[ntp:hackers] Drift of position with NTP/GPS.

Hal Murray hmurray at megapathdsl.net
Wed Oct 8 18:50:09 UTC 2008


> Now, some 10 years ago, when we were first putting the ONCORE driver
> together,  I did a several day run to determine my position, averaging
> the positions I  got over that time, and have used that position ever
> since. 


> I recently drug out that graphics program (I was recompiling the
> graphics  library it depends on and wanted to see that it still
> worked) and have let it run for  two days.

> I find that my position has changed. The changes are 2-3m in each of
> the three directions, viz: 

I'll bet that is "normal".

Random suggestions....

Split your data into 1 day chunks and see what each one says.

Collect some more data and see what happens.

Are you discarding outliers?  I've seen some data from GPS units that was 
marked as "valid" but is way off.

Plot the data you have and see if you see any interesting patterns.  You can 
graph lat or lon as a function of time, or you can make a scatter plot of 
lat/lon dots.   If you draw lines between successive points, the outliers 
stick out.

With the scatter plot, it's pretty easy to eyeball where the center of the 
blob is.  How does that compare to what you calculate?


-- 
These are my opinions, not necessarily my employer's.  I hate spam.





More information about the hackers mailing list