UNAVCO Home UNAVCO Home
   |    |   |  
UNAVCO Home UNAVCO Facility

Science & Technology Development & Testing
Receivers & Antennas
Testing AMCS
Communications & PC Configuration
Communications
Power
Boot Bar Testing Equipment Power Table Wind Power Publications 		The Effect of Antenna Covers On GPS Baseline Solutions
[Next] [Previous] [Top]

The Effect of Antenna Covers On GPS Baseline Solutions

Data Processing and Analysis Strategy

Individual data files were translated into RINEX format and processed using the Bernese Processing Engine (BPE). Gross outliers were detected using pseudorange observations on the zero difference level. Observations with pseudorange point positioning residuals greater than 100 meters or more than 5 times the RMS of all the residuals were deleted. Pseudorange observations above 15 degrees elevation angle were used to compute receiver clock corrections. Cycle slip repair and carrier phase editing were done on single difference files using triple difference observations.

Carrier phase data were edited down to nine degrees elevation. Integer phase biases (ambiguities) were introduced for data gaps longer than five minutes and where cycle slip repair was not possible. For each baseline, multiple solutions were computed. The various processing runs are summarized below.

  • L1 phase solution, all observations were used (30 second sampling), and ambiguities were resolved.

  • L2 phase solution, all observations were used, and ambiguities were resolved.

  • L3 (ionosphere free linear combination of L1 and L2 carrier phase observables) phase solution, all observations used, ambiguities from L1 and L2 processing were introduced.

  • L1 phase solution, with hourly troposphere estimate at one of the two stations, all observations were used, ambiguities from L1 phase solution were introduced.

  • L2 phase solution, with hourly troposphere estimates at one of the two stations, all observations were used, ambiguities from L2 phase solution were introduced.

  • L3 phase solutions, with hourly troposphere estimate at one of the two stations, all observations were used, ambiguities from L1 and L2 phase solution were introduced.

In addition to the above processing schemes, baselines were computed with three different elevation cutoff angles (10, 15, and 20 degrees). This report summarizes the results by presenting only the L3 solutions with and without additional troposphere parameter estimation. These two solutions represent the two most common ways that high accuracy GPS baseline solutions are computed. Also, the errors in these solutions are products of the errors in the other solution types.

After processing the GPS data, the baseline components were compared to a "ground truth". Generally, conventional surveying techniques provide measurements accurate to a few millimeters. For the type of antenna tests conducted here, it would be desirable to compare the solutions to a ground truth that is precise to less than one millimeter. This is accomplished by computing a ground truth using GPS. For short baselines less than a few tens of meters, a network of the same type of GPS receivers and antennas set up to minimize multipath, and processed using only the L1 phase observations will produce results repeatable to a less than 1 mm. This is as good, if not better than conventional surveying results. UNAVCO has conducted this type of calibration ground truth survey for the marks use in these tests. In addition, the calibration survey has been repeated several times over the past year and the agreement between surveys is better than one millimeter for most of the monuments.

The Effect of Antenna Covers On GPS Baseline Solutions - 17 JUN 1997
[Next] [Previous] [Top]

Last modified Thursday, 17-Nov-2005 03:58:00 UTC
 

Home | About Us | Contact Us | Support | Search | Facility | PBO | Education & Outreach

Comments: webmasterATunavco.org
© 2009 UNAVCO, Inc.