Talk:Global Positioning System

Relativistic corrections again

User:Marquinho keeps deleting any mention of relativistic time corrections based on their interpretation of a 1996 paper from Aerospace Corporation. I agree that the section needs to be improved and cited, but I object to both their arguments and their methods. First, the 1996 paper states that the relativistic corrections it dismisses would need to be considered for space-based receivers as well as for receivers that are not subject to selective availability and therefore need better accuracy. Second, the paper does not claim that relativistic corrections aren’t needed; it states that the mixture of reference frames in use, plus a relativistic time correction applied to satellite time, result in terms cancelling sufficiently for Newtonian mechanics to suffice for the real-time calculations in ground-based receivers. Meanwhile, both the size of the time correction in the satellite’s clock, and the mixture of reference frames chosen, were derived from relativistic considerations. What the 1996 paper claims is that the formula used by receivers for real-time interpretation of the satellite signal does not incorporate relativistic effects. That is very different than claiming that relativistic effects can be ignored in GPS. This paper even refutes a claim in the 1996 paper, as well as emphasizes the many ways that relativistic effects come into play in GPS. And finally, I object to this business of just deleting instead of discussing and improving. I would appreciate some collaboration from other editors. Strebe (talk) 19:16, 8 December 2022 (UTC)

There is no doubt relativistic corrections to the **radio frequency** - but it doesn't take a reference to know that no corrections are needed to timekeeping (though I did include a reference)

The atomic clocks themselves, when produced, have a delta, i.e. they don't all run at the same rate. That delta is greater than the delta from time dilation. So the GPS system already needs to account for the delta between clocks regardless of time dilation & as such adjustments due to this are unnecessary Marquinho (talk) 18:55, 9 December 2022 (UTC)

The paper that Strebe just cited is quite clear that it is referring to shifts in the frequency of the **clock**, not the radio frequency. MrOllie (talk) 20:51, 9 December 2022 (UTC)

The issue is that the paper Strebe points to is a random paper by a single physicist, whereas the one I linked to is literally a manual by employees of the GPS Joint Program Office, giving direction on how to implement in high-speed aircraft.

Aren't the people who actually created the system & understand the software for it more trustworthy than some random person writing a paper assuming how it works? Marquinho (talk) 22:33, 9 December 2022 (UTC)

And even the paper included by Strebe, in fact, says that the clocks are synchronized on the ECI coordinate system:

"For the GPS it means that synchronization of the entire system of ground-based and orbiting atomic clocks is performed in the local inertial frame, or ECI coordinate system"

In other words, as I mentioned previously, because each clock can have a different drift (not all atomic clocks are exactly the same), they already have a synchronization mechanism which synchronizes the clocks and as such whether or not there's time dilation is irrelevant as the clocks are synchronized anyway Marquinho (talk) 22:47, 9 December 2022 (UTC)

An additional reference:

"The master clock for GPS is provided by the United States Naval Observatory (USNO). In two facilities, with an ensemble of masers and cesium and rubidium atomic clocks, the USNO keeps GPS clocks accurate. Without intervention, GPS clocks could drift nanoseconds a day, giving errors unacceptable for navigation"

In other words, GPS was designed with the expectation that satellite clocks would drift & they are synchronized by the base station. So, even if time dilation applies, or clock error, etc. it wouldn't matter, because the clocks are synchronized to ground station & as such there's no need for complex time dilation calculations. Marquinho (talk) 23:03, 9 December 2022 (UTC)

You've got it backward, Wikipedia considers peer-reviewed sources (especially more recent peer-reviewed sources) as the best to follow. And while it is true that clocks do need to be synchronized periodically, that does not actually contradict the sources we have that detail the corrections needed due to relativity. MrOllie (talk) 23:25, 9 December 2022 (UTC)

So the fact that both sources give an exact measurement for the internal frequency to compensate for relativity (10.22999999543 MHz) but neither gives an exact measurement of the delta in time drift doesn't at all give you pause?

