[FOSS-GPS] rtklib: simple DGPS post-processing experiment

[Timo Juhani Lindfors]

Hello Tomoji,

as promised, here's a report of my attempts to use rtklib for
post-processing.

Hardware setup
==============

1) neo1 is an openmoko gta02V5.
2) neo2 is an openmoko gta02V7.
3) Both devices are outside on a wooden table.
4) The LCD of each device faces downwards.
5) The "up" direction of each device points to South-East.
6) Distance between the devices is one meter (give or take a few
centimeters).
7) neo2 is North-East of neo1.

Captured data
=============

http://iki.fi/lindi/gps/experiment/1/neo1.ubx
http://iki.fi/lindi/gps/experiment/1/neo2.ubx

Post-processing steps
=====================

$ rtklib-convbin neo1.ubx
$ rtklib-convbin neo2.ubx
$ rtklib-rnx2rtkp -p 1 -l 60.217480929 24.393206102 71.2617 neo1.obs neo2.obs neo1.nav > neo1-neo2.pos
$ head -n20 neo1-neo2.pos
% program   : rnx2rtkp ver.2.3.0
% inp file  : neo1.obs
% inp file  : neo2.obs
% inp file  : neo2.nav
% obs start : 2010/07/22 06:25:13.0 GPST (week1593 368713.0s)
% obs end   : 2010/07/22 08:29:45.0 GPST (week1593 376185.0s)
% ref pos   : 60.217480929   24.393206102    71.2617
%
% (lat/lon/height=WGS84/ellipsoidal,Q=1:fix,2:float,4:dgps,5:single,ns=# of satellites)
%  GPST                  latitude(deg) longitude(deg)  height(m)   Q  ns   sdn(m)   sde(m)   sdu(m)  sdne(m)  sdeu(m)  sdun(m) age(s)  ratio
2010/07/22 06:29:01.998   60.217692722   24.393138007    61.6161   4   4   2.3673   0.9796   2.9740  -1.1745  -1.0261  -0.5434   0.00    0.0
2010/07/22 06:29:02.998   60.217716772   24.393110643    65.7936   4   4   2.3680   0.9798   2.9741  -1.1751  -1.0262  -0.5407   0.00    0.0
2010/07/22 06:29:03.998   60.217727627   24.393090583    69.9502   4   4   2.3688   0.9801   2.9742  -1.1758  -1.0263  -0.5379   0.00    0.0
2010/07/22 06:29:04.998   60.217727239   24.393083772    73.7900   4   4   2.3695   0.9803   2.9744  -1.1764  -1.0265  -0.5352   0.00    0.0
2010/07/22 06:29:05.998   60.217700360   24.393116329    72.5172   4   4   2.3702   0.9805   2.9745  -1.1770  -1.0266  -0.5324   0.00    0.0
2010/07/22 06:29:06.998   60.217646591   24.393137804    75.9859   4   4   2.3709   0.9807   2.9747  -1.1776  -1.0267  -0.5296   0.00    0.0
2010/07/22 06:29:07.998   60.217605988   24.393149454    77.8729   4   4   2.3717   0.9809   2.9748  -1.1783  -1.0269  -0.5268   0.00    0.0
2010/07/22 06:29:08.998   60.217573994   24.393164286    76.2576   4   4   2.3724   0.9812   2.9750  -1.1789  -1.0270  -0.5240   0.00    0.0
2010/07/22 06:29:09.998   60.217562704   24.393172695    76.4654   4   4   2.3731   0.9814   2.9751  -1.1795  -1.0271  -0.5212   0.00    0.0
2010/07/22 06:29:10.998   60.217558401   24.393176600    73.0197   4   4   2.3738   0.9816   2.9752  -1.1802  -1.0272  -0.5183   0.00    0.0

Evaluation
==========

Scatter plots of neo1.pos, neo2.pos and neo1-neo2.pos are at

http://lindi.iki.fi/lindi/gps/experiment/1/scatter.html

Histograms of distance to average (simple sqrt((lat-lat_avg)**2 +
(lon-lon_avg)**2)) are at

http://lindi.iki.fi/lindi/gps/experiment/1/histo.html

Open questions
==============

1) Is it normal that neo1-neo2.pos has 1380 points while both neo1.pos
and neo2.pos have over 7200 points?

2) Did this really improve anything? ;-) Since the number of points is
different I am not quite sure how to interpret the scatter plots.

-Timo