Talk:Global Positioning System/Archive 7

Archive 1

←

Archive 5

Archive 6

Archive 7

Archive 8

Archive 9

Links to commercial site

For the time being I have removed two links to a commercial site per wp:ELNO #5. I'm not sure whether saying to exeptionally disagree because the sources are excellent readings to make probably too long already a list of external links even longer by adding two links on a commercial site — even if the articles are really excellent. I think that by allowing this, we risk getting flooded with more manufacturers with websites having more technical articles. DVdm (talk) 13:40, 10 January 2011 (UTC)

I understand your concern. I am aware that the list of external links is already quite long, and in this regard have carefully considered the appropriateness of the links. My conclusion was that the benefits outweigh the concern about a commercial offering for the following reasons: The compendium document succeeds in providing an excellent introduction and overview of the technical concepts, an aspect in which the current article is a disgrace. Several of the current external links discuss very specific technical issues, and none of them provide a good overview or historical introduction. Finally, I think that wp:ELNO #5 does not exactly apply - obviously, any company has financial interests. However, the linked documents hardly contain any explicit advertisements or promotion. The document presenting the origins of GPS contains nothing additional but the copyright of the company. Of course I respect the opinion of others, so let's see what consensus we can get. Nageh (talk) 14:04, 10 January 2011 (UTC)

Good idea. I personally think that thay are good sources and so do you, but in the end, who are we to decide? :-) Cheers - DVdm (talk) 14:19, 10 January 2011 (UTC)

General Relativity

re: "In 1956 Friedwardt Winterberg proposed a test of general relativity using accurate atomic clocks placed in orbit in artificial satellites. To achieve accuracy requirements, GPS uses principles of general relativity to correct the satellites' atomic clocks."

Wouldn't be better to move this from "History" to the section "Error sources and analysis", along with a short discussion describing the use of general relativity to correct the atomic clocks? Psalm 119:105 (talk) 17:54, 16 January 2011 (UTC)

Maybe. I am not really happy with the current layout of the article anyway. I think it should start with an introductory (1) Principles of satellite navigation, where the basic concepts are described, then go on with (2) Development of the GPS, which discusses both historical development and more advanced concepts (including the influence of time drifting due to relativistic effects, which is significant for GPS applications, btw), then describe (3) System structure and (4) Applications, next (5) Technical specifications, and finally (6) Accuracy enhancement techniques (and (7) Other systems), leaving Navigation equations as an advanced topic in a separate article (with basics described in sections 1 and 2). Something like this. Nageh (talk) 18:58, 16 January 2011 (UTC)

Now why would anyone want to put in something as uninteresting as technical specifications yet leave out the most interesting part of the entire document, Navigation Equations. The mathematics used in the navigation equations is certainly not advanced. It is very basic and fundamental. The physics is very basic and fundamental. We certainly don't need to dumb down the article to such an extent that everything that in any way resembles mathematics or physics goes into an advanced section. RHB100 (talk) 20:30, 18 January 2011 (UTC)

This is nowhere near what I intended. Sorry, my bad. I was somehow assuming that Methods of solution of navigation equations would discuss advanced techniques, similar to section Accuracy enhancement and surveying, without taking a second look. I agree that these are basics that need to be described in the article. If not right in section (1), which was my actual intention, of the structure I proposed, then in a separate section (where best?). Still, advanced calculations are possible pertaining to accuracy enhancement techniques; they are certainly best discussed in separate articles (e.g., the GPS augmentation articles). Regarding technical specification, as a mathematician you may find them uninteresting yet they are an important aspect. I moved them pretty much to the end in my list, anyway. Hope that clarifies things. (Btw, I'm surprised you suspected me to dumb down articles. I have recently argued from the opposite position on the Scientific guidelines talk page.) Nageh (talk) 21:14, 18 January 2011 (UTC)

With regard to specifications, I think the opening paragraph describing the capabilities of GPS covers a good part these speicifications. Most of the remaining part can best be covered in an error analysis. Incorporating ""Error analysis for the Global Positioning System"" into the GPS main article would complete the technical specifications and at the same time make the article more interesting. RHB100 (talk) 19:38, 19 January 2011 (UTC)

The "error analysis" material was in this article (see this version). In October, the extensive discussion of relativistic effects was removed from this article and is now in Error analysis for the Global Positioning System. Its relevance isn't limited to error analysis, though. This article should include at least a summary of the applicability of relativity to GPS: that the satellites orbit so high and move so fast that relativistic effects are unusually large; that the system accuracy depends on precision timing to the microsecond level, so the tolerance for error is unusually small; and that, as a result of these factors, GPS system design must -- and does -- take account of relativity. We could mention that there are some smaller relativistic effects (velocity and elevation of receiving stations, for example) that aren't large enough to worry about, and refer the reader to Error analysis for the Global Positioning System for further detail on that subject. I agree with Psalm 119:105 that such a section could also include the 1956 proposal and the actual test that was conducted.

Many aspects of the treatment of relativity were discussed in great detail over the summer in this thread and the next few after it in the archive. JamesMLane t c 01:17, 21 January 2011 (UTC)

The article, Error analysis for the Global Positioning System, contains material which is even more important for GPS than the material on relativity. For example the material on the calculation and derivation of geometric dilution of precision (GDOP) is also contained in this article. I think the contents of the article, Error analysis for the Global Positioning System, should again be made a part of the GPS article. RHB100 (talk) 19:53, 21 April 2011 (UTC)

Satellite positioning inconsistency in figure

The "visual example of the GPS constellation in motion" figure seems to be inconsistent with the article. The article states "the angular difference between satellites in each orbit is 30, 105, 120, and 105 degrees" but the angular difference in the figure is 90 degrees for all satellites.

This inconsistency should be resolved one way or the other. I have also added an equivalent note to the discussion page for the figure. 134.134.139.74 (talk) 19:50, 17 February 2011 (UTC)

I don't know how you are able to determine that "the angular difference in the figure is 90 degrees for all satellites". First of all the figure provides a two dimensional depiction of three dimensional motion. It is hard to measure angles accurately in this situation. Second the satellites are continually moving. There are not only difficulties in measuring the angles between moving objects in a given plane but also the distraction of the motion in other planes. But you may have found a way to measure the angles in the figure. If so, I would like to see an explanation of how you do it. RHB100 (talk) 20:21, 22 April 2011 (UTC)

Doppler shift discovered with Sputnik? I don't think so.

As of 4/5/2011 the article reads:

A team of U.S. scientists led by Dr. Richard B. Kershner were monitoring Sputnik's radio transmissions. **They discovered that,** because of the Doppler effect, the frequency of the signal being transmitted by Sputnik was higher as the satellite approached, and lower as it continued away from them. They realized that because they knew their exact location on the globe, they could pinpoint where the satellite was along its orbit by measuring the Doppler distortion (see Transit (satellite)).

The phrase "They discovered that," is completely out of place. These scientists did not have to receive Sputnik to "discover" Doppler shift of radio signals. Doppler is taught in applied math 101. If no Doppler shift had been observed, then *that* would have been a significant discovery.  ;-)

Therefore, please, someone, move this phrase alone to the next sentence. I suggest: "A team of U.S. scientists led by Dr. Richard B. Kershner were monitoring Sputnik's radio transmissions. Because of the Doppler effect, the frequency of the signal being transmitted by Sputnik was higher as the satellite approached, and lower as it continued away from them. Because they knew their exact location on the globe, they discovered that they could pinpoint where the satellite was along its orbit by measuring the Doppler distortion (see Transit (satellite))."

-Al Roxburgh —Preceding unsigned comment added by 98.174.141.194 (talk) 21:48, 5 April 2011 (UTC)

Multidimensional Newton-Raphson calculations

As a licensed professional engineer with advanced engineering degrees from both the University of Arkansas and UCLA, I would like to point out that there are both advantages and disadvantages associated with the Multidimensional Newton-Raphson approach. As the licensed professional engineer who documented this method for Wikipedia I am uniquely well qualified to make this statement. It is generally true that all methods have advantages and disadvantages. The fact that a method works well certainly provides no exception. In the case of a one dimensional root finding method you know for a continuous function that when positive and negative values are found a root lies somewhere in between. For a multidimensional root finding method, you never know in the process of iteration that a root even exists except in the unusual case that you hit upon the exact solution. Almost all methods of solution occasionally fail. It is theefore important that we be aware of all disadvantages. We can make a judgement that it might be wise to disregard the disadvantage but we should never make the mistake of assuming that a disadvantage does not exist. RHB100 (talk) 01:35, 10 May 2011 (UTC)

The sweeping statement that "There are no good general methods for solving systems of more than one nonlinear equations." is almost vacuous and therefore nonsensical and irrelevant. Multidimensional non-linear equations are routinely being solved quite dependably in a host of engineering environments. For the problem on hand there does not seem to be a problem in the application of NR, especially since a good starting value is easily available. Perhaps the most pessimistic that can be said in this context is "there is no guarantee of NR's convergence in all cases". −Woodstone (talk) 06:30, 10 May 2011 (UTC)

The statement is a direct quote from the book "Numerical Recipes". It is obviously true since you cannot bound a multidimensional root. Nothing in the "Numerical Recipes" statement of this disadvantage contradicts what you have said after your first sentence. And on the other hand nothing you have said contradicts this disadvantage. You need to come up with a sourced statement which backs up your statements above and add this as an advantage of this method. The statement "there does not seem to be a problem in the application of NR" may well be true. But you have got to come up with a source. Such a source would not eliminate the disadvantage but it would be an advantage that shoulod be listed' RHB100 (talk) 02:49, 11 May 2011 (UTC)

Even if "there are no good general methods for X" would be true, it does not imply that "there is no specific good method for a specific case of X". Your conclusion is invalid. Since NR is used in many implementations of GPS, it is evidently a good method for it. See for example " Determination of GPS receiver position using Multivariate Newton-Raphson Technique for over specified cases, where NR is concluded to be as good as Bankcroft's method. −Woodstone (talk) 08:24, 11 May 2011 (UTC)

I agree with you Woodstone that the paper you cite indicates an important advantage of the multidimensional Newton-Raphson method. So important that I have added this advantage to the relevant section in the GPS article. I thought about waiting to let you add this advantage but I decided it was so important that I should add it without further delay. RHB100 (talk) 20:07, 11 May 2011 (UTC)

Glad you appreciate the cite. Your latest addition makes the view rather contradictory: like "it cannot work, but performs as well as the best algorithm". Anyway, we should clarify that the NR is not applied to the direct equations, but to a least squares optimisation. −Woodstone (talk) 10:38, 12 May 2011 (UTC)

Listing both the advantages and disadvantages is not contradictory. It is instead just reallity. On the one hand it has certain advantages but on the other hand it has disadvantages. I think your statement, "it cannot work", is a misinterpretation of the disadvatage. RHB100 (talk) 23:37, 12 May 2011 (UTC)

In what cases do any of these methods actually fail to converge? Using a simple least squares method, you can literally set the initial guess to {0, 0, 0} and still get converges in 2-3 iterations. The linear range of the equations is sufficiently large as to be extremely robust, in my experience, and I've only seen a failure when something is grossly wrong in other ways (e.g., bad hardware, massive iono disturbances, &c.). I'm concerned about implying to the lay reader that these solution methods are somehow unreliable. siafu (talk) 02:38, 14 May 2011 (UTC)

Well the language used certainly does not suggest that the multidimensional Newton-Raphson method is somehow unreliable. However, if you can add a sourced statement regarding the reliability of this method, I think it might be a useful addition. RHB100 (talk) 03:14, 17 May 2011 (UTC)

Solving the location in X,Y and Z or any other variables is multidimensional. Solving the location using a digital computer as all gps systems do is numerical. So all solutions do a numerical multidimensional solution. Even implementations of algebraic solutions are implemented as numerical solutions. So if there is a 'general' disadvantage it affects all digitally implemented solutions tackling this problem. Crazy Software Productions (talk) 13:22, 3 June 2011 (UTC)

An addition was made to split off methods for more than 4 satellites. That is not quite correct. Even in the case of 4 satellites, the system is overdetermined. There are 4 equations and 4 unknowns, but that does not imply there is a solution. Actually, even is there is no clock error, other errors in the measurements or the model would make an exact solution almost sure not to exist. Three satellites determine two possible solutions. The fourth one will practically never match precisely. Instead of solving for equality, an optimisation needs to be done. There is no sharp distinction between using 4 satellites or more. −Woodstone (talk) 05:56, 7 June 2011 (UTC)

The above is not quite correct. First I am assuming a normal GPS receiver with a quartz clock. Normally a quartz clock has an accuracy of less than 1 in 1 000 000. So that clock can not be used for accurate positioning. Then 4 satellites are needed. There are 4 equations and 4 unknows. Mathematically (for the GPS situation) this will result in 1 or 2 exact solutions.

The clock is only used for timing the differences of the reception time.

1. One satellite does not give any position at all.
2. The solution for receiving the signals of two satellites is a plane with the shape of an hyperboloid.
3. The solution for receiving the signals of three satellites is the intersection of two hyperboloids. This is a closed or an open curve.
4. The solution for receiving the signals of four satellites is the intersection of th closed or the open curve with an extra hyperboloids. If the curve was closed this will result in two solutions. If the curve was open this will result in one or two solutions.

Mathematically these solutions are exact. In reality the errors in the signals (and timing between the signals) will give small positional errors, but mathematically there are exact 1 or 2 solutions.

2D solution. Using a fixed height (fixed distance from the centre of the earth). Three satellites put you on a curve. This curve will intersect the earth a two points. Allthough it is possible that the curve exactly touches the earth, this will be rare and only be for an infinite small time. So with three satellites and the earth there are mathematically two solutions. Only one of the solutions will be stable (Over time). (A curve that intersects with a sphere has to intersect at at least two points, one point can be considered the entry point and on point can be considered the exit point. Mathematically there are endless curves which can intersect a sphere at only one point, but this type of curve isn't the type produced by the intersection of two hyperboloids.)

3D solution. Again starting with the curve of three satellites. Introducing a fourth satellite. This introduces another hyperboloid the curve will intersect this hyperboloid in one or two points. Only one of the points will be stable.

When there are two solutions, the method to calculate these solutions does not alter that. With Bancroft both solutions are provided. With pseudorange calculation the first estimate will determine which point it comes up with. With pseudorange calculations the calculation will automatically fix on the stable point after a few seconds.

Four satellites is not an overdetermined system.

Crazy Software Productions (talk) 19:24, 18 June 2011 (UTC)

NAVSTAR name or acronym still unresolved

I may have been a bit hasty reverting the latest addition of NAVSTAR as being an acronym, or not. This was last discussed incompletely, and therefore unsatisfactorily here and here. Could the imperious authority "John Walsh" be a hoax? —EncMstr (talk) 20:18, 20 June 2011 (UTC)

The NAVSTAR acronym definition (NAVigation System Timing And Ranging) was created by Rockwell International Space Division as published the "GPS NAVSTAR Space Vehicle Description Handbook" when they manufactured the GPS satellites in Seal Beach, CA. It was the GPS JPO that started referencing the satellites as GPS vs. NAVSTAR GPS.Robapodaca (talk) 00:04, 21 June 2011 (UTC)

Cost

What were the cost of each semgment of the GPS? Who paid for it? Article does not answer this basic question!!! — Preceding unsigned comment added by 79.80.138.51 (talk) 19:01, 22 June 2011 (UTC)

There will probably never be a full accounting for the costs. It was 1) a U.S. government project, and 2) a military project. It might be possible to dig up some contributing costs, like the cost of launching a satellite, but any price quoted by NASA is highly questionable since government accountants are in a completely different world, where even the conversion rate to ours is unknowable; besides, they don't really have to account for their costs. That combined with various off-the-books costs for security and secrecy reasons mean that any hard number presented would be only the tip of the iceberg and attract various wild guess multipliers to get the true figure (probably varying from 50 to 1000). —EncMstr (talk) 19:27, 22 June 2011 (UTC)

That's a bit of a silly answer. It's not completely mysterious, just complicated, and someone would have to do a more detailed investigation. Most of the costs are not classified; for example, the recently announced contract for the new control segment gives a figure of $1.5 billion [1]. There will be some variability in estimates, because there remain elements of the payload that are classified, but this would account for a relatively small part of the budget. I'm not aware off the top of my head of any actual estimates, but I may be able to find something in a couple days. siafu (talk) 20:09, 22 June 2011 (UTC)

"L4 Band" and Ionospheric Correction

I have been looking for a reliable citation for the note in the 'Satellite Frequencies' section about the use of an 'L4' transmitted frequency for ionospheric correction.

I can find no independent reference to the existence of an L4 transmitted frequency; the most recent official 'Global Positioning System / Standard Positioning Service / Performance Standard' dated Sept 2008 (http://www.pnt.gov/public/docs/2008/spsps2008.pdf) makes no mention of L4, only L1 and L2 (and future L1C, L2C and L5 signals).

However I have found references to a calculated L4 signal in several papers about ionospheric correction of GPS signals where L4 is some kind of differential error signal calculated from L1 and L2. Example: 'Optimal Noise Filtering for the Ionospheric Correction of GPS Radio Occultation Signals' http://nldr.library.ucar.edu/repository/assets/osgc/OSGC-000-000-001-401.pdf

Is there some confusion here between a fictitious calculated signal (that by convention is referred to as L4) used purely for correction purposes and a real signal that does not actually exist?

Hughdel (talk) 17:20, 25 July 2011 (UTC)

As far as I understand there was an internal study in the 90's on how the signal structure of the GPS system could be enhanced, and part of this study suggested the introduction of a new carrier (L4), which should be made available to the civil sector and was also intended for ionospheric correction. Current research papers discuss alternate ways for carrying out ionospheric correction, but make reference to the L4 carrier by equally calling the linear combination of L1 and L2 as such. In other words, there are two different meanings for L4, one is the actual separate carrier from L1 and L2, and the other is the linear combination of L1 and L2 (i.e., a "virtual" carrier). Here is a reference that discusses the introduction of the L4 band: [2]. And here are some slides stating the frequency of the carrier: [3]. Nageh (talk) 18:33, 25 July 2011 (UTC)

The iono-free linear combination of L1 and L2 is almost always called L3, and not L4 (see Misra & Enge, e.g.). Prior to looking at your posted sources, Nageh, I've never seen the use of the term L4 anywhere. siafu (talk) 02:39, 26 July 2011 (UTC)

Images instead of text

The images used for some of the mathematics are not rendered well on my system. Part of the problem is that I have the pages enlarged somewhat to compensate for poor vision. Could someone replace these images with text? If that is not practical, it would help if the images were redone in .SVG. SlowJog (talk) 23:42, 6 October 2011 (UTC)

Before anyone changes this, try adjusting your Math rendering option of your user preferences on the Appearance tab (at the bottom of the page). I show six options in Firefox 7.0. If you are using an older browser or a non-standards complying one (like Internet Explorer pre-8.0 or so), consider upgrading. Be sure to WP:CLEAR your browser cache after saving a new setting. —EncMstr (talk) 23:56, 6 October 2011 (UTC)

The definition of variable should be corrected

Some definitions of variable and equations in this article should be corrected. The symbols of correct and apparent variables should be distinguished.

