I wonder how fast the GPS position estimates generally change over time for a stationary receiver. I want this information to form some idea around what the sampling rate needs to be to estimate the distance of two GPS receivers that are not synchronized.
5
-
2Theoretically, wouldn't it be the data collection rate of the receiver, say 1 Hz? I don't see how it could be any faster than that.– Stu SmithCommented Jul 27, 2021 at 5:06
-
Does it mean the faster we sample, the faster the estimates will vary as the error in the estimates is totally random? If so, I guess to have a good estimate of the distance between two receivers, they should get GPS fixes at the same time.– RezaCommented Jul 27, 2021 at 11:53
-
1I don't think that the errors are "totally random". Rather, they are the sum of a variety of errors that are propagated between the time a signal is generated by a satellite and subsequently received. These errors - from atmospheric conditions to receiver noise - are generally well understood, not random.– Stu SmithCommented Jul 27, 2021 at 13:50
-
1rtklibexplorer recently wrote a blog post with some plots that illustrate fluctuation over a 24 h period: rtklibexplorer.wordpress.com/2021/07/14/…– ErikCommented Aug 2, 2021 at 14:28
-
Maybe you are looking for how autocorrelated measurements are so you can sample slow enough to ignore the autocorrelation? It really would depend on the receiver, antenna, the constellation of the satellites visible in the location of interest, and the processing that the GPS does on the data, and the accuracy of the estimate you want to produce.– Dave XCommented Aug 10, 2021 at 4:10
Add a comment
|
1 Answer
This will depend heavily on the GNSS receiver which is used and what positioning algorithm is used use. Older receivers often determined each position based only on measurements made since the previous position was generated, so they jump around significantly on a second-by-second basis. Newer receivers tend to use more advanced positioning (insert the term "Kalman Filter" or similar mumbo-jumbo to add technical credibility if you like) so every position also depends on previous states, so positions tend to drift slowly rather than jumping. What GNSS receivers are you using, are you using real-time corrections, are both of your receivers stationary?
