|WikiProject Technology||(Rated Start-class)|
This article is poorly worded, confusing, and repetitive. It says the same thing several different times in slightly different words. It should probably be cleaned up alittle if anyone could. 18.104.22.168 23:01, 6 April 2007 (UTC)
i'm left a bit unclear after reading this article. i'm left with questions like:
what calculations can the device not make ?
- none. Simply, A-Gps is a normal gps with "remote assistance" regarding satellites
what do devices without agps do ?
- Simply, non-agps devices needs more time to fix.
are they innaccurate ?
- not necessarily.
what sort of innaccuracy ?
- a-gps is used when it is difficult to find satellites; non-agps are quites slowly in fixing.
what most affects the innaccuracy ?
- when you have few satellites in view (e.g. 3) accuracy is reduced in non-agps
who owns the AGPS servers ?
- phone manufacturer, or mobile operator, or whoever has infos about cells/satellites
does the device always connect to them via the cellular network ?
- Normally, yes.
what if the device is a GPS device but it's not cell phone & so not connected to the cellular network ?< br /> - It works like a normal GPS
is agps slower than gps?
- It is faster in fixing, but location is still provided at 1 Hz (1 sample per second)
what's the accuracy differential?
- see differential GPS
does an agps device always use agps or can it use gps ?
Is this service free of charge ? It depends to who offers the service. Tcoombs 16:01, 13 February 2007 (UTC)
Answers added by a GPS user
A little addition / clarification
I'm not up to the task of integrating this, but I'm leaving it for anyone who does feel like --
What you have to understand about A-GPS is the challenge of finding the satellites to begin with. The GPS signal as received at the surface of the earth is very, very weak. Were it not for the processing gain, the signal would be way down in the noise and impossible to detect at all. Besides that, the satellites are whizzing through space and their signals are undergoing doppler shift. A GPS receiver doing a "cold start" doesn't know where it is, where the satellites are flying or which ones are visible, etc. So it needs to employ a large number of processing units (called "correlators") to try to pinpoint the signals from the satellites, when those signals have unknown doppler shift, unknown code sequences, and unknown code offsets.
With A-GPS, the receiver is in communication with a base station that knows
- The correct time
- The satellites' orbital components and code assignments
- The receiver's approximate position
So the base station can now tell the receiver which codes correspond to satellites that are potentially above the horizon, approximately what the doppler shifts of the signals will be, approximately what the correct position is (aids in rejection of bad data), etc. -- meaning that the receiver can get a good lock, in less time, with less processing power. The first consideration is good for availability, and the second is good for battery life. 22.214.171.124 (talk) 09:20, 26 October 2008 (UTC)
- This is not really the case. In practice modern devices made within last few years, such as dedicated Garmin, tomtom etc do not have a hard time at all acquiring and holding satellites and calculating position as accurately as aGPS (of which there are many types, not just the types you have outlined).
- Cold Starts (resets of the device) are not the norm of most dedicated units as they hold almanac and ephmeris in memory even when off. Warm and hot starts are more common. The aGPS advantage is in practicality, a matter of reducing TTFF (Time to First Fix) by seconds, not minutes. Carwon (talk) 14:34, 15 February 2009 (UTC)
When read in combination with the GPS article, I think most, if not all of these points are addressed. I don't think it's quite appropriate to - in a general article at least - go into technical details that most will not understand. Whole books have been written on GPS, with good reason. If you've got a specific point that you feel is missed, could you mention it? --Speedevil (talk) 13:54, 26 October 2008 (UTC)
The article seems to be coming along nicely though the "Drawbacks" section is still a bit sloppy. Some of the context isn't expressed clearly; for example it perhaps implies but definitely doesn't say explicitly that the majority of GPS devices, and almost but not quite all A-GPS ones, are mobile phones. Is it time to make a separate GPS phone article, splitting the technology from its principal commercial application? Or at least a section of the present article? Jim.henderson (talk) 18:06, 29 August 2009 (UTC)
- the great majority of aGPS devices are in fact mobile phones at this point.Carwon (talk) 20:59, 22 September 2010 (UTC)
Fallback to standalone GPS
The article says, "Some A-GPS devices do not have the option of falling back to standalone or autonomous GPS." It would be helpful to list some examples of devices that can and cannot fall back to standalone. (I realize a comprehensive list is not possible, but I would really like to know whether the iPhone 3G or 3GS have this feature.) GPS Pilot (talk) 16:05, 28 November 2009 (UTC)
- If my memory from glancing through a book about iPhone this summer has not failed me, the 3GS can fall back to standalone GPS and even to navigating by WiFi addresses when all GPS fails. Somewhere there should be a site, if not a Wikipedia page (this field is rapidly developing this year), comparing the various smartphone nav capabilities. Jim.henderson (talk) 04:51, 30 November 2009 (UTC)
- Jim, I do not understand why you removed the term standalone for "standard." I have never seen the term "standard" and it is highly ambiguous since it is now "standard" to include aGPS, with hudnreds of millions of of data connected devices eclipsing traditional dedicated GPS devices. The industry term for units that can work without data is "Standalone GPS," this refers to a modality of autonomy from networks. There is an advantage to devices that do either depending on what resources are available.