My suggestion is as follows: (1) $\,{\tilde {t}}_{\text{r}}$: the apparent time of signal reception, which is indicated by a GPS receiver. (2) $\,t_{\text{r}}$: the time of signal reception (unknown). (3) $\,(x_{\text{r}},y_{\text{r}},z_{\text{r}})$: the receiver position (unknown). (4) $\,p_{i}:=c({\tilde {t}}_{\text{r}}-t_{i})$: pseudorange, which is observed by a GPS receiver. (5) $\,b:={\tilde {t}}_{\text{r}}-t_{\text{r}}$: the clock advance of a GPS receiver (unknown). (6) relations: $\,t_{\text{r}}=p_{i}/c+t_{i}-b$. $\,(x_{\text{r}}-x_{i})^{2}+(y_{\text{r}}-y_{i})^{2}+(z_{\text{r}}-z_{i})^{2}=c^{2}(t_{\text{r}}-t_{i})^{2}$. Kkddkkdd (talk) 14:52, 18 September 2011 (UTC)

---

We already have a variable for the clock bias called b. Adding another variable for essentially the same thing would do harm rather than good. RHB100 (talk) 01:48, 9 November 2011 (UTC)

Undue weight on Bancroft

Three times clearly wp:UNSOURCED wp:POV phrases, inserted by user RHB100 (talk · contribs) here and here, about Bancroft's method ("the most important method of solving the navigation equations because it involves an algebraic as opposed to numerical method," and that it "has the advantage that it can be used for the case of four satelites or for the case of more than four satellites."), were removed or rephrased by 78.147.75.16 (talk · contribs) here and by myself (here and here). I have left a warning on the user talk page about insertion of wp:POV.

On the other hand, do we really need this in a separate section? Shouldn't we instead just mention the name "Bancroft" once in the section about the least squares method? - DVdm (talk) 08:11, 9 November 2011 (UTC) Surely this books search has a proper source that is more suitable than the current source, a student assigment, sitting here. For instance, this book source says that Bancroft, Krause, Abel and Chaffee, and Hoshen developed non-iterative closed-form solutions to the nonlinear GPS pseudorange equations. Rather than having the current section, based on a weak source and merely stating the obvious, shouldn't a short referenced mention along these lines be largely sufficient? Any other suggestions? - DVdm (talk) 12:37, 9 November 2011 (UTC)

Bancroft, S.; , "An Algebraic Solution of the GPS Equations," Aerospace and Electronic Systems, IEEE Transactions on , vol.AES-21, no.1, pp.56-59, Jan. 1985, doi: 10.1109/TAES.1985.310538. This is the original paper, found here, though not freely available online, this would clearly be the best source for the actual method. There's nothing at all wrong with using print sources, why not just stick with this? siafu (talk) 17:44, 10 November 2011 (UTC)

Good. I have added the source. No problem if it is not free - see wp:SOURCEACCESS. Thanks. - DVdm (talk) 18:08, 10 November 2011 (UTC)

The IEEE, like many (most?) professional societies, allows authors to post papers on the their personal or work web site, subject to certain restrictions. Has anyone asked Bancroft if he'd be willing to do this? LouScheffer (talk) 18:20, 10 November 2011 (UTC)

Bancroft method made even more neutral, current reference retained, new may be added

So long as the section "Bancroft's method" remains as it is now, I have no objections to removing "Bancroft's method is perhaps the most important method of solving the navigation equations". I don't think it was a violation of the neutrality policy since it is like saying, I think GPS is perhaps a better navigation system than dead reckoning. RHB100 (talk) 22:25, 9 November 2011 (UTC)

The referebce, http://www.macalester.edu/~halverson/math36/GPS.pdf, must be retained. It is the best written and most clear description of the Bancroft method of any free source I have been able to find on the internet. Additional references may be added but it is important that this source be retained. To judge the best source, it is necessary to read and comprehend the material. I don't believe DVdm has read the source material with sufficient comprehension to understand the Bancroft method. RHB100 (talk) 22:42, 9 November 2011 (UTC)

If I would not have a masters degree in mathematics (and another one in IT, by the way), my understanding of the Bancroft method would be just as irrelevant as it is now (see your talk page and this old WQA) . It does not matter whether you and I understand it, or whether you and I are qualified in anything. It matters whether Wikipedia text is wp:verifiable by means of wp:reliable sources. Trust me, I do understand the equations, and I am quailified to judge that, and they are good—very good. There is no question about that, but for Wikipedia our judgment is irrelevant. My point is just this: having noticed that this particular source is merely a student assignment (and therefore not peer-reviewed), albeit, as you said "on the website of a highly respected Canadian university", and although 100% OK for you and for me, I wondered whether we didn't have "Wikipedia-better" sources, i.e. published books or peer reviewed articles. Don't be afraid, I will not remove "your" source, but if someone removes it and replaces it with something more solid (in the Wikipedia sense), I will support that action. That said, what do you (and others) think about adding the statement I suggested in the previous section, together with the book-source I provided? - DVdm (talk) 17:24, 10 November 2011 (UTC)

Meanwhile, ref to original added - see prv section. - DVdm (talk) 18:09, 10 November 2011 (UTC)

The reference resides on the website of a Minnesota college, http://macalester.edu, not a Canadian university, 'highly respected' or otherwise. - Pirround (talk) 00:43, 7 February 2012 (UTC)

I don't know what statement in the previous section you are talking about. RHB100 (talk) 00:55, 11 November 2011 (UTC)

potential WSJ resource

Beijing Launches Its Own GPS Rival by Jeremy Page 28. December.2011 (page A9 in print). 99.181.153.29 (talk) 02:12, 29 December 2011 (UTC)

Possible problematic edits

I noticed that an editor who made several edits to this article may be linked to a PR company (see WP:COIN#Qorvis for background). Could someone with more knowledge of GPS check the edits that WeatherBug17 (talk · contribs) made to the article? Some of the content has already been removed but some remains. Thanks SmartSE (talk) 21:38, 30 December 2011 (UTC)

Similarly, this section in GPS modernization and the entirety of Joint Polar Satellite System was written by them if anyone has the time to check them over. SmartSE (talk) 21:44, 30 December 2011 (UTC)

Opening Comment

The comment, "This article's use of external links may not follow Wikipedia's policies or guidelines. Please improve this article by removing excessive or inappropriate external links, and converting useful links where appropriate into footnote references. (January 2012)", appears more likely to cause harm than to result in benefits. I recommend that it be removed. RHB100 (talk) 02:32, 22 February 2012 (UTC)

I agree with that comment, and recommend that it be followed in stead of removed. See wp:ELNO. - DVdm (talk) 07:30, 22 February 2012 (UTC)

Neutrality of LightSquared section

Whoever wrote the section documenting the LightSquared/Coalition to Save Our GPS controversy is clearly biased in favour of the coalition. These two lines particularly caught my attention: "In the face of demonstrated disruption to GPS operations, LightSquared has turned to a strategy of blaming GPS manufacturers for building receiving equipment which "..looks into their (LightSquared) spectrum". This, despite the fact that the spectrum in question was never envisioned as being used for terrestrial broadcast." The language of this section is not neutral enough for Wikipedia, and should be substantially rewritten or removed. --The Editor 18:17, 26 October 2011 (UTC)

Reality is somewhat biased on this issue. It's certainly true that LightSquared has been trying to place the responsibility on the GPS community by insisting that GPS users add filters to their receivers, and it's also true that this part of the spectrum was never envisioned as being used for terrestrial broadcast in that the entire neighboring band was used for satellite communications. siafu (talk) 18:48, 26 October 2011 (UTC)

I just wanted to add my thoughts - Siafu, the statement in this section is indeed factually correct, however, the wording is quite biased. I think phrasing like "LightSquared says that GPS manufacturers are to blame for building receiving equipment which "..looks into their (LightSquared) spectrum". They say that the GPS industry has had almost 10 years to prepare or object, but has chosen not to until recently. However, the spectrum in question was never envisioned as being used for terrestrial broadcast." would be much more neutral. I'm leaving this for someone else to make any actual changes, but wanted to help. If no one changes this in the near future, I will make it myself. (I'm leaving it for others because: a. I have never made a change before and am somewhat uncomfortable doing so, and b. because I'm not actually happy with my version, though I think it would be a big improvement from the existing wording.) — Preceding unsigned comment added by 65.46.168.178 (talk) 05:51, 13 December 2011 (UTC)

LightSquared's public statements have been technically preposterous, only sounding vaguly plausible to someone completely unfamiliar with radio frequency communications. It is flatly not possible to build a receiver with 100% rejection of out-of-band signals, and GPS receivers are particularly problematic because sharp cutoff filters necessarily have messy phase response, which corrupts the very precise timing required for GPS to work. Higher-resolution receivers, such as used in surveying and automated tractors, generally have the highest receiver bandwidths.

Later processing stages, particularly the spread-sectrum demodulation, do an excellent job of ignoring unwanted signals, but they are limited by the dynamic range of the front end. While another satellite signal (which is what the spectrum is reserved for) would not be a problem, LS wanted to use the frequencies for terrestrial broadcasts, which have far more power available, and are enormously closer to the receivers. The net recevied power would be a million to a billion times (60–90 dB) more powerful than the GPS signal.

This would saturate the automatic gain control circuity, and require a 20+-bit ADC (impossible to build at the necessary data rates) to digitize the GPS signal in the presence of LS's interference. For the receiver designer, this is almost the same as deliberate jamming, which civilian GPS receviers are not designed to resist.

LightSquared bought the spectrum cheap precisely because it was reserved for low-powered satellite applications. Then they said they wanted to use it for terrestrial transmitters. The FCC thought it was impossible to not interfere with existing GPS receivers, but LightSquared insisted it was. So the FCC let them try. And, because the laws of physics still apply, they fell on their faces.

It's like buying land zoned low-density and asking for permission to build a syscraper because you can do it without casting shadows on the adjacent houses. And then whining because, after being given an opportunity to demonstrate this miraculous ability and failing, your application for a zoning variance is denied. 71.41.210.146 (talk) 18:50, 28 March 2012 (UTC)

I've added an update on the current status of LightSquared before the FCC. LightSquared is discussed in two places in this article, which I think should be consolidated. Also the first paragraph's discussion of Part 15 oversimplifies the situation. The fact that consumer grade GPS receivers carry the must accept interference notice does not mean that anyone is free to radiate signals that interfere with GPS. --agr (talk) 11:23, 2 May 2012 (UTC)

You may in fact add that intentionally jamming in the GPS band is not just not allowed, but is in fact a felony in the US. siafu (talk) 16:12, 2 May 2012 (UTC)

SuperGPS

Added in

"* SuperGPS - a form of GPS for land-based navigation" at other systems.

A new seperate article should be made about it. The system proposed is allot more accurate (upto 10 cm), country-independant, impossible to jam and best of all, uses no satellites (easier to repair, lower costs). The only downside seems to be that it doesn't work at sea (needs substations).

91.182.205.137 (talk) 11:25, 12 April 2012 (UTC)

I removed this from the list of other systems, primarily because AFAICT it only exists as a proposal as of now, and does not appear to be even being funded for implementation. Also, from that powerpoint, it seems that this system is only for time and frequency transfer, and not navigation or positioning. As an aside, that proposal makes a number of claims about GPS that are quite misleading, so I would also challenge its reliability. siafu (talk) 19:32, 12 April 2012 (UTC)

Criticism

The article do not contain any criticism of the US GPS system. The article on Galileo comments that the GPS system can be shut down at the behest of the US president. Should this not be in there? Gnurkel (talk) 08:50, 11 April 2012 (UTC)

This criticism is a bit overblown, based solely on the fact that the US government could, in theory, shut down the GPS system if it decided to. Per a 1996 Presidential Policy Directive signed by President Clinton ([4]), since reiterated by all subsequent presidents: "We will continue to provide the GPS Standard Positioning Service for peaceful civil, commercial and scientific use on a continuous, worldwide basis, free of direct user fees." The more recent document ([5]), signed by President Obama, states: "[the United States shall] Provide continuous worldwide access, for peaceful civil uses, to the Global Positioning System (GPS) and its government-provided augmentations, free of direct user charges." So they've made it a matter of national policy that GPS will be free and not shut off, and there's the simple fact that disabling the GPS service without warning would be crippling to the economy, and certainly not in the national interests of the United States. So, sure, in principle it could be done, but this is simply a result of GPS being controlled by a single government, whereas Gallileo represents an international cooperative effort including both government and business; the EU could, in theory, kill Gallileo at any point just by withdrawing official support just as easily as the US government could disable GPS. siafu (talk) 15:37, 11 April 2012 (UTC)

I agree. The article makes it pretty clear that the system is controlled and operated by the U.S. which conceivably could switch it off at a whim. But it also makes it pretty clear that the system is heavily relied upon by many important users, U.S. and others. Anyone capable of understanding the major points of the article can easily infer the latter. —EncMstr (talk) 20:13, 11 April 2012 (UTC)

I feel like I should add that I think that comment DOES belong in the article about Galileo, as American military control of GPS was one of the stated motivators for the development of the EU system. This does not mean that intentional disabling of GPS is anything but a fringe possibility, against the stated intent of all parties responsible for its operation and maintenance, and definitely does not mean that it merits mentioning in this article. siafu (talk) 19:37, 12 April 2012 (UTC)

Has anyone pointed out that the Galileo consortium could also switch off Galileo at a whim? Why should this not be pointed out in the GPS article as a reasoning for folks to use GPS? The system is clearly controlled and operated by the EU. The EU could easily make a case that in order to preserve their safety in a time of war, that it would be necessary to shut down the Galileo system every bit as "off" as the US could turn off the GPS. 14 June 2012 — Preceding unsigned comment added by 132.3.57.68 (talk) 17:13, 14 June 2012 (UTC)

The reason that US ownership of GPS is important is because it was one of the stated reasons for the creation (funding, development, etc.) of Galileo in the first place. The reverse statement is not true, and it's not really an important reason to "rely" on GPS-- it is most likely that in the future, GNSS receivers will avail themselves of all the satellite signals that they can use, including GPS, Galileo, GLONASS, and potentially Beidou/COMPASS as well. siafu (talk) 18:19, 14 June 2012 (UTC)

correction! It will not be "shut down"! the implementation allow to turn back selective signal aka turning off the high accuracy for non military users and that can be turn on/off anytime in any part of the globe (selective. Iraq for example) without affecting other parts of the system. 72.185.61.209 (talk) 00:27, 25 July 2012 (UTC)

wayback of a dead link

wayback machine has an archive of a dead link, specifically, citation #40. I would change it myself but i'm only familiar with very, very simple wikipedia markdown x.x

link: http://web.archive.org/web/20081114182739/http://www.navmanwireless.com/uploads/EK/C8/EKC8zb1ITsNwDqWcqLQxiQ/Support_Notes_GPS_OperatingParameters.pdf — Preceding unsigned comment added by 76.67.36.49 (talk) 01:20, 16 April 2012 (UTC)

 Done I've updated it to the version from March 28, 2009. - M0rphzone (talk) 05:52, 5 May 2012 (UTC)

Damage

It appears that the section on how GPS operates has been substantially damaged by back-and-forth edits and reverts. Someone who is an expert on the material should review and provide references. Someone more comfortable with reading edit histories than I should, for the time being, pull text from an earlier version of the page so the section is at least not broken. This is not an appropriate place for people's opinions of how they think GPS works. — Preceding unsigned comment added by 192.76.175.3 (talk) 18:30, 24 July 2012 (UTC)

The experts are providing references.LouScheffer (talk) 20:33, 24 July 2012 (UTC)

People, let's not get into an edit war. Let's come to a consensus here first before we start madly editing the article. With the references being provided, that shouldn't be a problem. In the mean time, you could add a Disputed-section to the article if necessary. Martijn Meijering (talk) 22:21, 24 July 2012 (UTC)

Disruptive editing

The anonymous IP keeps reinserting his point of view against a consensus of other editors. This is clearly against the rules. It is also pointless, because if this is escalated, as it will be eventually, he will most certainly lose as these rules are enforced very strictly. The way to influence an article in case of a dispute is to engage constructively on Talk and to try to achieve a consensus. The alternative will end in a ban and no influence at all. Martijn Meijering (talk) 18:29, 25 July 2012 (UTC)

hey sock puppet, where is the consensus and voting? This is not how free wiki works.

72.185.61.209 (talk) 02:50, 26 July 2012 (UTC)

Your 3 satellite version has been reverted by 4 different editors (a total of 5 times). Your concern has been discussed pn the Talk page, but no-one else seems to agree with you. It looks like a concensus to me. Meters (talk) 03:26, 26 July 2012 (UTC)

You want to accept a positional error of hundreds of metres? Then by all means, use just three satellites. But if you don't want that error, you will need to have your receiver's clock corrected, which means you need four. --  Denelson83  07:28, 26 July 2012 (UTC)

In practice, you need to correct your clock even for a three satellite fix. You need a microsecond accurate clock even to compute distances to 300 meters or so, since light goes 300 meters in a microsecond. Your basic crystal clock can't do this, so you solve for a clock offset just as in the 4 satellite case. It's just that you can do this with one less variable since you have one less unknown. You'll get the degraded position and also a degraded clock adjustment, good to perhaps a microsecond.

