Does averaging improve GPS accuracy?

Hi!

I have an application for the PPC that gets the location using an I-
Blue 737 Bluetooth GPS. I get the GPGGA string and parse the latitude
and longitude. I throw the values with a bad fix number, less than 4
satellites and HDOP values above 6. Would averaging 10 GPS readings
over 10 seconds improve the accuracy of the latitude and longitude?

Thanks a lot!
Anne

No. There is a worthwhile improvement in accuracy if you average over
about 10 minutes. See http://users.erols.com/dlwilson/gpsavg.htm

HTH, Glenn.

Anne wrote:

Probably not, Anne.
http://edu-observatory.org/gps/gps_accuracy.html

Thanks for the quick responses!

I took 100 GPS measurements and looked at the data two ways:

1. Plotted the data using a 10 point running average
2. Plotted the data with no averaging

The data with the 10 point running average is much tighter (the std
deviation is smaller) than with no averaging. Am I just improving
the precision not the accuracy? Accuracy can only improved over
longer periods of time?

Thanks a lot!
Anne

Anne wrote:

The links to David L Wilson's pages (among others) are telling.
Also keep in mind that short term averaging will have some different
effects depending on the filtering implemented in the specific receiver
design. Accuracy must include a fixed reference of some kind.

Best regards Anne
-Sam