In other words, if the source was accurate, would it not give an exact measurement, instead of some hand-waving? Especially when it does give an exact measurement for the delta in radio frequency? Marquinho (talk) 19:23, 10 December 2022 (UTC)

And yet again I'll make the point that this wikipedia article contradicts itself, here saying that time dilation is calculated for (without an exact value of how the atomic clock reading is adjusted to account for this), while at the same times saying: "The GPS satellites carry very stable atomic clocks that are synchronized with one another and with the reference atomic clocks at the ground control stations; any drift of the clocks aboard the satellites from the reference time maintained on the ground stations is corrected regularly"

In other words, the system is designed with the expectation that the clocks will drift & as such has auto-correction with the reference time on the ground stations Marquinho (talk) 19:29, 10 December 2022 (UTC)

As of '96, perhaps. Changes have been made to improve accuracy since then - all of the satellites that were used for GPS when your source was written have been decommissioned and replaced. At any rate, wrangling about details like this is not really what we should be doing on Wikipedia (WP:NOR). That the clocks need to be synchronized is irrelevant. If no corrections are needed for relativity, you should be able to present sources that say exactly that - and given that we already have sources that say the opposite, what you present will need to be more numerous, more authoritative and recent. MrOllie (talk) 19:38, 10 December 2022 (UTC)

That's not how wikipedia should work "someone said this without proof, so until you give proof against it, it stays" - again, what is the exact value of the time drift?

Any reliable source would be able to give an exact value by which the satellite atomic clocks are adjusted and why this would be necessary when the GPS system already has time synchronizing to base station.

If a statement is going to be made about how GPS adjusts for time drift from time dilation, should it not also state what the value is & why the built-in synchronization mechanism doesn't suffice? Marquinho (talk) 19:45, 10 December 2022 (UTC)

That's a straw man argument. The citations in the article are the proof. You want to remove well-sourced content based only on your own say-so. That is not what we do here. MrOllie (talk) 19:46, 10 December 2022 (UTC)

Not at all - what I'm saying is that there are 2 sources, both reliable, which conflict & contradict each other

The one from the GPS joint program office says that adjustments to drift due to relativity are not necessary

The one by Neil Ashby:

1) Does not give an exact value of the time drift adjustment

2) Does not justify why it's necessary when the clocks synchronize with the base station (which, again, is a well-source fact)

As such, I am calling into question the Ashby article & making the case that the GPS Joint Program Office article has more sources (in regards to time synchronization with base station) to back it up Marquinho (talk) 19:55, 10 December 2022 (UTC)

They only contradict each other if you engage in original research - making inferences that are not actually in the sources. I contend that they do not contradict at all. If you want to make this argument, you need sources that actually contradict, by stating something plainly like 'no correction due to relativity is needed'. MrOllie (talk) 19:58, 10 December 2022 (UTC)

So, again, what is the actual numerical value being used for adjustments? A clear value is given for frequency adjustments.

If this source is so trustworthy, what is the adjustment value? And if one cannot be given, how can the source be trusted? Marquinho (talk) 20:01, 10 December 2022 (UTC)

This question is also irrelevant. If a source says 'the sky is blue' it does not have to specify the precise shade. MrOllie (talk) 20:02, 10 December 2022 (UTC)

We're not talking about verifiable facts, such as the color of the sky - more importantly, you're making the claim that if a study says "actually, that study that said the sky is blue, is incorrect - the sky is red" - that you would take that update to wikipedia!

We're talking about contradicting sources, one which calculates the value & claims it's irrelevant, the other which disputes that claim, claims it's necessary, without giving a quantitative value nor having an explanation as to why the corrective system of time synchronization doesn't suffice

Surely basic reason counts for something? Or am I to believe that if a statement doesn't make logical sense, that as long as it got a peer review, one can apply the changes, no matter how irrational the claim is? Marquinho (talk) 23:28, 10 December 2022 (UTC)

Or am I to believe that if a statement doesn't make logical sense, that as long as it got a peer review, one can apply the changes, no matter how irrational the claim is? Not how I would phrase it - but yes, that is how Wikipedia works. We value the judgment of reliable sources above that of individual Wikipedia editors. No amount of 'basic logic' can really overrule a high-tier source such as a peer reviewed publication. Also, once again, I do not agree that the sources are in conflict. MrOllie (talk) 23:38, 10 December 2022 (UTC)

Wiki Education assignment: Technology and Culture

This article was the subject of a Wiki Education Foundation-supported course assignment, between 21 August 2023 and 15 December 2023. Further details are available on the course page. Student editor(s): Adomale (article contribs).

— Assignment last updated by Thecanyon (talk) 05:33, 12 December 2023 (UTC)

Lead section too long warning

It's four sentences long. MOS:LEADLENGTH says remove. nogamehere1 (talk) 06:35, 10 December 2023 (UTC)

Wiki Education assignment: Technology and Culture

This article is currently the subject of a Wiki Education Foundation-supported course assignment, between 19 August 2024 and 7 December 2024. Further details are available on the course page. Student editor(s): Pvirden (article contribs).

— Assignment last updated by Pvirden (talk) 02:01, 29 August 2024 (UTC)