"In practice, you need to correct your clock even for a three satellite fix." This is basically saying that you need extra information; i.e., if you already know the exact atomic time, you don't need to solve for it. This is not "basic" GPS operation, but assisted GPS-- equivalent to already knowing with certainty what your ECEF Z-coordinate is, for example. Without knowing, or just assuming, one of the four values which need to be solved for, it is mathematically impossible to arrive at a position solution using only three satellites. siafu (talk) 13:55, 26 July 2012 (UTC)

Yes, it's just another way of saying the same thing. You could actually do the calculation without using a local clock at all. Using the differences in the received times puts your position somewhere on a curve that is the intersection of two hyperboloids. Then you use your extra information to find the spot on your curve, and you've found your location without ever explicitly computing an accurate time. The accurate time is still available - take your final location, then add the time of flight to the time of the transmitted signal - but it was never explicitly computed during the process. LouScheffer (talk) 14:36, 26 July 2012 (UTC)

It looks as if the IP has a history of edit warring and being blocked for it. I think we can see where this will end. Martijn Meijering (talk) 17:08, 26 July 2012 (UTC)

Basic operation

The basic mode of operation of GPS is to use (at least) 4 satellites to solve for x,y,z,t. See, for example, this explanation and many others. If you believe otherwise, please provide a reference. Thanks, LouScheffer (talk) 13:27, 24 July 2012 (UTC)

That is incorrect! THREE satellites are needed for 2D location fix (every satellite sends location and precise ATOMIC time, besides other information: Ephemeris and many more) , fourth satellite needed for altitude only. Furthermore GPS sattelites or Earth are not static (nothing in Universe is). I recommend some reading of GPS SIGNAL and Trilateration article. 72.185.61.209 (talk) 18:44, 24 July 2012 (UTC)

Are you seriously using some mathematical example (formula) created by teachers from Penn state university? Its a formula example how to solve mathematical problem not how GPS actually works. Time signal is send by EVERY GPS satellite (its actually precise ATOMIC clock time)

here is a real source Operation Guide for DAGR: http://webcache.googleusercontent.com/search?q=cache:WPAfxtkXAOsJ:www.i-mef.usmc.mil/external/wss/deployable_virtual_trianing_environment/dvte_handouts/gensim/tech_manuals/dagr/DAGR%2520Pocket%2520Guide.pdf+&cd=3&hl=en&ct=clnk&gl=us#47 72.185.61.209 (talk) 18:54, 24 July 2012 (UTC)

The operation guide says at least four satellites are needed. The only way that three could possibly work is if the receiver's elevation/altitude is known by another means. Otherwise, a GPS receiver must solve for three dimensions plus time. That requires at least four satellite signals. More than four are useful for extra precision and when some are have close angularity. —EncMstr (talk) 19:33, 24 July 2012 (UTC)

Here is from The Future of the Global Positioning System, straight from the Defense Department (who built GPS):

The GPS receiver uses the position and time information, broadcast in the navigation messages and traveling at the speed of light, to calculate approximate ranges to each of the satellites within line of sight to its antenna. These approximate ranges are called pseudoranges, since biases in the user receiver clocks prevent the precise individual ranges from being measured directly. The pseudorange from each individual satellite for a specific but unknown value of user clock error defines a sphere on which a user may be located in three-dimensional space. The intersection of three spheres defines a point, though the intersection is imprecise due to the aforementioned biases in the receiver clock (which in nearly all cases is not an atomic clock) and to effects of ionosphere and atmosphere on the signal transit time. Addition of a pseudorange from a fourth satellite allows calculation of the user receiver clock error and permits computation of the three physical dimensions of the precise intersection, as well as precise time.

Or try this one Introduction to GPS (italics and bold in the original):

GPS receivers are equipped with crystal clocks that do not keep the same time as the more stable satellite clocks (the satellite clocks can be nearly synchronised to GPST using the clock correction model transmitted in the navigation message). Consequently each range is contaminated by the receiver clock error. This range quantity is therefore referred to as pseudo-range, and in order for the user to derive position from pseudo-range data, the receiver equipment is required to track (a minimum of) four satellites, and solve for four unknown quantities: the three-dimensional position components and the receiver-clock offset (from GPST) -- see Section 1.3.5. This is the basis of GPS real-time navigation, and why GPS could be considered an example of a time-difference-of-arrival system.

Or many, many more.LouScheffer (talk) 20:39, 24 July 2012 (UTC)

darling you need to READ the article. THREE dimensional fix (four satellites) is needed for planes and superman (and you). Here is another source.

Three satellites are needed for 2D (basic) fix. http://www.gpsinformation.org/dale/gpsfix.htm 72.185.61.209 (talk) 20:44, 24 July 2012 (UTC)

LouScheffer is correct; four satellites are required for an unassisted GPS receiver-- unassisted in the sense of not being provided any PNT data outside of GPS signals. Probably the most standard refernce, IMO, is Global Positioning System: Signals, Measurements, and Performance by Pratap Misra and Per Enge. In my copy (1st ed.), on p. 23:

A user, therefore, needs a minimum of four satellites in view to estimate his four-dimensional position: three coordinates of spatial position, plus time.

You may be getting confused with the idea of a "2-D" position; all position fixes are, in fact, 3-dimensional, since we live in 3-D universe. Receivers that do "2-D" fixes are taking advantage of extra information, namely the known elevation map of the Earth's surface; for example, car GPS receivers take advantage of the fact that they can assume that the car is on the road, and not anywhere off-road, and can constrain the problem that way. This is essentially equivalent to assisted GPS, mentioned by EncMstr. However, almost all receivers built and designed, including the ones in your cellphones and cars, require 4 satellites in view to function, and the basic positioning problem requires 4 satellites in view. siafu (talk) 20:57, 24 July 2012 (UTC)

Did you even bother to read the source article? : http://www.gpsinformation.org/dale/gpsfix.htm

@72.185.61.209: Unless you are a Flat Earther, you should realize that any position along the surface of a sphere is a position in 3 space. 2D solutions are more apt to apply to short range positioning systems like the defunct LORAN system. —EncMstr (talk) 21:16, 24 July 2012 (UTC)

We use navigation generally for 2D maps! (latitude, longitude) Unless military or flying the plane (then we use altitude as well). Lets get real here. GPS units gets a fix from THREE satellites. That is a must have minimum! Furthermore in the real life scenario six or more satellites are used for increased accuracy. That does not change the fact, that a basic fix is 2D (most maps are in 2D not 3D) is just from 3 satellites using Trilateration. We are talking here about GPS satellite fix (from cold start), not general use. Lets not mistake those two! 72.185.61.209 (talk) 00:40, 25 July 2012 (UTC)

There are many practical reasons not to use the three satellite solution. First, note the error in position will be comparable to any error in elevation. Just to name a few:

• You need a map that has the elevation in every part of the world, otherwise you can be off by kilometers. Most GPS implementations don't have the storage room for such a map.
• If you at the bottom of a cliff, or anywhere the elevation changes rapidly, position may be way off.
• If you are in a city with skyscrapers, what altitude do you use?
• The same GPS might be used by a rafter on a river and a driver on the bridge high above. What altitude do you use?
• If you use it in a hot-air balloon, or an airplane, you get the wrong location.

Furthermore, all of these happen without warning, since the receiver has no way to check. Also, such a solution may well be outside the accuracy limit for cell-phone 911 calls. All these are reasons that 3 satellite is a special case, not normal operation. LouScheffer (talk) 01:28, 25 July 2012 (UTC)

@72.185.61.209: Please indent comments which are responses to a previous comment. See WP:TALK.

I am not sure what kind of map you are referring to—paper or stored digital. Even a GPS which does not display elevation/altitude must do 3D calculations to determine where it is unless it is explicitly told its elevation (or told once where it is so that it can deduce elevation). General purpose GPSs do not have the luxury of assuming an elevation as they may be carried by someone on board a ship or in an airplane. The difference of 60,000 feet (18,000 m) which the same GPS receiver might experience has to be accounted for. —EncMstr (talk) 01:46, 25 July 2012 (UTC)

From a Trimble GPS tutorial: Trimble Navigation is one of the oldest and technically strongest of the GPS companies. Presumably they know how their own receivers work. (Emphasis added) LouScheffer (talk) 02:54, 25 July 2012 (UTC)

The secret to perfect timing is to make an extra satellite measurement.

That's right, if three perfect measurements can locate a point in 3-dimensional space, then four imperfect measurements can do the same thing.

This idea is so fundamental to the working of GPS that we have a separate illustrated section that shows how it works. If you have time, cruise through that.

If our receiver's clocks were perfect, then all our satellite ranges would intersect at a single point (which is our position). But with imperfect clocks, a fourth measurement, done as a cross-check, will NOT intersect with the first three.

So the receiver's computer says "Uh-oh! there is a discrepancy in my measurements. I must not be perfectly synced with universal time." Since any offset from universal time will affect all of our measurements, the receiver looks for a single correction factor that it can subtract from all its timing measurements that would cause them all to intersect at a single point.

That correction brings the receiver's clock back into sync with universal time, and bingo! - you've got atomic accuracy time right in the palm of your hand. Once it has that correction it applies to all the rest of its measurements and now we've got precise positioning.

One consequence of this principle is that any decent GPS receiver will need to have at least four channels so that it can make the four measurements simultaneously.

how more simply can I explain this? you don't need elevation for navigation (unless airborne) general maps USE TWO DIMENSIONs. Navigation was, and still is done , by latitude and longitude. GPS receiver is correcting time with every signal received from the GPS satellite. Stop talking about nonsense about some mystery fourth satellite sending "time" only. They all do! And every GPS satellite knows exactly how far from the Earth it is located.(takes about 130ns for the signal to travel from GPS satelite. so time correction would be irrelevant from your mystery fourth satellite.Correction of time is done many times in second by THREE satellite signals. Care to read GPS SIGNAL article before discussing this even further? Those satellites are not just some simple beacon. They are sending lot more data. Thats why THREE of them are only needed for basic operation or cold fix. I am not saying that that is used in real life scenario. That would be 6 and more GPS satellites. BUT you are grossly wrong about assuming that FOUR satellites are needed for navigation or fix.

72.185.61.209 (talk) 05:06, 25 July 2012 (UTC)

I certainly agree all GPS signal contains the exact time they were sent. However, the receiver needs to know the exact time they were RECEIVED. That's because it takes the difference between these times, divides by the speed of light, and uses that to find the distance to the satellite. It does the receiver no good at all to know the signal was transmitted at 13.000001234 seconds after the hour unless it knows the received time to comparable accuracy.

The internal clock in a GPS is not nearly good enough for this purpose. Furthermore, the receiver cannot just use the times it gets from the satellites directly, since they will all be different - they have all traveled different differences, and suffered different delays. The net result is that the receiver must calculate the accurate received time, using the data (including the sent time) from the satellites, finding a location in space (x,y,z or lat,long,elevation) and an accurate time, that agrees with all the data received. That's four variables in four unknowns, so four satellites are needed.

As the article notes, you can do this with 3 satellites if you assume an elevation. But the receiver still needs to compute the super accurate time, because the time is needed to compute the delay, which is needed for the distances, which are needed for even a 2-D fix.

The reason this is not normally done is that assuming an elevation is error-prone. For example, from A US government publication on using GPS during wildfires

There are several different types of errors that can occur when using a GPS receiver, for example:

...

Unknowingly relying on a 2D position instead of a 3D position for determining position coordinates. This mistake can result in distance errors in excess of a mile.

So while a GPS system can work with three satellites, it then has to assume an altitude, which is error-prone. This is a desperation measure by a GPS when it can't find 4 satellites. Four satellites requires no assumptions and is much more accurate. This is why it's the normal mode of operation. LouScheffer (talk) 10:53, 25 July 2012 (UTC)

you are the one who is Assuimng. GPS receiver has time and location from 3 satellites and that is good enough to calculate location for normal human being on earth. its even good for emergency location and for cold fix. We are not talking about accuracy here. You are missing the point. four sattelite is needed for altitude measurement. there is no question about it. You dont need to know you altitude for location fix on the map. and again i have to repeat my self Trilateration do some reading about it. tri- means 3 BTW

72.185.61.209 (talk) 17:26, 25 July 2012 (UTC)

Yes, Tri means three - three distances. How does it find the distances? It knows what time they were sent, and what time they were received. How does it know what time they were received? It has to solve for that, that make four (4) measurements in order to do trilateration. (Or assume one of this distances, such as the altitude).

If you don't think GPS receivers need an accurate clock, how do you think they calculate the distance from each satellite? If it does need accurate times, how do you propose the GPS box figures out what they are? LouScheffer (talk) 18:06, 25 July 2012 (UTC)

gps receivers (used by general public) do not have atomic clock. but they have very good time correction from signal send by every satelite with location data.that signal travels abut 130ns so that makes every receiver very correct clock. So you theory about fourth satelite needed for time correction is incorect. every gps receiver does time correction with every signal received. And again, I have to repeat my self. Do some reading and educate your self about GPS SIGNAL

72.185.61.209 (talk) 02:48, 26 July 2012 (UTC)

I'm quite sure we don't need to continue this discussion any further. It's very clear that this anonymous contributor is not familiar with the functioning of GPS receivers, and we've shown enough references to convince anyone who is eager to learn. Further changes to the basic operation section by this contributor to the effect of 3 satellites being all that are necessary should be considered disruptive at this point. siafu (talk) 18:25, 25 July 2012 (UTC)

so to sum this up. just because i dont have account created i have nothing to say? stop saying anonymous.you yourself are using anonymous name. or is Siafu the name you pay your taxes with? didnt think so! I have provided clear references showing that 3 satellites are needed for cold fix and for latitude and longitude navigation. that is a fact. Educate your self about GPS SIGNAL before trying to insult somone!

72.185.61.209 (talk) 02:48, 26 July 2012 (UTC)

You should be aware that a GPS receiver does not directly calculate latitude and longitude. The position is always first calculated using a three-dimensional rectangular coordinate system, and then converted into whatever coordinate format is selected using whatever geodetic datum is selected. --  Denelson83  07:42, 25 July 2012 (UTC)

Reverted him again, and templated him for edit warring. Meters (talk) 03:35, 26 July 2012 (UTC)

and that makes you a hero?

72.185.61.209 (talk) 03:43, 26 July 2012 (UTC)

Answer me this. How do you think the GPS calculates how far away it is from the satellite? Obviously, this is needed for even a 2D fix. You just need a few words, not a formula or anything.

It is certainly true the GPS signals contain the time they were sent (which is exceedingly accurate) but they have differing delays from 65 to 85 milliseconds to the receiver, depending on where the satellite is in the sky. This is fine for your setting your watch, where you could just choose one, or an average or something. But please explain how it enables you to find the distance to each satellite. LouScheffer (talk) 10:42, 26 July 2012 (UTC)

No no, be specific. I'd like to see the math that can solve 4 unknowns from just 3 equations. siafu (talk) 13:50, 26 July 2012 (UTC)

The use of terms like 3 equations in 3 unknowns or 4 equations with 4 unknowns is an inappropriate over simplification of the GPS problem. The navigation equations are nonlinear and the attempt to apply rules for linear equations to these nonlinear equations indicates a superficial understanding of the problem. RHB100 (talk) 20:58, 5 September 2012 (UTC)

Most solution algorithms for the navigation equations (e.g. batch filter, kalman filter) require linearization of the equations. Also, it is not at all possible to solve an underdetermined system of non-linear equations any more than it is for a system of linear equations; your statement is effectively irrelevant. I'd thank you to keep comments about other editors' expertise out of the discussion. siafu (talk) 21:47, 5 September 2012 (UTC)

Siafu, the Bancroft method solves the nonlinear equations directly. Kalman filter is not a soloution method. Solution methods are listed in the section, "Navigation Equations". Batch filter is not listed. The statement, "Also, it is not at all possible to solve an underdetermined system of non-linear equations any more than it is for a system of linear equations", is meaningless to me. I hold advanced engineering degrees from both the University of Arkansas and UCLA. I am a Licensed Professional Engineer. Do you have any degrees from top quality engineeering schools? RHB100 (talk) 00:21, 6 September 2012 (UTC)

I do, in fact, hold such degrees, and I am an expert in GPS, specializing in GPS radio occultation. The batch filter is essentially a least squares method for solving for a position solution; it, along with the Kalman filter, are common in statistical orbit determination and precision GPS applications (e.g. seismic monitoring). The Bancroft method is interesting, but in my experience, very rarely referenced or used since linearization is more than sufficient to determine an accurate and precise solution. In addition, the Bancroft method is still going to be unable to solve this system of equations with only three measurements, since an ambiguity remains here just as it does in the linearized system. Congrats on your credentials, also, but they don't hold much weight here on wikipedia, I'm afraid, since there's no way to prove that they are real. siafu (talk) 04:15, 6 September 2012 (UTC)

Even if there was a way to prove that these credentials are real, they are 100% irrelevant here. RHB100 should know that - he was warned about this many times on this artcle talk page and on his own talk page, which he recently cleaned out. RHB100, please do not go that way again, but back your edits with reliable sources, not with an authoritative appeal to your personal experience. Thanks. - DVdm (talk) 08:07, 6 September 2012 (UTC)

I think the most recent changes that Lou Scheffer has contributed has the section "Basic concept of GPS" pretty much the way it should be as far as I can tell at this time. He has done a good job of utilizing the best parts of other's contributions along with his own contributions to come up with a well written section. Let's not make it worse by making poorly thought out edits. RHB100 (talk) 20:03, 6 September 2012 (UTC)

The clock bias

In the section of navigation equation, following to the GPS/GNSS convention, the clock bias should be its advance. The signal transit time, thus, should be $t_{\text{r}}-b-t_{i}$.

