If you go onto alt.sci.satellite-nav, there are many technical threads about why averaging GPS position readings has its limitations. It isn't like averaging an electronic measurrment where you are filtering random noise. Or maybe it is like an electronic measurement where you have to deal with drift.
-----Original Message----- From: John Nelson <[email protected]> Sender: [email protected] Date: Thu, 24 Jan 2013 16:13:36 To: <[email protected]> Reply-To: Discussion of precise time and frequency measurement <[email protected]> Subject: [time-nuts] Lady Heather survey accuracy I hesitate to ask this question on here since it's perhaps more one for position-nuts rather than time-nuts. My excuse is that it involves a Thunderbolt ;-) I've been a happy user of a Thunderbolt for a while as an accurate time and frequency source but until recently I hadn't got round to mounting it in a proper enclosure with its PSU and finding a permanent home for the result in the office. Having done so I asked Lady Heather to carry out a 48h 'precision survey' and got a result which although close to that done when I first acquired the unit was about 10m different. I then carried out two further surveys which gave similar disparate results. LH reports latitude and longitude results to eight decimal places, which in principle suggests centimetre accuracy. I don't know enough about the finer points of GPS to know whether this is actually achievable over a 48h period but the results I'm seeing suggest that it's not. So the question amounts to this -- how reliable is the position produced by LH after a 48h 'precision survey' and what level of accuracy is achievable in practice? I'm guessing it's more like four places of decimals than eight. John _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
