exercise is to help you read the specification sheet on the equipment.

For the AgNavigator look at the section Measurement Features
Single point position:* 25m(rms)
*with Selective Availability off
This says that 68.3% of the time that the receiver with no differential corrections will be within 25 meters of the correct location when selective availability is off.  This is consistent what was estimated above.

 DGPS Position: < 1 m (rms, PDOP<4)
With differential correction (from the Coast Guard for example) the accuracy should be better than 1 meter 68.3 % of the time.  The satellites should not be all in one part of the sky.  See page 26 of your text.

 Time to first fix: Typically>1 minute.  
When you turn on the unit it does not know where the satellites are.  It probably knows approximately where in the world it is because of previous readings.  Under these conditions it will be able to obtain it location in 1 minute or less.  For some receivers, if the previous location is not stored in the instrument, it can take a long time to initialize.

Tracks 12 satellites simultaneously.
This should be enough to track all satellites in view.  Inexpensive instruments may track only one or two satellites and have to cycle between satellites causing slower response.

Carrier-smoothed pseudo range.
The basic measurement is made with the C/A code.  The carrier wave is used to improve the measurement but the integer number of carrier waves to the satellite are not measured.

Half second update rate
The GPS position is updated ever one half second.  At 5 miles per hour you would move 3.66 feet in one half second.  Note your speed will help determine your require update rate

  

Real time Differential (RTCM 104)
This is the format required for a unit supplying the correction information.

In group discussion, discuss your interpretation of the specifications of the other two GPS receivers.

 

This site is funded by grants from USDA and ARI, and developed by California Polytechnic State University, San Luis Obispo, California State University, Fresno, and the University of California, Davis
Web site development by ATI-Net
Distribution of this material is under the GNU General Public License
http://www.gnu.org/copyleft/gpl.html