Kkddkkdd (talk) 15:42, 14 August 2012 (UTC)

Two sensible solutions with 3 satellites

This is just an academic point, but it seems to me it should be possible in some odd cases to get two sensible solutions with 3 satellites. The solution you get will be reflected through the plane of the satellites to get the other solution. In general this will be in outer space, since all satellites are above you. But if the three satellites are all in one GPS orbital plane, this plane cuts through the center of the earth, hence the reflected point will also be on the surface of the Earth. (In general there are 4 satellites in one orbital plane, so if they were evenly spaced you cannot see three at once. But they are unevenly spaced, and there may be spares, so maybe you could see three at once.) Then if you are stationary (with respect to the Earth's surface) then the other point will be moving with twice the Earth's rotational velocity at that point, and hence in general not be likely. But if you are in the far North, or far South, then this velocity can be reasonable as well. So in the very special case of three satellites, all the same plane, and a position in the far North or South, you could get two sensible solutions. Is this reasoning correct? LouScheffer (talk) 11:56, 7 September 2012 (UTC)

The 3 satellites do not have to be co-orbital to allow two solutions on the surface. In the archive of this talk, user "crazy software productions" shows a constructive proof. The "other" solution is always moving rather fast in time, so real confusion will be unlikely.Woodstone (talk) 16:39, 7 September 2012 (UTC)

His case gives two solutions near the Earth, but one solution will be moving with a good fraction of orbital velocity - this is why I call that solution not 'sensible'. In the case I outline, both solutions are near the surface *and* moving slowly, so you can't pick between them based on nearness to the Earth's surface, or speed. LouScheffer (talk) 17:29, 7 September 2012 (UTC)

This could still be useful for launch vehicles, which nowadays use GPS in addition to inertial navigation (I think), but unless we have a source describing it, shouldn't we avoid putting it in the article? Martijn Meijering (talk) 09:57, 8 September 2012 (UTC)

It looks like we agree. Now how about editing out the section trilateration (and 1-D root finding), which describes a theoretical method for exactly 4 satellites which I suspect is not used in commercial GPS devices. −Woodstone (talk) 04:48, 8 September 2012 (UTC)

Basic concept of GPS needs additional sourcing and explanation

The following is stated in Basic concept of GPS, "Three satellites might seem enough to solve for position since three measured distances define just two points". It occurs to me as I further review this section that there is a need for additional sourcing and explanation. How do we know that three measured distances define just two points? I think it is somewhat beyond being obvious. I think that to make this statement we need a source justification. That justification can be provided in a fairly easy and straightforward manner by mentioning that the two points are the intersections of the surfaces of three spheres and then referring to trilateration which shows that there are typically two intersections. RHB100 (talk) 04:12, 9 September 2012 (UTC)

The section has become too complicated. Its title Basic concept is a strong hint for readers there is no detailed nor rigorous theory of operation, and certainly they do not need to be prepared for or endure an in-depth geometric analysis or heavy duty physics. The section should only give the general mechanism of system operation in a conceptual way. It is intended for the intelligent layman as well as an accessible introduction for the detailed treatment later. In its original version, I have found it effective to squelch common myths:

• myth: satellites (therefore governments, companies, etc.) can track a person using a GPS receiver
• myth: GPS receivers transmit a signal
• myth: only one satellite signal is needed
• myth: that speed is directly determined by the unit
• myth: that setting the GPS calendar/clock is necessary for it to work
• myth: that once locked, a GPS receiver works in a tunnel or aboard a submarine
• myth: that wind affects accuracy
• why it can't tell where it is from a deep shaft or surrounded by tall buildings
• why altitude information is usually less accurate than horizontal position

It is sufficient to speak of distances to satellites and precise satellite positions as those principles can be applied intuitively. Mentioning the importance of clock accuracy is a less important concept for understanding the system, but it has been there since the inception; appropriately, it has explained how the system measures the distance from the receiver to each satellite. Unfortunately, that discussion has encouraged tinkering for rigorous accuracy which now seems to be leading to an increasingly rigorous geometric treatment. —EncMstr (talk) 19:33, 9 September 2012 (UTC)

There is one more issue that I have observed which needs to be taken care of. There is a note in a reference which says in part "The two positions are symmetrical through the plane of the satellites". This statement is quite true. However, Wikipedia requires that we have a source for the facts we state. Although the statement is true, it is not obvious. How do we know that it is true. We know because the article, trilateration, provides the source reference we need. In trilateration it is shown that the two solutions are symmetrical about the plane containing the three sphere centers. therefore I think the note should be modified so that a source reference is given as to the reason for the symmetry. RHB100 (talk) 23:24, 9 September 2012 (UTC)

I have gone ahead and implemented the suggestions I have made above. I will let somebody else take care of the suggestions of EncMstr. RHB100 (talk) 00:33, 10 September 2012 (UTC)

EncMstr has complained that the Basic concept section is too complicated. Although I do not fully agree with EncMstr, I think there is one way in which the section could be simplified. We say "Three satellites might seem enough to solve for position" and we then explain why three satellites might seem like enough and explain why three satellites is actually not enough. Going through and understanding this might be an unnecessary intelectual exercise. A better way might be to eliminate the remarks on "Three satellites might seem enough to solve for position" and instead focus the arguments on why four satellites is enough. RHB100 (talk) 18:18, 10 September 2012 (UTC)

I sympathize with RHB100, but agree with EncMstr. The basic operation section has again gotten too complicated. When most readers encounter something like "the intersection of the surface of three spheres", they will stop paying attention. It's not wrong, and maybe they *should* understand it, but many will be turned off and read no further. LouScheffer (talk) 11:26, 11 September 2012 (UTC)

Please do not leave out explaining the basic concept of triangulation in that section. I had no idea how GPS worked until several years when I attended a presentation on the workings of the GPS system, and once the method of triangulation was described the basic concept of GPS became so obvious. Really, this is fundamental, and it is just high-school math after all. The details are complicated, the basics are not. I would even suggest that the basic section should come as the first section in the article, before the history section. Do not dumb the article down to a "how can I use GPS with my new iPhone" style content. As for the myths presented by EncMstr, these would need to be reliably sourced if they are really common myths in order to be addressed in the article. Nageh (talk) 13:18, 11 September 2012 (UTC)

OK, I tried a re-write, with an analogy to make it (I hope) simpler, and address most of the points of EncMstr. Feel free to edit/change/revise/revert etc... LouScheffer (talk) 13:44, 11 September 2012 (UTC)

Not a bad rewrite, but it sweeps under the carpet that in the 2-D case, there are two intersections of the circles. I don't see an easy way around that. And it misses the opportunity of showing that if the watch is off by just a second, the error is in the order of 300 m. −Woodstone (talk) 16:59, 11 September 2012 (UTC)

As for the two intersections, I think it could be said that you can resolve this either by knowing your approximate location (you'll usually know whether you are on the northern or the southern hemisphere) or by using another lateration, which guarantees a unique solution. Nageh (talk) 17:09, 11 September 2012 (UTC)

Of course, the current proposed text has very much the form of a how-to. I think what would be needed is a simple example for 2D triangulation (possibly using light rather than sound waves), then explain how this can be amended to not require synchronized clocks, and then generalize to the 3D case. I am a bit disappointed that no good diagrams for trilateration are available on commons; it would be simple to explain the basic concepts using some good pictures. Nageh (talk) 17:05, 11 September 2012 (UTC)

I think LouScheffer has succedeeded in dumbing down the section. People in general are not as dumb as LouScheffer thinks. Practically everyone is familiar with the fact that we live on an approximately spherical earth, but Scheffer thinks the concept of a sphere would be just too difficult for people to understand. And these big words like intersect, LouScheffer would never want people to learn what for some people would be a new word. LouScheffer apparently wants to keep everybody as dumb as possible. This dumbing down of everything is disgusting. RHB100 (talk) 19:15, 11 September 2012 (UTC)

Personal attacks and commenting on editors instead of content is also pretty unhelpful and "disgusting". We do need a simplified section for the lay reader, and providing such a thing does not mean we have to do without more complicated and in-depth explanations elsewhere. This section is entitled "Basic Operation", and it should be really basic. siafu (talk) 19:22, 11 September 2012 (UTC)

Alright as siafu says this simplified section for the lay reader does not mean we have to do without more complicated and in-depth explanations elsewhere. Scheffer in providing a 2D analogy should not have removed the more realistic and clearly explained explanation ln terms of spheres and their intersections. I think there are a lot of people if not a majority of the readers of the section "Basic concept of GPS" who want to know how GPS really works. I think that for the readers of this section, the statement, "When most readers encounter something like "the intersection of the surface of three spheres", they will stop paying attention." is untrue. We should separate the 2D analogy from the "Bssic concept of GPS" section. RHB100 (talk) 05:00, 12 September 2012 (UTC)

Another thing to think about is the level of English of readers. Wikipedia is used quite a bit by students and those for whom English is not a first language. "Sphere" and "intersect" are not common words - "intersect" in particular in not even in the 10,000 most common words. Wiktionary's list. So I think these should be avoided in the "basic" section. My personal preference is for the simplest explanation that is not wrong in the basic section, and the full explanation in the "Navigation equations" section. But of course each person's opinion may vary... LouScheffer (talk) 00:49, 12 September 2012 (UTC)

This 2D example certainly doesn't qualify as an explanation of basic concepts. Even if it were correct it would not qualify as an explanation. Failure to think clearly in terms of number of intersections has caused errors in this example. RHB100 (talk) 05:00, 12 September 2012 (UTC)

The 2D example involving Church bells was most incorrect. This example implied that two circles intersect at one point but this is of course incorrect since they typically intersect at two points. The Basic concept of GPS section was therefore replaced with a correct description. RHB100 (talk) 06:09, 12 September 2012 (UTC)

In response to Nageh above, I with the help of others have two pictures showing how 2 spher surfaces intersect and how a third sphere surface intersects the other two but somebody took them down. They could be brough back if this appears to be the proper thing to do. Also there is an excellent diagram at the beginning of the trilateration article. RHB100 (talk) 01:56, 14 September 2012 (UTC)

Overanalysis of the 4 satellites minimum

I contracted:

Four sphere surfaces typically do not intersect. This can be seen from the fact that three sphere surfaces typically intersect at two points as shown in trilateration and for a fourth sphere surface to intersect the other three, it would have to go through one of the points at which the other three intersect. This is a special case not the general situation. But we know that the four sphere surfaces corresponding to the four satellites do intersect at a point, namely the position of the receiver. Thus we can say with confidence that when we solve the navigation equations, the solution gives us the position of the receiver along with accurate time thereby eliminating the need for a very large, expensive, and power hungry clock.

into:

From these signals, position, altitude and precise time can be computed.

because I felt that the original explanation was obtuse and User:RHB100 reverted that change. I'd like to discuss the change here.

I do not understand why we need an explanation of why these 4 spheres will intersect because: (1) Obviously they will intersect because they are all defined to contain your current position and (2) They won't intersect exactly because of lack of precision, noise, bad signal, reflection, etc. so you will have to guess at the most likely position based on the imperfect data you get.

In general, it seems like this section is trying to describe the theory behind how someone could triangulate position with perfect signal and no assumptions (like assuming a driver is on the surface of the earth, say). I think this is not very valuable and instead the minimum satellite number should be referenced to an actual specification/measurement. For example, does the US government specify how many satellites should be needed to accurately measure position? Even better would be if they have information on how much accuracy can be expected based on number of satellites we're reading from!

Anyway, even if you guys want to keep this section completely theoretical, I think the explanation could be made much less verbose. Perhaps something like:

Without any assumptions (like the user being on the surface of the Earth), for distances are needed for perfect three-dimensional trilateration. In addition to position and altitude, solving the navigation equations will also provide accurate time.

Rather than trying to explain why 4 distances uniquely determine 3d position, I'd suggest we just add a reference so that people who want to dig deeper, can. Cheers, — sligocki (talk) 20:52, 21 September 2012 (UTC)

The 4 satellite requirement is not a matter of an expert opinion, but a simple mathematical constraint. Four unknowns requires four measurements to determine. Specifically, statements like:

For example, does the US government specify how many satellites should be needed to accurately measure position? Even better would be if they have information on how much accuracy can be expected based on number of satellites we're reading from!

Don't really make much sense; without four satellites the accuracy will be zero in a theoretical sense because it's not possible to solve the system of equations involved. As an aside, of course the US Gov says that you need four satellites (in the GPS ICD, e.g., which can be found here), the same way that the designers of cars say that you need to put fuel in them to make them run. In the non-basic sense, there are special cases where you can use fewer satellites, but this requires additional data source(s) to substitute for the missing satellite measurements. I agree that the intersecting spheres discussion can get very obtuse very quickly, and isn't really needed, however, and I very much endorse a much less verbose description. siafu (talk) 21:27, 21 September 2012 (UTC)

We are trying to describe the basic concept of how GPS works. There are many people with scientific curiosity who want to understand the basic concepts of how GPS works not just that it does work. We have already achieved a great simplification by going from explaining why three satellites are insufficient to a concentration on explaining why four satellites are sufficient. The key to a clear explanation is to use sufficient words to explain with clarity. When you try to be cute by explaining too many things with too few words, the explanation ends up being unclear and confusing to many readers. RHB100 (talk) 22:00, 22 September 2012 (UTC)

Explaining basic concepts for all readers

There are some who have indicated that they desire to see less explanation of the fundamentak principles. Yet others have expressed an interest in understanding how GPS works. In order to satisfy these conflicting desires, some of the explanatory material has been moved to notes. RHB100 (talk) 20:28, 23 September 2012 (UTC)

Even greater conciseness was achieved through editing amd eliminating a sentence. RHB100 (talk) 23:04, 23 September 2012 (UTC)

GLONASS

Why is the NA in navigation capitalized on the N in GLONASS? Thanks!! If you have the answer tell me on my talk page.

--Miquaz1 (talk) 23:35, 26 September 2012 (UTC)

According to the page wikipedia has on GLONASS, the "GLO" is for Globalnaya ("Global"), so it would seem that someone at the Russian Federal Space Agency way back when just decided to capitalize the whole thing. siafu (talk) 00:38, 27 September 2012 (UTC)

You know, I might have been the one that did that years ago, but since then a) the GLONASS page itself doesn't show the mixed capitalization, and b) I've rarely seen a page other than the main article explain why something is acron-ized the way it is. I'll change it now but Miquaz1 was good to question it. - Davandron | Talk 04:31, 18 December 2012 (UTC)

Section needed

There needs to be a section on getting lost by following GPS. Some people get lost by asking for the shortest route, like this couple did: [6]. They were looking for the shortest route from Oregon to Jackpot, Nevada. --Auric (talk) 02:10, 3 October 2012 (UTC)

This is not relevant to this (GPS) article. If such content deserves to go anywhere, it would be in GPS receiver in a section on misuse, or–more likely, in an article of notable examples of this kind of tragedy: incidents involving GPS receiver misuse, failure to use good judgment, blind faith in technology, following sheep over the cliff, or the like. —EncMstr (talk) 04:28, 3 October 2012 (UTC)

Thanks. I'll do just that.--Auric (talk) 04:50, 3 October 2012 (UTC)

External link to GPS simulation

I restored a link to a Java based simulation of GPS and GLONASS that provides a graphical depiction of space vehicle motion and illustrates the variation of dilution of precision (DOP) with varying constellation configuration. The latter is hard to appreciate and compare between the two systems with out observing it over time. This link is certainly relevant to the article and is devoid of any advertizing or other commercial content. The link existed in the article for a number of years and was recently deleted without any specific explanation. Its purpose was explained on this talk page (since archived) and no counter comments were made. Roesser (talk) 00:22, 26 October 2012 (UTC)

Events on GPS

• GPS was disabled during Georgia conflict. (it should have been mentioned) — Preceding unsigned comment added by 189.113.75.27 (talk) 19:36, 8 November 2012 (UTC)

Not finding any detailed refs on this but there are mentions: 2008 South Ossetia war, [7]. -—Kvng 00:44, 24 November 2012 (UTC)

It was not disabled, it was just locally jammed. You can jam the public GPS band but if you don't have an alternate frequency to use you're basically without positioning as well. OTAN military generally jam public GPS in the area of operation but they can still use the encrypted military band when they do. Shutting down the whole GPS system would cause worldwide trouble. 62.42.72.17 (talk) 22:03, 7 December 2012 (UTC)

In Navigation equations Why replace $\,t_{r}$ with $\,{\tilde {t}}_{\text{r}}$ ?

Why replace $\,t_{r}$ with $\,{\tilde {t}}_{\text{r}}$ ? It adds nothing to understanding yet it creates more cumbersome notation. RHB100 (talk) 21:27, 16 January 2013 (UTC) hai

Clock bias

The sentence "The clock error or bias, b, is the amount that the receiver's clock is off. " in Navigation equations seems redundant. Kkddkkdd (talk) 17:39, 30 March 2013 (UTC)

I agree, I suggest you delete it Roesser (talk) 20:22, 30 March 2013 (UTC)

Threat to aircraft and ships from car based jammers.

Hi all, I don't think there is any ref to this issue in the article. See here [8] for some material. Is this a non-story I wonder (ie do the jammers actually not work?) Anyone have some good information? Springnuts (talk) 16:55, 13 April 2013 (UTC)

I've seen a few talks on the issue of interference and jammers. This is a definite issue, though, since because of the low power of the GPS signals, jammers do in fact work; there was a documented incident with a problem caused by a trucker using a GPS jammer at Newark airport, for example ([9],[10], pretty sure this was because of interference with a WAAS base station). In addition, it is illegal in the United States to buy, sell, or operate a device that produces interference in the GPS frequencies (L-band). siafu (talk) 17:11, 13 April 2013 (UTC)

Economic impact

Hey everyone, I was wondering if readers might benefit from a section on the economic impact of the GPS system since the U.S. government made it freely accessible? Jodayagi (talk) 19:31, 9 May 2013 (UTC)

Excellent suggestion! I look forward to reading it and seeing if I can find anything to add. —EncMstr (talk) 21:13, 9 May 2013 (UTC)

Talk about spheres

I believe just about everything in this section is wrong: http://en.wiki.x.io/wiki/Global_Positioning_System#Basic_concept_of_GPS

GPS receivers don't have atomic clocks (although the satellites do), so they can't use the time a signal was emitted from a satellite to determine a fixed sphere they are on. Rather, you know that the *difference* in times between two satellites puts you on a hyperboloid.

My understanding it that this is the real reason you need 4 satellites. The 4th satellite doesn't distinguish between 2 points (which are at the intersection of 3 sphere, it's the extra variable to make up for the fact that you don't know the absolute time (only time differences). Ambiguity between just two (or three or four) discrete points in space is probably resolved by assuming you're on the surface of the Earth. Jess (talk) 21:21, 19 August 2013 (UTC)

That section does not say that GPS receivers use atomic clocks, and does say it precisely timing the signals sent by [the] satellites. Can you point to any specific incorrect fact?

Also, there is no need to resolve sphere overlap ambiguity when there are four or more satellite signals. —EncMstr (talk) 22:33, 19 August 2013 (UTC)

Articles relating GPS and Trilateration

Woodstone has said "remove doubtful method with shaky refs". This is tottally and completely false. This section shows that the method involving trilateration and one dimensional iteration has good references. The next section shows that this method most certainly works.

There are many articles relating GPS and Trilateration as can be seen at GPS and Trilateration articles. Unfortunately most of these articles seemed to be dumbed down, they talk about circles instead of spheres as if the concept of a sphere were too difficult to comprehend.

One of the better articles in my opinion is "Position Determination with GPS".

It is not difficult to understand that there are so many such articles since one of the fundamental principles in GPS is the determination of location in part by determining the intersections of three spheres. Once the clock error has been approximately driven to zero so that four spheres intersect approximately, an estimate of position will have been obtained.

This is the reason why trilateration as a part of numerical root finding (i.e. finding the value of b which drives da to zero through an iterative process) is one of the two methods discussed in Position calculation, advanced.

Another method discussed is multidimensional root finding. This method does not involve the use of trilateration.

It should be kept in mind that these methods are discussed on the conceptual level. They are not descriptions of algorithms.

