otterbuff Posted July 23, 2002 Share Posted July 23, 2002 [Not sure how to title or ask this one :-)] If you placed two (or more) units on the ground right next to each other, what is the liklihood that they will read the same? In other words what is the probability that there will be a significant difference in their location readouts at the exact same time? What if you used identical units? Or same manufacturer but different units? Or different manufacturers? If you placed your single unit on the ground, let it get as accurate a fix as possible and then turned it off and then back on so that it would get another fix. What is the probability that the two fixes will be the same? (Yes, I KNOW I can test out my second question, but I am daydreaming at work now Quote Link to comment
+geospotter Posted July 23, 2002 Share Posted July 23, 2002 Could I ask one more (related) question? Do different manufacturers use the same formulas for calculating locations? Quote Link to comment
+mrp Posted July 23, 2002 Share Posted July 23, 2002 I've heard that placing units right next to each other may cause them to interfer because the few microwatts of RF leaking out of the oscilator circuts will cause problems. However, when I've had two units within a few feet of each other, they seemed to agree on the location down to +/- 1/1000th of an arcminute of Lat/Long, which is as much agreement as you can expect. As to whether different manufacturers use the same "formula", well.. yes and no. They're all trying ot solve the same problem, which is delta_x_n^2 + delta_y_n^2 + delta_z_n^2 - (c*delta_t_n)^2 = 0 where n represents the index of the satellite (there are a few other corrections in there, but that's the gist.) However, the algorithm the manufacturer uses to solve this problem is a closely guarded trade secret. I imagine that the different approaches will not greatly affect accuracy directly, but will instead have an effect on how quickly the answers converge (I'm assuming an algorithm of sucessive approximation), and how well they deal with occsionally spurious data from bad signals, reflections, etc. -- Mitchsuko [This message was edited by Pneumatic on July 23, 2002 at 09:25 AM.] Quote Link to comment
+majicman Posted July 23, 2002 Share Posted July 23, 2002 quote:Originally posted by Pneumatic:(snip) had two units within a few feet of each other, they seemed to agree on the location down to +/- 1/1000th of an arcminute of Lat/Long, which is as much agreement as you can expect. (snip) However, the algorithm the manufacturer uses to solve this problem is a closely guarded trade secret. I imagine that the different approaches will not greatly affect accuracy directly, but will instead have an effect on how quickly the answers converge (I'm assuming an algorithm of sucessive approximation), and how well they deal with occsionally spurious data from bad signals, reflections, etc. -- Mitchsuko Pneumatic (AKA Mitchsuko), I agree with point 1 - My Mag 315 and others have usually been within 1/1000th DecMin when settled. Point 2 - I know that algorithm that Magellan uses (it has to do with "Product Moment Coeffecients of Correlation" - an old set of math formulas from my Math Analysis days...) and you are again correct, it is a ever-centering circle average algorithm. - Basic PMCC stuff here: http://www.stat.ohio-state.edu/~tjs/494/lec16.pdf Point 3 - Is Mitchsuko really Pneumatic! - Caught ya' didn't we! --majicman Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.