Wikipedia:Peer review/Global Positioning System/archive1

This peer review discussion has been closed.
I've listed this article for peer review because there are plans to nominate the article, Global Positioning System, for Wikipedia Featured Article. This article has undergone considerable improvement over the past year or so with several new sections added and old ones updated. These improvements having been made, the article is now in a rather stable state with only minor changes being made on a day to day basis. We now have sections on position calculation (introductory and advanced), clock correction, and others as indicated in the table of contents. If you would like to comment on section 2 Basic concept of GPS, section 3.2 Navigation signals, section 3.4 C/A code, section 3.6 Error sources and analysis including the subsections, other sections which interest you, or make overall comments, it would be greatly appreciated. I think you will find the material interesting but we are interesting in hearing your opinions. Thanks, RHB100 (talk) 21:27, 30 July 2009 (UTC)[reply]

  • Titoxd's comments - In the lede:
    • It is the only fully functional GNSS in the world, can be used freely (unless technically restricted by the U.S. Department of Defense) - the "technically restricted" part seems a bit imprecise. Do you mean Selective Availability, the new M-codes, the anti-spoof or jamming?
    • ... satellites that transmit precise radiowave signals,... - you can afford to be more precise here. Use UHF instead.
    • Although NAVSTAR is not an acronym, a few backronyms have been created for it. - is this sentence really necessary? And the reference for the backronyms doesn't seem to pass WP:RS for starters...
    • The lede is rather short for an article of this size. A paragraph summarizing the history of the system should be added to the lede.
  • History section
    • In the 1970s, the ground-based Omega Navigation System, based on signal phase comparison, - jargon. Either more development here or a link would be nice...
    • GPS requires the equations of general relativistic corrections to these atomic clocks in satellites for sufficient accuracy. - this just sounds grammatically wrong. "Equations to correct relativistic effects based on the theory of general relativity" would be better, I think.
    • The general relativistic equations correcting signals of atomic clocks in satellites were first published in 1956 by Friedwardt Winterberg - be more precise in the reference (you may want to use one of the citation templates), include publisher, author, etc. Also, full stops go before the reference, not the other way around (that is buried somewhere in WP:MOS as well).
    • The design of GPS is based partly on similar ground-based radio navigation systems, such as LORAN and the Decca Navigator developed in the early 1940s, and used during World War II. - this sentence would flow better if it was split.
    • There is a big old comment in the source code: The converse is also true: if the satellite's position were known, they could identify their own position on Earth. (commented because I am not sure of this. At most, they would know the rate at which the distance between themselves and the satellite was changing. There would be at least two points (one each north and south of the equator) for which that would be true, and practically one would not get an exact position, especially with 1950s electronics, even if one knew the satellite's exact orbit, and the exact time - has that been resolved?
    • The entire second paragraph could use some references.
    • After Korean Air Lines Flight 007 was shot down in 1983 after straying into the USSR's prohibited airspace, - was this before or during the development of GPS? Also, this is repeated in the timeline...
    • Initially the highest quality signal was reserved for military use, while the signal available for civilian use was intentionally degraded ("Selective Availability", SA). Selective Availability was ended in 2000, improving the precision of civilian GPS from about 100m to about 20m. - this seems redundant to a whole section farther down the article...
    • The whole timeline section should be converted to prose. Keep the table, though.
  • GPS concept
    • The trilateriation info is repeated twice in the same section. I would recommend keeping the content in the subsection, and leaving the overview in the first paragraph out. I'd try merging them somehow for style, but the overview just seems too lacking in details, especially when the info is fully developed less than a page away. (That, or just merge the two paragraphs together. The discontinuity in the treatment doesn't seem efficient here.)
    • I'm not sure the clock correction section should go here. That seems as more advanced a topic than necessary for a basic overview of the technology. Maybe it should be moved farther down the article. A sentence explaining that the fourth satellite can be used to correct the GPS unit's internal clock and a pointer to another section should suffice.
  • More to come. Titoxd(?!? - cool stuff) 07:03, 31 July 2009 (UTC)[reply]