The Bancroft method is in my opinion the best method. RHB100 (talk) 19:48, 20 August 2013 (UTC)

The problem of course being that nobody actually uses the Bancroft method in computing position solutions. Batch linearized least squares is more than enough to get the job done, even from cold start-- if you provide an a priori receiver solution of {0, 0, 0} (in ECEF) it will still converge in 2-3 iterations. It seems inappropriate both to have Bancroft be the first method mentioned, and really even to include it in the article at all. siafu (talk) 20:03, 20 August 2013 (UTC)

The Bancroft is a method that involves least squares. The term "batch linearized least squares" is inadequate to describe what you're talking about. You say 2-3 iterations but the Bancroft method requires no iterations. RHB100 (talk) 00:15, 21 August 2013 (UTC)

Whether or not the Bancroft method requires iteration is irrelevant; the fact is that it's little more than a mathematical curiosity not actually applied in GPS technology except in a few rare cases. And yes, "batch linearized least squares" is more than enough to describe what I'm talking about. Linearize the system, in this case the range equations, and solve the overdetermined set of pseudoranges using a batch filter (i.e., ingesting all measurements at once as opposed to something like a Kalman filter). If you're unfamiliar with this, the math is laid out quite clearly in Linear least squares (mathematics) and the application to GPS is done in basically every GPS textbook. Rather than continuing the harp on about how wonderful the Bancroft method is, you could provide some sources to show that it's actually notable in the GPS community. siafu (talk) 17:32, 21 August 2013 (UTC)

First you talk about a method which requires 2-3 iterations. But then you refer to Linear least squares (mathematics). But linear least squares provides a closed form solution which implies zero iterations. So what you say is self contradictory. Now I am a licensed professional engineer with many years of experience and I find thw words "batch linearized least squares" to be vague and ambiguous and I have certainly studied and used linearized least squares. Now if these words do not clearly describe to me the computational steps you are talking about, then they are certainly likely to confuse the typical reader. RHB100 (talk) 20:01, 21 August 2013 (UTC)

Yes, I've heard you relate your list of qualifications before. I am myself a published research engineer specializing in GPS. As for least squares, you clearly neither read the article nor are apparently familiar with the topic. The least squares method for overdetermined systems is an iterative one, arriving at progressively more accurate estimates, assuming that the initial point was within the linear region of the system in question. See page 15, here: http://www.nbmg.unr.edu/staff/pdfs/Blewitt%20Basics%20of%20gps.pdf . Again, do you have any sources at all that indicate that the Bancroft method is anything more than just an elegant but completely non-notable solution? siafu (talk) 20:34, 21 August 2013 (UTC)

References 92 and 93 are given for the Bancroft method, one of which is in an IEEE publication and the other is free. These should certainly be adequate. What references do you have critical of the Bancroft method? How does the method you are talking about differ from what is called "Multidimensional Newton-Raphson calculations" in the Wikipedia GPS article and what is also discussed in the paper called, "The Mathematics of GPS" by Richard B. Langley of the University of New Brunswick? Once I understand the method you are talking about, then we can discuss the advantages and disadvantages as compared to the Bancroft method. RHB100 (talk) 21:02, 21 August 2013 (UTC)

Discussing the advantages and disadvantages of the Bancroft method is exactly what we should not be doing, see WP:OR. I don't need a reference that's critical of the Bancroft method; I'm not saying it's a bad method or that the article should say it's a bad method. I'm saying nobody every uses it. It's not used in precise positioning applications, it's not used in orbit determination, it's not used in navigation. Why do we have a section on it? The article needs to reflect the expert consensus, and experts are not particularly interested in Bancroft's method. siafu (talk) 22:04, 21 August 2013 (UTC)

Well we don't know whether what you say is true or not. If something is wrong with the Bancroft method then we need a reference and a discussion so that everybody will know. RHB100 (talk) 23:38, 21 August 2013 (UTC)

Can you or can you not provide any support at all for the assertion that the Bancroft method is notable in the GPS community as a method used for obtaining position solutions? It is not at all my responsibility to demonstrate that it isn't. Furthermore, I mentioned that you could read about how it's actually done in any GPS textbook, and went so far as to give you one link already. I also recommend Global Positioning System: Signals, Measurements, and Performance by P. Misra and P. Enge, as it's the one used when I took intro to GNSS years ago. You could also try Understanding GPS: Principles and Applications, Second Edition by E. Kaplan and C. Hegarty, which I actually prefer. There is nothing "wrong" with the Bancroft method, but it's just not widely used, and one has to wonder why the section on navigation equations does not provide a realistic discussion of the topic, or at least one that is recognizable to an expert in the field. siafu (talk) 05:55, 22 August 2013 (UTC)

I don't know how many times I have to keep repeating this to you but the fact the Bancroft method was published in an IEEE publication is more than adequate to show its notability. The fact that it was also published in an online paper also shows its notability. RHB100 (talk) 18:57, 22 August 2013 (UTC)

The fact that it's published in IEEE means only that it exists and it's a real thing. Why should it be in the article? Do we have similar sections on ARAIM, for example? It's certainly published in IEEE [11]. Bancroft's method deserves maybe two sentences given the current weight given to methods actually in use in GPS. The fact that there's a section that says "Additional methods for more than four satellites" is itself absurd. Except in rare cases (indoors, urban canyons) receivers will invariably see more than four satellites, and the idea that they would ever implement a scheme that fails to take advantage of more than 4 satellites is patently ridiculous. Unless you are able to provide a source indicating that Bancroft's method is not just an extant method but of actual note in the actual GPS community, then the section needs a major rewrite, which I plan to do this weekend. siafu (talk) 01:56, 23 August 2013 (UTC)

Code and results from using trilateration and one dimensional root finding

Woodstone has said "remove doubtful method with shaky refs". This is tottally and completely false. This section shows that the method involving trilateration and one dimensional iteration most certainly works. The previous section has shown that this method has good references.

There have been some who have expressed some doubt as to whether the method involving trilateration and one dimensional root finding as discussed in the GPS article would work. It is hoped that this section will clarify that misunderstanding by showing that this method certainly does work. The results of running a Fortran program which uses this method can be seen by clicking show on the bar labelled "Results from using trilateration and one dimensional root finding". Three of the routines that comprise the Fortran program can be seen by clicking show on the indicated bars below. Two of the subroutines, subroutines zbrac and rtbis, cannot be shown because they are protected by copyright. These subroutines can be found in the book on Numerical Recipes. The subroutine zbrac performs the task of finding a bracket for the solution given an initial guess. The subroutine rtbis performs the task of making a binary search to find a small enough bracket of the solution to meet the specified accuracy.

The results below show for each case the positions and psuedoranges of the three satellites which are used for trilateration and the one satellite which is not used for trilateration. The solution, the receiver position is then shown along with RBIAS which is the bias in dimensions of distance rather than time. RHB100 (talk) 21:14, 24 April 2010 (UTC)

Results from using trilateration and one dimensional root finding

 Trilateration Satellites
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 1    15524471.180   -16649826.220    13512272.390    22262088.180
 2    -2304058.534   -23287906.470    11917038.110    19908187.050
 3   -14799931.400   -21425358.240     6069947.224    21479180.580
  
 Non-trilateration Satellite
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 4    16680243.360    -3069625.561    20378551.050    24554242.170
 
                  X 	       Y 	   Z      RBIAS
 Solution =    -733186.0  -5443792.0   3231193.0 -300.0

STOP Normal completion in routine main of gps.

[rhb100@localhost gps]$ ./gps.exe

 Trilateration Satellites
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 1    15524471.180   -16649826.220    13512272.390    22262088.180
 2    -2304058.534   -23287906.470    11917038.110    19908187.050
 4    16680243.360    -3069625.561    20378551.050    24554242.170
  
 Non-trilateration Satellite
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 3   -14799931.400   -21425358.240     6069947.224    21479180.580
 
                  X 	       Y 	   Z      RBIAS
 Solution =    -733186.0  -5443792.0   3231193.0 -300.0

STOP Normal completion in routine main of gps.

[rhb100@localhost gps]$ ./gps.exe

 Trilateration Satellites
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 1    15524471.180   -16649826.220    13512272.390    22262088.180
 3   -14799931.400   -21425358.240     6069947.224    21479180.580
 4    16680243.360    -3069625.561    20378551.050    24554242.170
  
 Non-trilateration Satellite
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 2    -2304058.534   -23287906.470    11917038.110    19908187.050
 
                  X 	       Y 	   Z      RBIAS
 Solution =    -733186.0  -5443792.0   3231193.0 -300.0

STOP Normal completion in routine main of gps.

[rhb100@localhost gps]$ ./gps.exe

 Trilateration Satellites
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 2    -2304058.534   -23287906.470    11917038.110    19908187.050
 3   -14799931.400   -21425358.240     6069947.224    21479180.580
 4    16680243.360    -3069625.561    20378551.050    24554242.170
  
 Non-trilateration Satellite
Sat #     X 	            Y 	           Z 	 	PSUEDORANGE               
 1    15524471.180   -16649826.220    13512272.390    22262088.180
 
                  X 	       Y 	   Z      RBIAS
 Solution =    -733186.0  -5443792.0   3231193.0 -300.0

STOP Normal completion in routine main of gps. [rhb100@localhost gps]$

The main routine reads in and writes out the data. A call to trilat and a function evaluation of dacomp help to provide initial estimates of a bounding bracket between BIAS1 and BIAS2. A call to zbrac modifies BIAS1 and BIAS2 if necessary to assure that the solution is bracketed. The subroutine rtbis returns with the solution, RGBIAS, which is the value of the argument of function dacomp required to cause the function to evaluate close enough to zero to satisfy the specified accuracy requirement.

MAIN FORTRAN ROUTINE

      IMPLICIT DOUBLE PRECISION (A-H,O-Z)
C        1         2         3         4         5         6         7
C 3456789012345678901234567890123456789012345678901234567890123456789012
C
	EXTERNAL DACOMP
	DOUBLE PRECISION XSAT(5), YSAT(5), ZSAT(5), PRANGE(3), ERRMAT(3)
	DOUBLE PRECISION ERRTOT(2), SATTIM(3), RSAT(5,3), R4(2), DAA(2)
	CHARACTER HEAD1*70, HEAD2*70
	LOGICAL SUCCES
	COMMON /COMMDA/ RSAT, PRANGE, SATTIM, PRANTI
	INW = 25
	IOUT = 45
	CLIGHT = 299792458. ! meters per second
      OPEN(UNIT = INW, FILE = 'sat_position_prange.txt'
     2 , FORM='FORMATTED', ACTION='READ', STATUS='OLD')
      OPEN(UNIT = IOUT, FILE = 'out1.txt'
     2 , FORM='FORMATTED', ACTION='WRITE', STATUS='UNKNOWN')
 
	READ(INW,*) I_TIME, HEAD2 
! Satellite I_TIME is not for trilateration but for clock correction
C	WRITE(6,*) ' Satellite ', I_TIME, HEAD2
	
       READ(INW, *) HEAD1
       WRITE(6,*) ' Trilateration Satellites'
	WRITE( 6, *) HEAD1

	K = 1
	DO 10, I=1,4 ! Read header lines and data
	  IF(I .NE. I_TIME) THEN
            READ(INW, *)  ISAT, (RSAT(K,J), J=1,3), PRANGE(K)
            WRITE( 6,'(1X,I2,4F16.3)') I, (RSAT(K,J), J=1,3), PRANGE(K)
	    K = K + 1
	  ELSE
            READ(INW, *)  ISAT, (SATTIM(J), J=1,3), PRANTI
	  END IF
10	CONTINUE
	WRITE(6,*) '  '
       WRITE(6,*) ' Non-trilateration Satellite'
 	WRITE( 6, *) HEAD1
       WRITE( 6,'(1X,I2,4F16.3)')I_TIME, (SATTIM(J), J=1,3), PRANTI

	READ (INW,*) IPRINT, HEAD2
C	WRITE(6,*) ' IPRINT = ', IPRINT
	CLOSE(INW)

C	CALL TRILAT( XSAT, YSAT, ZSAT, PRANGE, NSOLNS)
	CALL TRILAT( RSAT(1,1), RSAT(1,2), RSAT(1,3), PRANGE, NSOLNS)

C	WRITE(6,*) ' RSAT= ' , RSAT


	DO 50 I=4,5 ! Distance to 4th sphere for solutions in 4 & 5
	  SQUARE = 0.0
	    DO 47 J=1,3 ! X, Y, Z components
		SQUARE = SQUARE + (  RSAT(I,J) - SATTIM(J)  )**2
47	    CONTINUE
	    R4(I-3)  = SQRT( SQUARE )
	    DAA(I-3) = R4(I-3) - PRANTI
50	CONTINUE

	IF( ABS(DAA(1)) .LE. ABS(DAA(2)) ) THEN !Closest solution
	  DA = DAA(1)
	  ISOLN = 1
	ELSE
	  DA = DAA(2)
	  ISOLN = 2
	END IF

	BIAS1 = DA
	BIAS2 = DACOMP ( BIAS1 )
	
      CALL ZBRAC(DACOMP, BIAS1, BIAS2, SUCCES)
      XACC = 0.01
      RGBIAS = RTBIS(DACOMP,BIAS1,BIAS2,XACC)

      WRITE(6,*) ' '
      WRITE(6,*)  '                  X 	       Y 	   Z      RBIAS'
      WRITE(6,150) ' Solution = ', (RSAT(ISOLN+3,J), J=1,3), RGBIAS

      CLOSE(IOUT)
      STOP ' Normal completion in routine main of gps.'
150   FORMAT(1X,A12,3F12.1, F7.1)
200   FORMAT(I3)
      END

The subroutine trilat computes any intersections of three surfaces given the sphere centers and radii as described in trilateration.

SUBROUTINE TRILAT

 	SUBROUTINE TRILAT( X0, Y0, Z0, RADIUS, NSOLNS )
C
C        1         2         3         4         5         6         7
C 3456789012345678901234567890123456789012345678901234567890123456789012
C
C TRILATERATION FINDS INTERSECTIONS OF SURFACES OF SPHERES
C  SPHERES CENTERED AT X0(I), Y0(I), AND Z0(I) WITH RADII, R(I)
        IMPLICIT DOUBLE PRECISION (A-H,O-Z)

	DOUBLE PRECISION X0(5), Y0(5), Z0(5), RADIUS(3)
	DOUBLE PRECISION X1(5), Y1(5), Z1(5), X2(5), Y2(5), Z2(5)
	DOUBLE PRECISION X3(5), Y3(5), Z3(5), X4(5), Y4(5), Z4(5)
	DOUBLE PRECISION ERRMAT(3)

C          WRITE( 6, * ) 'NOW IN SUBROUTINE TRILAT'
C	DO 10, I=1,3 ! Write data
C          WRITE( 6, * ) I, X0(I), Y0(I), Z0(I), RADIUS(I)
C10	CONTINUE

C Translate coordinate system so as to put center of sphere 1, p1, at	
C  origin.  That is compute expression of vectors, VP2 and VP3 in
C  coordinate system 1.

	DO 20, I=2,3
	  X1(I) = X0(I) - X0(1) 
	  Y1(I) = Y0(I) - Y0(1) 
	  Z1(I) = Z0(I) - Z0(1) 
20	CONTINUE


C First sphere center, P1, remains at origin, since all axes intersect
C  at origin.
	  X1(1) = 0.0
	  Y1(1) = 0.0 
	  Z1(1) = 0.0 

	  X2(1) = 0.0
	  Y2(1) = 0.0 
	  Z2(1) = 0.0 

	  X3(1) = 0.0
	  Y3(1) = 0.0 
	  Z3(1) = 0.0 

	  X4(1) = 0.0
	  Y4(1) = 0.0 
	  Z4(1) = 0.0 

C	DO 22, I=1,3 ! Write data
C          WRITE( 6, * ) I, 'CS1 = ', X1(I), Y1(I), Z1(I)
C22	CONTINUE

C Compute 1st coordinate system rotation angle, ALAMBD, required
C  to set the y component of P2 to zero, that is make Y2(2) zero.
C This rotation is about the positive Z1 axis.
	IF ( (Y1(2) .NE. 0.0) .OR. (X1(2) .NE. 0.0) ) THEN
	  ALAMBD = ATAN2( Y1(2), X1(2) )
	  CLAM = COS(ALAMBD)
	  SLAM = SIN(ALAMBD)
	ELSE
	  ALAMBD = 0.0
	  CLAM = 1.
	  SLAM = 0.
	END IF

	DO 30 I=2,3
	  X2(I) =  X1(I)*CLAM + Y1(I)*SLAM
	  Y2(I) = -X1(I)*SLAM + Y1(I)*CLAM
	  Z2(I) =  Z1(I)
30	CONTINUE

C	DO 32, I=1,3 ! Write data
C          WRITE( 6, * ) I, 'CS2 = ', X2(I), Y2(I), Z2(I)
C32	CONTINUE

CC	VP2MAG = SQRT( X2(2)**2 + Y2(2)**2 + Z2(2)**2 )
CC	VP2MAG = SQRT( X2(2)**2 + Z2(2)**2 )

C Compute 2nd coordinate system rotation angle, PHI, required
C  to put P2 on X axis.  That is make Z3(2) zero.
C This rotation is about the negative Y2 axis.
	IF ( Z2(2) .NE. 0.0 ) THEN
	  VP2MAG = SQRT( X2(2)**2 + Z2(2)**2 )
	  PHI = ASIN(Z2(2) / VP2MAG)
	  CPHI = COS(PHI)
	  SPHI = SIN(PHI)
	ELSE
	  PHI =  0.
	  CPHI = 1.
	  SPHI = 0.
	END IF

	DO 40 I=2,3
	  X3(I) =  X2(I)*CPHI + Z2(I)*SPHI
	  Y3(I) =  Y2(I)
	  Z3(I) = -X2(I)*SPHI + Z2(I)*CPHI
40	CONTINUE

C	DO 42, I=1,3 ! Write data
C          WRITE( 6, * ) I, 'CS3 = ', X3(I), Y3(I), Z3(I)
C42	CONTINUE

C Compute 3rd coordinate system rotation angle, THETA, required
C  to make Z compenent of P3 zero.
C This rotation is aboutthe -X3 axis.
	IF ( (Z3(3) .NE. 0.0) .OR. ( Y3(3) .NE. 0.0) ) THEN
CCC	VP3MAG = SQRT( X3(3)**2 + Y3(3)**2 )
	  THETA = ATAN2( Z3(3), Y3(3) )
	  CTHETA = COS(THETA)
	  STHETA = SIN(THETA)
	ELSE
	  THETA =  0.
	  CTHETA = 1.
	  STHETA = 0.
	END IF

	DO 50 I=2,3
	  X4(I) =  X3(I)
	  Y4(I) =  Y3(I)*CTHETA + Z3(I)*STHETA
	  Z4(I) = -Y3(I)*STHETA + Z3(I)*CTHETA
50	CONTINUE

C          WRITE( 6, * ) 'STILL IN SUBROUTINE TRILAT'      C
C	DO 52, I=1,3 ! Write data
C          WRITE( 6, * ) I, 'CS4 = ', X4(I), Y4(I), Z4(I)
C52	CONTINUE


C Now find intersections of 3 sphere surfaces expressed in 
C  coordinatesystem 3.

	D = X4(2)
	D_SQR = D**2
	R1_SQR = RADIUS(1)**2
	R2_SQR = RADIUS(2)**2
	R3_SQR = RADIUS(3)**2
	AI = X4(3)
	AJ = Y4(3)

	X = ( R1_SQR - R2_SQR + D_SQR ) / (2.*D)
	X_SQR = X**2
	Y = (R1_SQR - R3_SQR - X_SQR + (X - AI)**2 + AJ**2) /(2.*AJ)
	Y_SQR = Y**2
	Z_SQR = R1_SQR - X_SQR - Y_SQR
	IF(Z_SQR .GT. 0.0) THEN
	  NSOLNS = 2
	  ZP = SQRT(Z_SQR)
	  ZM = -ZP
	ELSE IF (Z_SQR .EQ. 0.0) THEN
	  NSOLNS = 1
	  ZP = 0.0
	  ZM = 0.0
	ELSE
	  NSOLNS = 0
	END IF

C	WRITE(6,*) NSOLNS, ' solutions with X, Y, ZP, ZM = '
C     2               , X, Y, ZP, ZM

C Verify solutions

C	ERRMAT(1) = RADIUS(1) - SQRT( (X-X4(1))**2 +  (Y-Y4(1))**2 
C     2       +  (ZP-Z4(1))**2 )
C	ERRMAT(2) = RADIUS(2) - SQRT( (X-X4(2))**2 +  (Y-Y4(2))**2 
C     2       +  (ZP-Z4(2))**2 )
C	ERRMAT(3) = RADIUS(3) - SQRT( (X-X4(3))**2 +  (Y-Y4(3))**2 
C     2       +  (ZP-Z4(3))**2 )
C	WRITE(6,*) ' Errors = ', (ERRMAT(I), I=1,3)

	X4(4) = X
	Y4(4) = Y
	Z4(4) = ZP

	X4(5) = X
	Y4(5) = Y
	Z4(5) = ZM

	DO 60 I=4,5
	  X3(I) =  X4(I)
	  Y3(I) =  Y4(I)*CTHETA - Z4(I)*STHETA
	  Z3(I) =  Y4(I)*STHETA + Z4(I)*CTHETA
60	CONTINUE

C	ERRMAT(1) = RADIUS(1) - SQRT( (X3(4)-X3(1))**2 
C     2       + (Y3(4)-Y3(1))**2 + (Z3(4)-Z3(1))**2 )
C	ERRMAT(2) = RADIUS(2) - SQRT( (X3(4)-X3(2))**2  
C     2       + (Y3(4)-Y3(2))**2 + (Z3(4)-Z3(2))**2 )
C	ERRMAT(3) = RADIUS(3) - SQRT( (X3(4)-X3(3))**2  
C     2       + (Y3(4)-Y3(3))**2 + (Z3(4)-Z3(3))**2 )
C	WRITE(6,*) ' Errors X3 = ', (ERRMAT(I), I=1,3)

	DO 70 I=4,5
	  X2(I) =  X3(I)*CPHI - Z3(I)*SPHI
	  Y2(I) =  Y3(I)
	  Z2(I) =  X3(I)*SPHI + Z3(I)*CPHI
70	CONTINUE

C	ERRMAT(1) = RADIUS(1) - SQRT( (X2(4)-X2(1))**2 
C     2       + (Y2(4)-Y2(1))**2 + (Z2(4)-Z2(1))**2 )
C	ERRMAT(2) = RADIUS(2) - SQRT( (X2(4)-X2(2))**2  
C     2       + (Y2(4)-Y2(2))**2 + (Z2(4)-Z2(2))**2 )
C	ERRMAT(3) = RADIUS(3) - SQRT( (X2(4)-X2(3))**2  
C     2       + (Y2(4)-Y2(3))**2 + (Z2(4)-Z2(3))**2 )
C	WRITE(6,*) ' Errors X2 = ', (ERRMAT(I), I=1,3)

	DO 80 I=4,5
	  X1(I) =  X2(I)*CLAM - Y2(I)*SLAM
	  Y1(I) =  X2(I)*SLAM + Y2(I)*CLAM
	  Z1(I) =  Z2(I)
80	CONTINUE

C	ERRMAT(1) = RADIUS(1) - SQRT( (X1(4)-X1(1))**2 
C     2       + (Y1(4)-Y1(1))**2 + (Z1(4)-Z1(1))**2 )
C	ERRMAT(2) = RADIUS(2) - SQRT( (X1(4)-X1(2))**2  
C     2       + (Y1(4)-Y1(2))**2 + (Z1(4)-Z1(2))**2 )
C	ERRMAT(3) = RADIUS(3) - SQRT( (X1(4)-X1(3))**2  
C     2       + (Y1(4)-Y1(3))**2 + (Z1(4)-Z1(3))**2 )
C	WRITE(6,*) ' Errors X1 = ', (ERRMAT(I), I=1,3)

	DO 90, I=4,5
	  II3 = I-3
	  X0(I) = X1(I) + X0(1) 
 	  Y0(I) = Y1(I) + Y0(1) 
	  Z0(I) = Z1(I) + Z0(1)
90	CONTINUE

C	WRITE(6,*) ' X0 = ', (X0(I), I=1,5)
C	WRITE(6,*) ' Y0 = ', (Y0(I), I=1,5)
C	WRITE(6,*) ' Z0 = ', (Z0(I), I=1,5)

C	WRITE(6,*) ' Now leaving subroutine TRILAT. '

	RETURN
	END

The function dacomp computes the closest distance from three surface intersections to the fourth sphere surface as a function of its argument, a range bias.

FUNCTION DACOMP

 	FUNCTION DACOMP ( BIAS )
	IMPLICIT DOUBLE PRECISION (A-H,O-Z)
C        1         2         3         4         5         6         7
C 3456789012345678901234567890123456789012345678901234567890123456789012
C
	DOUBLE PRECISION PRANGE(3), ERRMAT(3), PRTEMP(3)
	DOUBLE PRECISION ERRTOT(2), SATTIM(3), RSAT(5,3), R4(2), DAA(2)
	COMMON /COMMDA/ RSAT, PRANGE, SATTIM, PRANTI 
	
	DO 10 I=1,3
	  PRTEMP(I) = PRANGE(I) + BIAS
10	CONTINUE
	PTMTMP = PRANTI + BIAS

	CALL TRILAT( RSAT(1,1), RSAT(1,2), RSAT(1,3), PRTEMP, NSOLNS)


	DO 40 I=4,5 ! Compute errore for each of 2 solutions
	    DO 35 J=1,3   ! Compute distances to J=1,2,3 satellites
	      SQUARE = 0.0
	        DO 33 K=1,3 ! X, Y, Z components of error
	          SQUARE = SQUARE + (  RSAT(I,K) - RSAT(J,K)  )**2
33	        CONTINUE
	      ERRMAT(J) = PRTEMP(J) - SQRT( SQUARE )
35	    CONTINUE
	    ERRTOT(I-3) = SQRT( ERRMAT(1)**2 +  ERRMAT(2)**2 +
     2       ERRMAT(3)**2 )
40	CONTINUE

	IF( (ERRTOT(1) .GE. 1.E-6) .OR. (ERRTOT(2) .GE. 1.E-6)  ) THEN
	  WRITE(6,*) ' Excessive trilateration error '	    
	  WRITE(6,*) ' ERRTOT = ', (ERRTOT(I), I=1,2)
	  IPRINT = 1
	END IF
	
	DO 50 I=4,5 ! Distance to 4th sphere for solutions in 4 & 5
	  SQUARE = 0.0
	    DO 47 J=1,3 ! X, Y, Z components
		SQUARE = SQUARE + (  RSAT(I,J) - SATTIM(J)  )**2
47	    CONTINUE
	    R4(I-3)  = SQRT( SQUARE )
	    DAA(I-3) = R4(I-3) - PTMTMP
50	CONTINUE

	IF( ABS(DAA(1)) .LE. ABS(DAA(2)) ) THEN
	  DA = DAA(1)
	  ISOLN = 1
	ELSE
	  DA = DAA(2)
	  ISOLN = 2
	END IF

C	WRITE(6,*) ' R4 = ', (R4(I), I=1,2)
C	WRITE(6,*) ' DAA = ', (DAA(I), I=1,2)
C	WRITE(6,*) ' Solution ', ISOLN, ' is closest.', ' DA = ', DA
	DACOMP = DA
	RETURN
	END

RHB100 (talk) 19:48, 20 August 2013 (UTC)

How do you feel about putting all the code on Wikisource? —EncMstr (talk) 20:17, 20 August 2013 (UTC)

Right off the top of my head,I say that I would be glad to do it if they want it. But I don't know too much about Wikisource. RHB100 (talk) 00:21, 21 August 2013 (UTC)

It is of no importance if the method works or not (I actually do not doubt it does for the special case of 4 satellites). What counts is whether it's really used in GPS devices on the market. The latter I sincerely doubt, because of its inherent asymmetry and limitation to 4 satellites. Practical algorithms in devices would use all available satellites in a symmetric way. The references given now are popular naive approaches and generic root finding. So could you point out a reference stating its professional implementation in devices? −Woodstone (talk) 07:13, 22 August 2013 (UTC)

It is important to show that the method works because by so doing it shows that your use of the words "doubtful method" in reference to this method are false and untrue. It is necessary to show that it works in order to make clear that your statement is totally false and untrue. More than adequate references have already been given. RHB100 (talk) 18:40, 22 August 2013 (UTC)

Please read up on wikipedia policies, particularly the injuction against original research, which is exactly what this is. siafu (talk) 01:59, 23 August 2013 (UTC)

Least squares method has replaced trilateration method and Multidimensional Newton Raphson method

The least squares and the multidimensional Newton Raphson methods are very similar. In fact the method of updating the estimated solution from one iteration to the next appears to be identical for the case of four satellites. Thus we have algorithms, one developed from the discipline of numerical analysis and the other developed from the discipline of statistics which are very similar. One difference is that the multidimensional Newton Raphson method in its current form does not include the case of more than four satellites. Thus it appears that we can concentrate on the least squares method.

One thing I would like to point out is that the accusation that I engaged in personal research in writing the sections on the Multidimensional Newton Raphson method and the Trilateration method are false. I have formulated the problem to be solved and then selected the appropriate method from "Numerical Recipes" to obtain the solution. The Wikipedia guidelines allow the use of known facts to establish other facts and this is exactly what I have done. — Preceding unsigned comment added by RHB100 (talk • contribs) 18:13, 23 August 2013 (UTC)

Wikipedia does not rely on the expertise of editors, but rather the reliability of sources. To prove your point on the trilateration issue, the proper approach would be to look through the literature and find a source that confirms that it works, NOT to try and implement it yourself and show that it works, which is original research. Look at it this way; supposing you presented some gobbledegook FORTRAN code and fudged results that looked right. It would require other experts to look through all the code attempt to do it themselves to invalidate the claim, and such experts cannot be relied upon to exist here on wikipedia. siafu (talk) 18:53, 23 August 2013 (UTC)

I can't count the number of times that this one particular user RHB100 has tried to push his personal research into Wikipedia onto this article. The talk page archives talk for themselves. "Personal research" is not just an accusation — it is documented fact on the archives and history of Talk:Global Positioning System and User talk:RHB100. - DVdm (talk) 19:38, 23 August 2013 (UTC)

These accusations by DVdm are false. When he says that I have tried to push personal research into Wikipedia, he is a liar an outright liar and nothing but a liar. I stongly resent these false accusations by DVdm. He is an outright liar. DVdm through his ignorance doesn't understand that Wikipedia policy allows the use of mathematical facts to establish other mathematical facts. RHB100 (talk) 03:31, 24 August 2013 (UTC)

Goodness, is he still doing that after all these years? Maybe we should nominate him for a topic ban. Martijn Meijering (talk) 22:01, 23 August 2013 (UTC)

As for you, Martijn Meijering, you should find out the truth before making statements. When you say "is he still doing that", You are accusing me of having pushed personal research onto Wikipedia at some time in the past. Well you are an outright liar, I have never pushed personal research onto Wikipedia. Maybe we should nominate you as the outright liar of the week. You are an outright liar. I have never pushed personal research onto Wikipedia. RHB100 (talk) 03:31, 24 August 2013 (UTC)

Application of numerical methods is not personal research

In this era of high speed computers, the use of numerical methods is an indispensable tool of many engineers. Solution methods for many problems can be found in such books as "Numerical Recipes". Methods of solving the GPS navigation equations can be found in the chapter on root finding in this book. One such method is multidimensional Newton Raphson. In utilizing this method, the engineer may have to perform some work such as the evaluation of partial derivatives. When the engineer uses multidimensional Newton Raphson including the necessary evaluation of partial derivatives, the engineer is not performing personal research. The engineer is instead using a well known method to solve a given problem. The evaluation of partial derivatives is straightforward everyday work, there is no personal research involved.

Also the engineer may choose to use one dimensional Newton Raphson in conjunction with trilateration to solve the GPS navigation equations. This is a straightforward application of well known methods to a given problem. Again this is straightforward everyday work. There is no personal research involved.

We should therefore keep in mind that for the purposes of Wikipedia, the straightforward application of numerical methods to solve a given problem is certainly not personal research. — Preceding unsigned comment added by RHB100 (talk • contribs) 21:36, 24 August 2013 (UTC)

Doesn't work indoors

I have just recently acquired my first GPS device - a smartphone. For half an hour it has been lying on my desk helplessly trying to get a lock. That's something that I didn't know: GPS doesn't work inside buildings. It's a basic and important fact that is not in the current article, unless I missed it.

I do know that some wavelengths do pass through concrete or other materials, others don't. Some reflect on surrounding buildings, on cliffs, on water, others don't. GPS is in the 20-30cm range. Somewhere the transparency and reflection such waves should be specified.

More generally, is there a page that sums up in a practical way the transparency and reflection data for various wavelengths? Electromagnetic spectrum is too theory-oriented. I just want to know: will waves of a certain frequency enter buildings (through a direct route, through the windows?), will they pass the upper atmosphere (or perhaps be reflected), or through clouds, how far will they go underwater?

David Olivier (talk) 09:56, 5 September 2013 (UTC)

Most of your questions belong on our wp:reference desk/science — see wp:talk page guidelines. However, your remark that the article should mention the basic fact that GPS doesn't work inside buildings, is of course very on topic on this talk page. Go ahead, be wp:bold and fix it. Wikipedia is yours. DVdm (talk) 10:50, 5 September 2013 (UTC)

The wavelength of the GPS L1 carrier signal is 19 cm (1575 MHz). The primary reason GPS doesn't work indoors is that the signal is extremely weak in the first place-- it can be blocked by trees, fences, people standing in the way (happened numerous times doing data collection on Pike's Peak...), or even sometimes waving your hand in front of the antenna. Moreover, reflected waves are not helpful in GPS as they take a longer path than the one straight from the satellite to you, and will confuse the receiver (they are also weaker as well). There are, indeed, only certain wavelengths that penetrate the atmosphere ([12]), the L-band (1-2 GHz) was selected for GPS for this reason. If you're curious about these topics, I recommended the articles on penetration depth, path loss, and radio propagation. siafu (talk) 13:51, 5 September 2013 (UTC)

Thanks for your answer. Yes, I'm curious about such topics, generally, but that wasn't the point of my question, which was a practical one. Yes, I could be bold and add a section ("Availability", perhaps), but that would imply a lot of research on points that I guess most editors of this page already know. For instance: is the service available (with sufficient quality) at all latitudes? I gather from the section "3.1 Space segment" that it is, but that should be stated more clearly somewhere. OK, the signal is weak, but still some things will be more transparent than others. My experience is that it usually will work inside a bus. It doesn't work in most buildings, but does in my mother's house that is of wood with a tiled roof. But if I say that on the page someone will jump on me for inserting OR :D Will it work on a street surrounded by many skyscrapers? Also: does the weather affect the signal? Maybe I will go ahead sometime and be bold, and put in all I know and think I know and see what happens. David Olivier (talk) 13:58, 14 September 2013 (UTC)

To reduce the chances of being reverted for OR, find yourself some sources with this little tool —producing for instance this— and then be bold :-) - Cheers - DVdm (talk) 18:48, 14 September 2013 (UTC)

Section on Regulatory spectrum issues concerning GPS receivers is quite biased towards LightSquared

Having lived through the LightSquared testing while in the GPS industry, I find many omissions from the posting to make LightSquared appear to be a victim. Most notably, that GPS is first and foremost a military system that is owned and operated by the US Air Force and paid for by tax dollars from US citizens. It is not a private system. Military receivers would also be subject to jamming from a high-powered close-in signal, but obviously this was not widely discussed. Any major changes to the GPS system would be paid by citizens whereas new civilian receivers would have to be purchased by consumers. If LightSquared had been allowed to go forward, private GPS receiver companies actually would have made a lot of money by having to restock all GPS receivers in the US with new, more expensive receivers containing more complex filters. Customers outside of the US would still be able to use simple, inexpensive GPS receivers since the LightSquared contract is only for the US. It would have been painfully ironic if the GPS system, one of the greatest "public goods" (to use the economic term) provided by the US government, were to operate poorly in the US while operating very well in the rest of the world.

Other items that were left out of this section include: Section 911 of the 2012 Defense Appropriations Act, signed on Dec 31, 2011, forbade the FCC from lifting the prohibition against terrestrial operations (related to LightSquared) until concerns of interference with GPS have been resolved. Any proposed resolution must pass committees in the US House of Representatives and the US Senate. The spectrum that LightSquared owns is Mobile Satellite Service in the L-band, allocated by the FCC for satellite to earth communications. The FCC had a clause that allowed ATC or Ancillary Terrestrial Communications, but this was intended only to augment MSS, not to create a nationwide terrestrial network. LightSquared obtained the L-band spectrum via the purchase of SkyTerra in March of 2011 for $280M. This price was *far* below the price of terrestrial spectrum of similar bandwidth. In effect, LightSquared was attempting to buy terrestrial bandwidth at a satellite to earth price. If the LightSquared proposition was really financially sound, it would have been able to make money with their service by purchasing appropriate terrestrial bandwidth.