Inital comments from JMiall

  • 'can be used freely (unless technically restricted by the U.S. Department of Defense) by anyone,' - this is a clunky bracketed clause in the most important part of the article. 'can be used freely by anyone (unless technically restricted by the U.S. Department of Defense),' would be an improvement or maybe 'can be used freely by anyone unless the system is technically restricted by the U.S. Department of Defense,'
  • 'anywhere'- it can't actually be used anywhere. The lead should probably be more specific about where GPS/GNSSs can be used ie on the earth's surface or in the atmosphere with line of sight to several satellites
  • the lead should probably also mention the approx location and time accuracy available as these are important aspects of why GPS use is widespread
  • the link for 'phase comparison of signal transmission from pairs of stations' could be a ref rather than direct link. As could others.
  • 'and hobbies such as geocaching' - there is no mention made of this in the main part of the article
  • 'Friedwardt Winterberg [1] proposed a test of General Relativity using accurate atomic clocks placed in orbit in artificial satellites' - when? can this sentence be combined with next. get rid of 1 of the 2 GR links in these.
  • There should be a (non-breaking) space before all units
  • link White Sands Missile Range
  • 'In 1998, GPS technology was inducted into the Space Foundation Space Technology Hall of Fame' - is this worth mentioning?
  • link 1st use of Galileo
  • What was the 1st civilian product using GPS? This should be in the timeline.
  • 'Using messages received from a minimum of four visible satellites, a GPS receiver is able to determine the satellite positions and time sent' - presumably 'times sent'. If each satellite broadcasts its own position then it only needs 1 visible satellite to determine that satellite's position.
  • Are the procedures in 'Position calculation introduction' and 'Correcting a GPS receiver's clock' actually close to what is generally done (I haven't got far enough in the article yet)? Or is it best-fit a solve for all 4 parameters simultaneously?
  • 'but this was modified to six planes with four satellites each' - why?
  • There is very little information about what is actually on each satellite including in the article GPS satellite. For example do they have onboard power/thrust supplies that will run out? What method(s) do they use to correct alignment?
  • 'GPS functionality has now started to move into mobile phones en masse' - when is 'now'? JMiall 22:35, 5 August 2009 (UTC)[reply]

RHB100 (talk) 01:35, 8 August 2009 (UTC) I thank all of the reviewers for the reviews they have submitted thus far. Your comments have been very useful and I hope you are able to work into your busy schedule the time to submit more. I am trying to update the article in response to your comments and will continue to do so. Both of you have commented on Winterberg. One has asked for a more precise reference and the other has requested getting rid of one of the links. I will try to respond to your comments and they are certainly appreciated. However, I do not have very good information on Winterberg and I don't know if more research on Winterberg is justified for a GPS article. RHB100 (talk) 21:47, 6 August 2009 (UTC)[reply]

Sorry I meant get rid of 1 of the 2 GR links, not both. I've changed it above. JMiall 23:03, 6 August 2009 (UTC)[reply]

Many changes have been made in response to your comments. Some but not all are mentioned in the following. A link was added to help explain how the location of Sputnik was determined. Typical position accuracy put in 1sr section. I finally realized that GR was an abbreviation for general relativity and I removed the 2nd link to general relativity. Links to White Sands and Galileo system added. The procedures in 'Position calculation introduction' and 'Correcting a GPS receiver's clock' are simplified expolanations of what is actually done. Changed time sent to times sent in Position calculation, introduction section. I added a diagram to the section on GPS receiver clock in order to make it more understandable and more appropriate for an introductory section. The Timeline section has been improved and the Modernization has been merged into the Timeline section. More links have been added at the beginning of the Space segment section in order to make more information on the satellites available. I have attempted to make sure that there is adequate space between numbers and units whether achieved by the math editor, table editor, or non-breaking spaces. Please let me know of any specific cases I have missed. Natural and artificial sources of interference have been put under Error sources and analysis.
RHB100 (talk) 01:58, 8 August 2009 (UTC)[reply]

With regard to the table at the beginning of Error sources and analysis, we acknowledge that it is not made clear whether the numbers are standard deviatios or worst case errors. However, we are limited by what our source reference at the end of the 1st paragraph provides. In the analysis, we have taken the conservative approach by considering them to be standard deviations.
RHB100 (talk) 02:16, 8 August 2009 (UTC)[reply]