108.213.70.59 (talk) 04:55, 22 October 2013 (UTC)

I believe there was an article in GPS Solutions in 2012 that basically laid all this out, in fact. Don't have it to hand, though. siafu (talk) 05:16, 22 October 2013 (UTC)

section "Navigation Equations". a mistake in pseudorange equation?

currently:

p_i = \left ( t_\text{r} - t_i \right )c

shouldn't t_r be replaced by t_r tilde ? — Preceding unsigned comment added by 77.125.77.35 (talk) 10:17, 17 December 2013 (UTC)

Blacklisted Links Found on the Main Page

Cyberbot II has detected that page contains external links that have either been globally or locally blacklisted. Links tend to be blacklisted because they have a history of being spammed, or are highly innappropriate for Wikipedia. This, however, doesn't necessarily mean it's spam, or not a good link. If the link is a good link, you may wish to request whitelisting by going to the request page for whitelisting. If you feel the link being caught by the blacklist is a false positive, or no longer needed on the blacklist, you may request the regex be removed or altered at the blacklist request page. If the link is blacklisted globally and you feel the above applies you may request to whitelist it using the before mentioned request page, or request its removal, or alteration, at the request page on meta. When requesting whitelisting, be sure to supply the link to be whitelisted and wrap the link in nowiki tags. The whitelisting process can take its time so once a request has been filled out, you may set the invisible parameter on the tag to true. Please be aware that the bot will replace removed tags, and will remove misplaced tags regularly.

Below is a list of links that were found on the main page:

• http://web.archive.org/web/20080610092154/http://www.army-technology.com/contractors/computers/lago/lago6.html

  Triggered by \barmy-technology\.com\b on the local blacklist

• http://www.army-technology.com/contractors/computers/lago/lago6.html

  Triggered by \barmy-technology\.com\b on the local blacklist

If you would like me to provide more information on the talk page, contact User:Cyberpower678 and ask him to program me with more info.

From your friendly hard working bot.—cyberbot II ^Notify_Online 12:58, 3 April 2014 (UTC)

 Fixed See [13]. - DVdm (talk) 13:19, 3 April 2014 (UTC)

Accuracy?

This is an excellent article, but I came to this page hoping to find out how accurate GPS is, and cannot find this information which I would have thought basic. I can see that there are sections and linked articles on improving accuracy, but a top-level summary of accuracy would be very useful. Even better would be a summary of how the accuracy of GPS has changed during its development, and/or how accuracy is affect by the number of satellites viewed, how accurate can it be? Also, how does GPS accuracy compare with Galileo/ Glonass etc? Gebjon (talk) 10:51, 9 April 2014 (UTC)

I have added some basic figures in the section Global Positioning System#Accuracy enhancement and surveying. Feel free to hone and add more. - DVdm (talk) 11:32, 9 April 2014 (UTC)

Marketing literature

"Built on a flexible architecture that can rapidly adapt to the changing needs of today's and future GPS users allowing immediate access to GPS data and constellations status through secure, accurate and reliable information."

This reads like marketing literature. — Preceding unsigned comment added by 5.67.191.234 (talk) 08:45, 21 October 2014 (UTC)

Please sign your talk page messages with four tildes (~~~~). Thanks.

I have somewhat redacted the layout of your section—hope you don't mind. Wikipedia is yours, so feel free to demarketise. - DVdm (talk) 09:07, 21 October 2014 (UTC)

Proposed Changes to 'Basic concept of GPS'

As currently written, this section does not agree with my understanding of the basic GPS concept. I propose to change the first portion of it as follows. I'm soliciting feedback before making changes to the article. It needs references, which I will work on providing.

Using signals received from four GPS satellites above the Earth (approximately 10,000 nautical miles), a GPS receiver calculates (a) its three-dimensional position, and (b) the offset of its clock from GPS system time used by the satellites. Each GPS satellite continually broadcasts a signal (carrier frequency with modulation) that include:

• a pseudorandom code, from which the time of arrival (TOA) of a code epoch can be found (in the receiver clock time scale)
• a message the include the time of transmission of the epoch (in GPS system time scale) and the satellite position at that time.

Conceptually, the receiver measures the TOAs (according to its own clock) of four satellite signals. Since the signals are continuous, the TOA is associated with a code epoch. From the four TOAs, it forms three time differences of arrival (TDOAs). The receiver then computes its three-dimensional position from the three TDOAs. Each TDOA defines a hyperboloid (see Multilateration), so the receiver is located at the point where the three hyperboloids intersect.

Given it position and those of the satellites, the receiver can associate each TOA with a specific satellite (e.g., the smallest TOA is associated with the nearest satellite). Using the speed of light and the distance to a satellite, the receiver computes the transmission time of the epoch in the receiver clock time scale. Since the satellite broadcast message includes the epoch transmission time in GPS system time, the receiver computes the offset between its time scale and GPS system time scale.

This description is conceptual. In practice the receiver position (in three dimensional Cartesian coordinates with origin at the earth's center) and code epoch time of transmission (according to the receiver clock) can be computed either sequentially (as described) or simultaneously, using the navigation equations.

The receiver's earth-centered solution location is usually then converted to latitude, longitude and height relative to an ellipsoidal earth model. The height may then be further converted to height relative the geoid (e.g., EGM96) (essentially, mean sea level). These coordinates may be displayed, perhaps on a moving map display and/or recorded and/or used by other system (e.g., vehicle guidance).

Basic GPS measurements yield ... CONTINUE CURRENT ARTICLE

--NavigationGuy (talk) 10:14, 5 November 2014 (UTC)

I would not call that basic. Perhaps you could insert that as a new More advanced details section after the basic section? In my opinion, there has been little regard for the lay reader trying to understand how the GPS fundamentally works. Over time, well-meaning editors have enhanced the basic section with more and more technically correct detail. A more fundamental explanation would be:

GPS satellites transmit data continuously which contains their current time and position.

GPS receiver listens to multiple satellites and solves complex equations to determine the exact time of day, distance to each satellite, and from that the exact position of the receiver.

—EncMstr (talk) 19:30, 5 November 2014 (UTC)

--NavigationGuy (talk) 11:46, 6 November 2014 (UTC)Thanks. Your point is valid. I'm also soliciting comments from co-workers.

NavigationGuy (talk) 12:04, 7 November 2014 (UTC)Second Shot. As currently written, this section does not agree with my understanding of the basic GPS concept. I propose to change the first portion of it as follows. I'm soliciting feedback before making changes to the article. It needs references, which I will work on providing.

--NavigationGuy (talk) 12:13, 7 November 2014 (UTC)

Proposed Changes to 'Basic concept of GPS' (version 2)

Fundamentals

The GPS system concept is based on time. The satellites carry atomic clocks which are synchronized, and their locations are known precisely. User receivers have clocks as well, but they are not synchronized with the satellites. They are less stable and only capable of measuring differences in time between signals received from satellites. GPS satellites transmit data continuously which contains their current time and position. A GPS receiver listens to multiple satellites and solves equations to determine the the exact position of the receiver and the exact time of day. At a minimum, four satellites must be in view of the receiver in order to compute four unknown quantities (three position coordinates and time).

--NavigationGuy (talk) 12:25, 7 November 2014 (UTC)

More Detailed Description

Each GPS satellite continually broadcasts a signal (carrier frequency with modulation) that include:

• a pseudorandom code (sequence of ones and zeros) that is known to the receiver. By time-aligning a receiver-generated version and the received version of the code, the time of arrival (TOA) of a defined point in the code sequence, called an epoch, can be found in the receiver clock time scale
• a message that includes the time of transmission of the code epoch (in GPS system time scale) and the satellite position at that time.

Conceptually, the receiver measures the TOAs (according to its own clock) of four satellite signals. From the TOAs, the receiver forms three time differences of arrival (TDOAs), which are (given the speed of light) equivalent to receiver-satellite range differences. The receiver then computes its three-dimensional position from the three TDOAs .

Given its position and those of the satellites, the receiver can associate each TOA with a specific satellite (e.g., the smallest TOA is associated with the nearest satellite). Using the speed of light and the distance to a satellite, the receiver computes the transmission time of a code epoch in the receiver clock time scale. Since the satellite broadcast message includes the epoch transmission time in GPS system time, the receiver computes the offset between its time scale and GPS system time scale.

This description is conceptual. In practice the receiver position (in three dimensional Cartesian coordinates with origin at the earth's center) and the offset of the receiver clock relative to the satellite clocks are computed simultaneously, using the navigation equations to process the TOAs. The TDOAs are not explicitly formed.

The receiver's earth-centered solution location is usually converted to latitude, longitude and height relative to an ellipsoidal earth model. The height may then be further converted to height relative the geoid (e.g., EGM96) (essentially, mean sea level). These coordinates may be displayed, perhaps on a moving map display and/or recorded and/or used by other system (e.g., vehicle guidance).

--NavigationGuy (talk) 12:33, 7 November 2014 (UTC)

User-Satellite Geometry

Although usually not formed explicitly in the receiver processing, the conceptual TDOAs define the measurement geometry. Each TDOA corresponds to a hyperboloid of revolution (see Multilateration). The line connecting the two satellites involved (and its extensions) forms the axis of the hyperboloid. The receiver is located at the point where three hyperboloids intersect.

It is sometimes incorrectly said that the user location is at the intersection of three spheres. While simpler to visualize, this is only the case if the receiver has a clock synchronized with the satellite clocks (i.e., the receiver measures true ranges to the satellites rather than range differences). There are significant performance benefits to the user carrying a clock synchronized with the satellites. Foremost is that only three satellites are needed to compute a position solution. If this were part of the GPS system concept so that all users needed to carry a synchronized clock, then a smaller number of satellites could be deployed. However, the cost and complexity of the user equipment would increase significantly. --NavigationGuy (talk) 12:51, 7 November 2014 (UTC)

Receiver in Continuous Operation

The description above is representative of a cold-start situation. Most receivers have a tracker algorithm that, in effect, combines sets of satellite measurements collected at different times. After a set of measurements are processed, the tracker predicts the receiver location corresponding to the next set of satellite measurements. When the new measurements are collected, the receiver uses a weighting scheme to combine the new measurements with the tracker prediction. In general, a tracker can (a) improve receiver position and time accuracy, (b) reject bad measurements, and (c) estimate receiver speed and direction.

The disadvantage of a tracker is that changes in speed or direction can only be computed with a delay, and that derived direction becomes inaccurate when the distance traveled between two position measurements drops below or near the random error of position measurement. GPS units can use measurements of the doppler shift of the signals received to compute velocity accurately. More advanced navigation systems use additional sensors like a compass or an inertial navigation system to complement GPS.

--NavigationGuy (talk) 12:51, 7 November 2014 (UTC)

Other Applications

In typical GPS operation as a navigator, four or more satellites must be visible to obtain an accurate result. The solution of the navigation equations gives the position of the receiver along with the difference between the time kept by the receiver's on-board clock and the true time-of-day, thereby eliminating the need for a more precise and possibly impractical receiver based clock. Applications for GPS such as time transfer, traffic signal timing, and synchronization of cell phone base stations, make use of this cheap and highly accurate timing. Some GPS applications use this time for display, or, other than for the basic position calculations, do not use it at all.

Although four satellites are required for normal operation, fewer apply in special cases. If one variable is already known, a receiver can determine its position using only three satellites. For example, a ship or aircraft may have known elevation. Some GPS receivers may use additional clues or assumptions such as reusing the last known altitude, dead reckoning, inertial navigation, or including information from the vehicle computer, to give a (possibly degraded) position when fewer than four satellites are visible.

Please refrain from including unsourced material in the article. Fgnievinski (talk) 20:09, 8 November 2014 (UTC)

Solution based on intersection of at least four spheres, not TDOA and not Multilateration

Position is based on the intersection of four or more spheres. See the navigation equations. These equations describe spheres. The solution has nothing to do with hyperboloids or multilateration. RHB100 (talk) 05:20, 19 January 2015 (UTC)

The navigation equations to be solved are:

$(x-x_{i})^{2}+(y-y_{i})^{2}+(z-z_{i})^{2}={\bigl (}[{\tilde {t}}_{\text{r}}+b-t_{i}]c{\bigr )}^{2},\;i=1,2,\dots ,n$

or in terms of pseudoranges, $p_{i}=\left({\tilde {t}}_{\text{r}}-t_{i}\right)c$, as

$p_{i}={\sqrt {(x-x_{i})^{2}+(y-y_{i})^{2}+(z-z_{i})^{2}}}-bc,\;i=1,2,...,n$ .

These equations describe spheres. The solution of these equations is at the intersection of n spheres with the necessary clock bias. The writer of the section, User-satellite geometry, does not seem to be aware of this fact. RHB100 (talk) 18:21, 19 January 2015 (UTC)

The statement below quoted from "User-satellite geometry" seems to be incompatible with the navigation equations. The time for receiver clocks is synchronized with the satellite clocks as a part of the solution process. It is certainly true that the receiver location is at the intersection of 3 spheres since it is at the intersection of n spheres where n is greater than 3. "It is sometimes incorrectly said that the user location is at the intersection of three spheres. While simpler to visualize, this is only the case if the receiver has a clock synchronized with the satellite clocks (i.e., the receiver measures true ranges to the satellites rather than range differences). There are significant performance benefits to the user carrying a clock synchronized with the satellites. Foremost is that only three satellites are needed to compute a position solution. If this were part of the GPS system concept so that all users needed to carry a synchronized clock, then a smaller number of satellites could be deployed. However, the cost and complexity of the user equipment would increase significantly." Licensed Professional Engineer RHB100 (talk) 19:41, 21 January 2015 (UTC)

Also in ""User-satellite geometry" it is stated, "Although usually not formed explicitly in the receiver processing, the conceptual time differences of arrival (TDOAs) define the measurement geometry." If it is not a part of receiver processing don't confuse people by mentioning it. RHB100 (talk) 19:54, 22 January 2015 (UTC)

The "Basic concept" section is getting pretty disjoint. It should probably be re-arranged and re-sectioned. "User-satellite geometry" now has little to do with geometry, for example, and clock info is split between that section and "Non-navigation applications". Kendall-K1 (talk) 21:59, 22 January 2015 (UTC)

Can anyone explain how the statement, "It is sometimes incorrectly said that the user location is at the intersection of three spheres", is compatible with the navigation equations? The navigation equations are the equations stated above and in the Navigation equations section of the article. The navigation equations appear to be clearly equations describing spheres. A solution of these n equations is at the intersection of these n spheres. So clearly it would seem this solution is at the intersection of three spheres. But someone says, this is incorrect. Why? Can anyone explain this? RHB100 (talk) 21:17, 23 January 2015 (UTC)

Primarily, four (n >= 4) ranging measurements are necessary, since the receiver must resolve its internal clock bias, so "3 spheres" doesn't really make sense when you're dealing with a 4-dimensional problem. Secondly and more pedantically, the idea of a sphere, here representing a wavefront (of constant phase) centered on a satellite is also inaccurate as the propagation is very definitely anisotropic because of the effects of the atmosphere and other effects (e.g. relativity). The "intersection of 3 spheres" idea is a handy teaching tool, but does not actually represent or explain how GPS works. siafu (talk) 22:11, 23 January 2015 (UTC)

Thank you but what you are telling me is what I already know and have stated above in the subject that n is at least 4. All you have done is make some comments which have nothing to do with the question that was asked. I understand that there are errors in signal arrival time measurement as discussed in Error Analysis for the GPS at [14]. This takes care of the anisotropic propogtion. You have taken "intersection of 3 spheres" out of the context in which the question was asked and your comments on it have nothing to do with the question that was asked? RHB100 (talk) 01:58, 24 January 2015 (UTC)

Now let me make this clear, Siafu. I am a Licensed Professional Engineer in the field of Control System Engineering. I hold advanced engineering degrees from both the University of Arkansas and UCLA. Because of the fact that I have been educated at better quality engineering schools such as the University of Arkansas, I clearly understand that the GPS navigation equations consist of n (n greater than 3) equations for the surfaces of n spheres and that the solution of these n equations consists of a clock bias and a point at the intersection of these n sphere surfaces. I furthermore understand that a point at the intersection of more than 3 sphere surfaces is clearly at the intersection of 3 sphere surfaces. Are you able to comprehend these facts, siafu? RHB100 (talk) 01:58, 24 January 2015 (UTC)

And here I thought giving you a year to stew might mean that you wouldn't try to pull out your supposed credentials again the instant you get confused. Congratulations on your advanced degrees; I have some of those too, and they're actually relevant to the field of GPS. It seems that despite knowing the reasons why describing the solution to the navigation equations as the intersection of three spheres is incorrect, as I just noted, you claim to not understand why describing the solution to the navigation equations as the intersection of three spheres is incorrect. If I recall correctly, you also didn't know what a batch filter was, so I guess that's really all there is to say about that. siafu (talk) 13:43, 24 January 2015 (UTC)

Credentials don't matter much here. Wikipedia content is based on verifiable information from reliable sources. We also treat each other with respect and civility. See Wikipedia:Five pillars. If you want to make a particular point, you can't just write something up based on your own knowledge, you must find a source that supports what you want to say, then incorporate material from that source. Kendall-K1 (talk) 15:47, 24 January 2015 (UTC)

That's credentials revisited from September 2012—see closing comments at the end of this archived section. - DVdm (talk) 16:58, 24 January 2015 (UTC)

Now here you, Siafu, go misquoting me again. I did not describe the solution of the navigation equations as the intersection of three spheres. I said, "I clearly understand that the GPS navigation equations consist of n (n greater than 3) equations for the surfaces of n spheres and that the solution of these n equations consists of a clock bias and a point at the intersection of these n sphere surfaces". This of course implies that the position part of the solution is at an intersection of any three of the n spheres but this is not a description of the solution but instead an implication of the solution. RHB100 (talk) 17:58, 24 January 2015 (UTC)

Kendall, credentials certainly do matter since if you don't have the proper education then you can't understand the material and are certainly unqualified to write. Also, Kendall, pride is a virtue. I am very proud that I have received a superior quality of education and that I have had a highly successful engineering career. The fact that I have the great virtue of pride does not mean that I do not treat others with respect and civility. RHB100 (talk) 17:58, 24 January 2015 (UTC)

DVdm, you start your post with the word, "That's". This usage of the word "that" is very vague and ambiguous in this context. Also you need to say what you are talking about rather than make some reference. RHB100 (talk) 17:58, 24 January 2015 (UTC)

Woodstone, you have removed important material from the article section, "Problem Description". You removed a statement that the equations to be solved are not equations for a hyperbola. Rather than state what you had done, you gave the vague description, removed contrarian remark. This was a terrible edit on your part because not only is it absolute truth that the equations to be solved are not equations for a hyperbola, it is extremely important that the readers be told this fact. The reason for this is because there is much discussion relating to hyperbolas in the section which is quite likely to mislead the reader into believing that the equations to be solved are equations for a hyperbola. The fact that you only say, removed contrarian remark, indicates that you were very superficial in your thinking. It is therefore concluded that we do need a statement alerting readers to the fact that the equations to be solved are not equations for a hyperbola. RHB100 (talk) 18:30, 26 January 2015 (UTC)

Wikipedia is not in the business of alerting readers. This is a blatant example of wp:EDITORIALIZING. I have removed it. - DVdm (talk) 18:47, 26 January 2015 (UTC)

DVdm, You say "This is a blatant example of wp:EDITORIALIZING", but you don't say what you mean by "This". What specific statement that I made are you referring to when you say "This". I don't believe that I made any statement that could be classified as blatant editorializing. I have taken a neutral point of view. RHB100 (talk) 20:10, 26 January 2015 (UTC)

Read wp:EDITORIALIZING and see if you can find how it relates to what you just wrote here on talk and here in the article. - DVdm (talk) 21:01, 26 January 2015 (UTC)

You have made this vague accusation that I have engaged in blatant editorializing. You are unable to come up with any specific statement I have made that shows anything vaguely resembling blatant editorializing and you tell me to go read wp:EDITORIALIZING and figure out what it was that you were talking about. Well I have read it and I maintain I am innocent of the accusations you have made. Furthermore you are now accusing me of what you call, blatant editorializing on the talk page. So according to you, a discussion on the talk page cannot take place without being subject to having someone like you make the accusation of blatant editorializing. Well I think you are wrong, this wp:EDITORIALIZING doesn't even apply to the talk page, it only applies to the writing of the article. I still maintain that I am innocent of engaging in anything remotely resembling blatant editorializing. You have not been able to come up with a single statement that I have written that in anyway resembles blatant editorializing. You are guilty of having made totally and completely false accusations against me. RHB100 (talk) 00:36, 27 January 2015 (UTC)

It's pretty obvious to me that it's editorializing. You really don't see that? Compare "The reader should note" to "notably" or "it should be noted." If anything, "The reader should note" seems like stronger editorializing than the examples from the MOS. Kendall-K1 (talk) 01:17, 27 January 2015 (UTC)

Notably is an adverb which carries the connotation of something which is outstanding. The word note when used in "The reader should note" is a verb which carries no such connotation. The two words have completely different meanings. wp:EDITORIALIZING talks about the use of adverbs. It doesn't say anything about the use of verbs. There is nothing wrong with the use of the verb note in "The reader should note". RHB100 (talk) 17:58, 27 January 2015 (UTC)

Again, blatant wp:EDITORIALIZING, and on top of that:, whether "the reader should observe" is wp:UNSOURCED. Reverted for obviouis reasons. - DVdm (talk) 18:24, 27 January 2015 (UTC)

My god, every time I make statements on the talk page, you call it blatant editorializing. When I say "notably is an adverb" and "note is used as a verb" you don't respond with an agreement or disagreement. You just holler "blatand editorializing". RHB100 (talk) 19:03, 27 January 2015 (UTC)

I am referring to this, in the article, not on the talk page. - DVdm (talk) 20:07, 27 January 2015 (UTC)

back to the subject

If there were no receiver clock errors, the simplest solution would indeed be at the intersection of three spheres. The intersection of four spheres is not a single point unless the fourth sphere is a duplicate of one of the other three unique spheres. With the presence of a receiver clock error (still no satellite clock errors), the spheres have unknown radii, so the original geometrical picture is no longer valid. It turns out a geometrical representation is still possible, with the introduction of hyperboloid surfaces. See sec. 2 in [15] for details. All of the above neglects moving transmitters/receivers, atmospheric delays, and more esoteric relativity corrections. As for the terminology, we need a source stating that the use of absolute ranges is called trilateration and the use of pseudo-ranges (clock-biased ranges) is called multilateration. When more than the minimum number of observations are available, there are multiple possible solutions, and a unique one can be found via least squares; thus least squares triangulation if using absolute ranges and least-squares mutilateration if using pseudo-ranges. The min number of obs would be two on a plane and three in three-dimensional space. Fgnievinski (talk) 05:00, 27 January 2015 (UTC)

Another way to look at the equations is considering them as virtual time travel following the clock bias. So there are not fixed spheres, but expanding bubbles. For some values of the bias the bubbles do not touch each other and there is no solution. For some values of the bias the bubbles are so large that they practically coincide and the solution becomes very unstable. Solving the system is to find the bias value where the solution is unique, or more usual, to find a location and a bias that minimise the distance from all bubbles. −Woodstone (talk) 10:47, 27 January 2015 (UTC)

The two main problems with the "intersection of spheres" model is that a) the equations describe not spheres, but irregular 4-D hypersurfaces with non-constant radii, and b) in actual practice solving the navigation equations is really a matter of performing a least squares minimization on an overdetermined set of noisy pseudoranges-- that is, the surfaces described by the equations don't actually intsersect properly when you try to solve them with actual pseudorange measurements. Really, I think the whole idea of trying to describe the solution as the "intersection of spheres" can get very complicated with caveats if we insist on it being a true statement in any way; it's only useful as an educational tool. As such, it would be better to start with explaining the concept of trilateration, and then expand to the more complicate real situation by noting the similarity. This way, we don't have to justify the geometrical as valid, since it isn't, and we don't have to give up the illustrative power of the analogy. siafu (talk) 13:57, 27 January 2015 (UTC)

Siafu, according to the sources we have referenced, these equations for sphere surfaces are what we actually solve to determine receiver position and clock bias. According to our documentation, these are the equations we solve. They are not good for education only, they are actually used in the GPS receiver software. RHB100 (talk) 22:21, 27 January 2015 (UTC)

I think all of these discussions above are making the issue much more confusing and difficult to understand than it is. Based on the sources we reference, the equations we solve are

$(x-x_{i})^{2}+(y-y_{i})^{2}+(z-z_{i})^{2}={\bigl (}[{\tilde {t}}_{\text{r}}+b-t_{i}]c{\bigr )}^{2},\;i=1,2,\dots ,n$

or in terms of pseudoranges, $p_{i}=\left({\tilde {t}}_{\text{r}}-t_{i}\right)c$, as

$p_{i}={\sqrt {(x-x_{i})^{2}+(y-y_{i})^{2}+(z-z_{i})^{2}}}-bc,\;i=1,2,...,n$

We find the simultaneous solution of these n equations. This means we find an (x, y, z, b) which satisfies all of these equations. There may be some approximation in the solution of these equations, but it is sufficiently close. Physically we know we are at the intersection of n spheres since we know we are at some radial distance from each of the n satellites, thereby proving existence. The problem is to correct the error in the radial distance to each of the satellites. This is done as part of the solution process. Uniqueness for n greater than or equal to 4 assures us that when we find a solution, it is the receiver position at least approximately. Since the equations for sphere surfaces are what we solve, we could completely eliminate all the discussion of hyperbolas, hyperboloids, multilateration, etc. This would make the section much more straightforward and much easier for the reader to comprehend. RHB100 (talk) 18:39, 27 January 2015 (UTC)

The use of spheres for visualization is amply sourced so it deserves to be mentioned in this article. The use of hyperboloids is less common though still well sourced, so I think it deserves mention as well. @Woodstone:, the bubbles looks interesting; can you source it? @Siafu:, visualizations are pedagogical tools for the defining mathematical equations; it's not meant to be a replacement. I'm okay referring to the sphere-intersection as trilateration, or vice-versa. Likewise, multilateration and the hyperboloid-intersection problem. @RHB100: the equations are for x, y, z, b; the spheres/hyperboloids only appear as an optional interpretation. Furthermore, the spherical interpretation only applies to true ranges not to pseudo-ranges. And it is not correct that more than three true ranges necessarily intersect at a unique point. If you have n true ranges, you can take 3 elements at a time, and each combination will produce a different intersection. So for 4, 5, 6, 7 true ranges, there are respectively up to 4, 10, 20, 35 unique solutions. The over-determined least-squares solution amounts to a spatial average of such combinatorial solutions, average which does not necessarily lies on the surface of any of the spheres. Again, the geometrical visualizations are in addition to the equations. If you don't find the geometrical visualizations useful, that only means you're an analytical thinker; we should assumed a more diverse readership. Can we try and agree on the order of presentation? I propose: true-range equation; system of 3 such equations; trilateration; sphere-intersection; more than 3 true range equations; least-squares trilateration; average sphere-intersection. pseudo-range equation; system of 4 such equations; multilateration; hyperboloid-intersection; more than 4 pseudo-range equations; least-squares multilateration; average hyperboloid-intersection. Fgnievinski (talk) 01:13, 29 January 2015 (UTC)

Fgnievinski, trilateration is not one of the solution methods we reference. I have documented these in the past but some have said they don't use them. The solution methods we reference are Least Squares and Bancroft method. Besides the Bancroft method is much better. Read the online paper on the Bancroft method. This will gives you a much better understanding of how GPS works. I made drawings showing how two sphere surfaces intersected in a circle, three sphere surfaces intersected at two points, and four sphere surfaces intersect at one point in the past but someone took them down. RHB100 (talk) 01:47, 29 January 2015 (UTC)

Fgnievinski, four sphere surfaces intersect at least approximately. Both the Bancroft and Least Squares methods are based on the at least approximate intersection of four or more sphere surfaces. The fact that the speed of light may not be the same in all direction does not invalidate the use of modelling the receiver as being on the surfaces of n spheres. Simplified models of reality are quite sufficient. To get an understanding of how GPS works, read the online paper on the Bancroft method. If everybody reads this paper, I think we all can get a much better understanding of how GPS works and we will be able to decide what information we need to present to the readers. RHB100 (talk) 01:47, 29 January 2015 (UTC)

I think to make sure we all understand how GPS works, it is necessary that we all read the paper on Bancroft's method at paper on Bancroft's method. Even if we decide to use the Least Squaress method instead, understanding the Bancroft method will give us an understanding of how GPS works. Unless we do this, I don't think we will have adequate understanding of how GPS works to explain it to our readers. RHB100 (talk) 03:02, 29 January 2015 (UTC)

The repeated mention of Bancroft's method is a mistake; this is not a method that is actually used in practice. The least squares minimization is. I suggest, RHB100, that you familiarize yourself with the actual behavior of GPS receivers before attempting to lecture everyone else on the topic. siafu (talk) 14:16, 29 January 2015 (UTC)

The classic book Linear Algebra, Geodesy, and GPS by Strang & Borre says (p.499) [16]: "To find the absolute position of a point is a very fundamental problem in positional GPS. We already have mentioned several methods to achieve the goal. We shall deal with one more method which is described by Bancroft (1985)." So Bancroft's is a nice-to-have method, not necessarily the most basic, common, or best method. As for the spheres and hyperboloids, the same authors offer this popularization of science article in SIAM News (1997): [17]. Fgnievinski (talk) 20:00, 29 January 2015 (UTC)

Content Dispute

There is a content dispute about wording in the article. There are minor civility violations, but nothing that should stop use of the dispute resolution policy. There is no need to go to WP:ANI at this time. If talk page discussions do not work, follow a procedure described in the policy, such as use of the dispute resolution noticeboard or a Request for Comments. Robert McClenon (talk) 22:28, 28 January 2015 (UTC)

DVdm, the Manual of Style says: There are no forbidden words or expressions on Wikipedia, but certain expressions should be used with care, because they may introduce bias. Strive to eliminate expressions that are flattering, disparaging, vague, clichéd, or that endorse a particular point of view.

The advice in this guideline is not limited to the examples provided and should not be applied rigidly.

I am not attempting to editorialize. I am just trying to let readers know, we have made a sudden shift from equations for hyperboloids to equations for surfaces of spheres. If you can think of a better way of accomplishing this, please let me know. RHB100 (talk) 02:08, 29 January 2015 (UTC)

How about changing "These equations can be solved by algebraic or numerical methods" to "These equations describe the surfaces of spheres and can be solved by algebraic or numerical methods"? Kendall-K1 (talk) 04:19, 29 January 2015 (UTC)

Alright that sounds like a good, intelligent suggestion, Kendall-K1 . RHB100 (talk) 06:05, 29 January 2015 (UTC)

Of course. - DVdm (talk) 07:55, 29 January 2015 (UTC)

Except when it's not: it's spheres fox xyz and hyperboloids for xyzb (Strang & Borre, 1997). Fgnievinski (talk) 20:12, 29 January 2015 (UTC)

Not quite. For fixed b they describe spheres. For concurrent variables x, y, z, b they are more like a 4-dimensional spherical cones. −Woodstone (talk) 15:53, 29 January 2015 (UTC)

You mean hyperboloids. First you gave bubbles now it's cones; can you source any of that? Fgnievinski (talk) 20:12, 29 January 2015 (UTC)

Cones or growing bubbles are two consistent views on the 4D equations. If the 4th dimension is seen as a timeline, the view of expanding bubble is appropriate. If one considers just an abstract 4D space, then a cone is more adequate. Compare the 3D case. Look at a geometric circular cone. If the height (3rd dimension) of the cone is seen as timeline, then moving away from the top, one sees a growing circle. I cannot source these views, but they are straightforward mathematical interpretations of the equations. The view mentioning hyperboloids is clearly valid, but does not represent each equation as written. It is a construct using pairs of satellites together. That introduces redundancy, since there are 6 pairs (out of 4 satellites), while just 3 define a unique intersection. Similarly one could look at the solutions for a particular clock bias (if one exists) for each combination of spheres around 3 satellites. For a range of biases there will be two solutions for every combination. By varying the bias, half of the solutions will come close together and the closest we can bring them is called the solution. In actual practice using combinations of satellites is impractical, especially since with the advent of the European satellites, there are often over 10 in view, but also because some combinations may yield very unstable solutions (satellites coplanar, are at close angles). So the only feasible approach is a concurrent least squares solution, which does not admit interpretation as spheres or hyperboloids. −Woodstone (talk) 15:21, 30 January 2015 (UTC)

@Woodstone: Unless you can source any of your interpretations, there's no need to spend any time explaining it here; WP:Verifiability, please. Fgnievinski (talk) 19:48, 30 January 2015 (UTC)

I found several quite explicit sources for the spherical cones:

• here
• there is also this one
• and this
• there are many more, do a search for 'gps spherical cone'.

−Woodstone (talk) 16:17, 31 January 2015 (UTC).

And of course I would not be surprised (if I may editorialise here on talk, while not having done the maths) if the intersection of two spherical cones is a hyperboloid. −Woodstone (talk) 17:10, 31 January 2015 (UTC)

Sourcing "User-satellite geometry"

Section GPS#User-satellite geometry would benefit from a rewrite. We got explicit sources for the following interpretations: (a) 3D spheres for true-ranges [18], (b) 3D hyperboloids for pseudo-ranges [19], and (c) 4D hypercones for pseudo-ranges [20], [21], [22]. How about this:

While the GPS equations can be solved directly by numerical and analytical methods, geometrical interpretations are possible. At least three cases of geometrical analogy exist. In case (a) measured ranges are assumed synchronized; the locus of the minimum solution lies at the intersection of spheres in three-dimensional space (the everyday "physical" space where positions are described); each sphere is centered at one of three broadcasting satellites. In the case (b), measured ranges are allowed to be non-synchronized (pseudo-ranges), corrupted by receiver clock errors; the locus of minimum solution can then be found at the intersection of three hyperboloids in three-dimensional space; each hyperboloid is aligned with a different satellite pair, for a total of four satellites. In case (c), the same four pseudo-range observations form a hypercone in four-dimensional spacetime (the three positional dimensions augmented with a temporal dimension). These cases use the minimum number of observations (3 synchorized ranges and 4 pseudo-ranges); over-determined solutions are found via least squares and do not have an exact geometrical interpretation.

Please try to wikilink as much as possible. Fgnievinski (talk) 02:47, 1 February 2015 (UTC)

I think this is written in a style more appropriate for journal submission. I would suggest smoothing it out a bit, e.g.:

While the GPS equations can be solved directly by numerical and analytical methods, geometrical interpretations are possible, and at least three cases of geometrical analogy exist. In a simplified case in which the measured ranges are synchronized, they represent the radii of spheres, each centered on one of the transmitting satellites. The solution, the position of the receiver, is then at one of the two intersections of three of these spheres. However, this case assumes the receiver clock bias is known; including the unknown receiver clock bias allows the ranges to become unsynchronized. In this case, the solution lies at the intersection of three hyperboloids each aligned with a different satellite pair amongst four satellites. Alternately, the same case can be seen as representing a hypercone in four-dimensional spacetime (three spatial dimensions plus time).

In all these cases, only the minimum number of satellites, four, are considered. In reality, many more than four satellites are typically in view leading to a system that is over-determined and cannot be solved analytically. Additionally, the effects of relativity, space weather, and the atmosphere cause the signals to propagate differently in different directions such that the surfaces discussed above are not actually spheres but irregular shapes. As a result, solving the GPS positioning problem is typically via a least squares minimization that does not have an exact geometrical interpretation.

Better? Worse? siafu (talk) 14:35, 1 February 2015 (UTC)

Contrary to what the writer seems to be saying above, the navigation equations can be solved analytically when there are more than 4 satellites with the Bancroft method by using the pseudo-inverse matrix. RHB100 (talk) 20:13, 1 February 2015 (UTC)

I agree with the contention (although notice that "The pseudoinverse provides a least squares solution to a system of linear equations" , see Moore–Penrose pseudoinverse#Linear least-squares); I also think that for this particular section, User-satellite geometry, we should not elaborate about errors sources other than rx clock bias. So how about this. Fgnievinski (talk) 00:19, 3 February 2015 (UTC)

These geometrical analogies consider only the minimum number of satellites, three or four (depending on whether a receiver clock bias is allowed or not). In reality, more than four satellites are typically in view leading to a system that is over-determined, in which case the geometrical analogy is no longer valid. As a result, solving the GPS positioning problem is typically via a least squares minimization, although other methods also exist (see #Closed-form solution methods).