Comments from Mike1024:

  • There are details on how position solutions are calculated in section 2, then section 3.4.3, then section 9, and the calculations are pretty involved. I would consider combining the three sections into a separate article, and maybe simplifying section 2 a bit and giving a link to the more detailed article.
  • The article includes three pictures showing two spheres intersecting to form a circle - then goes off talking about using 4 satellites to calculate 4 unknowns. I realise it would be difficult to make a clear 3D image with 4 satellites in it, but I would consider using a 2D image to show that 3 satellites are necessary for the 2D case, and hence 4 satellites are required for the 3D case. I'm thinking something like figure 1 on <a href="http://lea.hamradio.si/~s53mv/navsats/theory.html">this page</a>?
  • In the 'Error sources and analysis' section there are a list of error sources such as "Signal Arrival C/A ± 3 m" - it's not clear whether that's a range (i.e. the error never exceeds 3m) or a standard deviation (i.e. the error exceeds 3m about 30% of the time), especially when it gets to calculations using the symbol sigma, which is the usual symbol for standard deviation.
  • There's a Section 3.6, "Error sources and analysis", and a Section 4, "Possible sources of interference" - do these belong in the same section? I suppose arguably not though.
  • The section "Geometric dilution of precision computation (DOP)" dives straight into how DOP is calculated, but does not explain what DOP is, or what it represents. I would consider a diagram as it's hard to explain in words. If you go for the separate article on calculations that I suggest above, you could also put the derivation of DOP equations section in there. Also, the section on DOP does all its calculations with values named x, y and z which implies (though it isn't specified) ECEF positioning - but in order to be able to calculate HDOP and VDOP isn't it typical to perform the calculations in the local tangential plane (ENU)?
  • I would consider cleaning up the civilian applications section; for example the line about GPS tours looks like it was added later, and there is more text about CDMA synchronization applications than about personal navigation. Also, does the stuff about export regulations really belong in the 'civilian applications' section?
  • The 'See also' section could be trimmed a bit, - I'm thinking the links to Exif, skyhook wireless, Bluetooth, things like that. Also, given that we have the page GNSS applications to list things that use GPS, I would have thought that Geodashing, Geocaching, Geotagging, Geofence, GPS drawing and so on could be moved to that page.

Needless to say these are just my observations on the article - it might be that I'm suggesting things that the article's more regular editors have a good reason not to do, in which case by all means discard the suggestions you don't like! Mike1024 (t/c) 09:54, 7 August 2009 (UTC)[reply]

I agree with you that a diagram would be helpful in explaining dilution of precision. DOP can be explained in words but a diagram would be helpful. A diagram may be added at some time in the future. Please do not assume that the use of x, y, and z imply that they are expression of the components of a vector in an ECEF coordinate system or any other coordinate system. I have always regarded x, y, and z as strictly generic unless otherwise stated and that x, y, and z can be used to express the components of a vector in any Cartesian coordinate system whatsoever including but not limited to a North, East, Down coordinate system a South, East, Up coordinate system, or a geocentric equatorial coordinate system. In making a numerical computation, the unit vector components in the A matrix should be in the coordinate system of interest. This will usually be a North, East, Down or South, East, Up coordinate system to correspond to our usual interpretation of the meaning of horizontal and vertical. RHB100 (talk) 21:44, 8 August 2009 (UTC)[reply]

  • The table "Summary of Satellites Launched" is a bit confusing - for example the column "Satellites launched" containing both launched and unlaunched satellites, plus signs, and three footnotes. I would consider splitting that into two columns, 'launched' and 'planned', with one footnote to denote launch failures. I would also consider changing the 'Currently in service' column from wording "12 of the 12 launched" to simply "12", as the number launched will be in a previous column. Mike1024 (t/c) 15:15, 9 August 2009 (UTC)[reply]

I agree with you. I think the table would be better if there were more columns and less use of superscripts. RHB100 (talk) 22:54, 9 August 2009 (UTC)[reply]