- - A-GPS, generally acquires faster, and can acquire in more adverse conditions (all things being equal, and they are not equal in smartphones with high attenuation from cramming all the hardware into tiny form factors)
- - "Standalone GPS" works with satellites alone, with no need to communicate with network for initial acquisition
- - Both Standalone and aGPS function -- the best of both worlds
- - Dedicated GPS: a device (as opposed to modes above) whoso primary function is GPS localization and navigation Carwon (talk) 21:20, 22 September 2010 (UTC)
Suggestion for division
I have to agree that the article is confusing.
One thing I can easily identify as being confusing is the division into two groups of aGPS. While writing this, they are called.
- "Using information known to the assistance server but not the GPS receiver"
- "Calculation of position by the server using information from the GPS receiver"
From my understanding, the first one makes little sense. The assistance server will always have some information that the receiver is not able to obtain only via the satellite broadcast signals, right? And aGPS is about using that information, right? So the discription of that group of assistance applies to all aGPS. That description also implies that this information is not to be shared with the GPS receiver, which, I think, is exactly what the author meant.
What about making a division, smth. like this:
- Letting the GPS receiver acquire supplementary data from the assistance server and use this for calculation, to improve time to first fix, yield and accuracy.
- Letting the GPS receiver send available satellite measurements, which might be insufficient for a proper or any localilization in itself, to an assistance server, that has greater computing power and supplemental data to localise the receiver.
It is basically client side vs. server side calculation.
What do you think?
- The client vs server side calculation is the main dividing line in aGPS. This is called MS-based or MS-assisted (with MS meaning the receiver containing device). MS bases means the calculation are done by the MS, and MS-assited means the calculations is done by a provider server using GPS data collected by the MS and relayed to the network and then positing is relayed back to the MS. In practical terms this would be used for sporadic GPS location (not suitable for navigation) on devices with GSP receivers but inadequate processing power.
- Also we do need to be clear that the main purpose and affect of a-GPS is not accuracy refinement. Most inexpensive consumer GPS units and mobile phones utilizing aGPS have adequate RF hardware and calculating power to produce highly accurate positions. The main utility of aGPS remains TTFF (Time To First Fix).Carwon (talk) 21:09, 22 September 2010 (UTC)
GPS devices with no localisation ability in itself?
From hearing other people talk about aGPS a few years back, I was under the impression that some mobile phones and cars had emergency GPS units, that could only take snapshots of satellite measurements, but not make any kind of triangulations from these. Instead they would, in the case of a 911 call or when airbags released, send these GPS snapshots to some 911 server, which would calculate a position.
Are there such units?
- There are many localization methods. Some involve GPS (global position from satellites), many do not. Many countries including the US have a scaled mandate for localization for mobile phones. This means every couple of years the average localization accuracy of handsets available to 911 and law enforcement has an increasing accuracy requirement mandate. This, plus the typcial and changing technology of the handsets means a variety of methods are used. For example carriers are able to discern on non GPS dumphones, or dumphones with GPS, but outside of GPS signal conditions, which tower terrestrial the phone is communicating with. this s the most rough estimate. Signal strength can refine that. Triangulation of towers can refine it further, and timing signals can give even more precise non GPS location.
- But these methods are not part of GPS or aGPS which ultimately require by definition use of orbital satellites.. —Preceding unsigned comment added by Carwon (talk • contribs) 17:33, 12 October 2010 (UTC)
Its said in this article that A-GPS is less accurate than regular GPS, how much is that? Regualr GPS has a +- deviation of 2 meters. How much is the deviation in the A_GPS in lets say an iPhone 3GS? 126.96.36.199 (talk) 16:11, 1 June 2010 (UTC)
- aGPS is no less accurate. A prior editor may have conflated or confused aGPS with land tower triangulation methods, which are not aGPS. aGPS is full use of GPS satellites, with assistance in initially locating them. A decreasing fractional use, MS assisted, while as accurate is not updated as often causing reporting lag and making it unsuitable for navigation, but it can resolve to full accuracy. Carwon (talk) 20:42, 22 September 2010 (UTC)
Mostly, A-GPS means that the phone does not have an integrated GPS receiver and you need an external GPS device to get your location. —Preceding unsigned comment added by 188.8.131.52 (talk) 08:25, 1 November 2010 (UTC)
Is A-GPS assisted GPS or is it any form of locating a device? One part of the article says it makes using the satelites for an accurate fix quicker, another part says the device may not even be able to read satellite data.. which is it? Reading Google's global WiFi AP location database surely can't be GPS, although it is "global positioning" if in range of a known terrestial transmition point. The definition of GPS involves using satelite data, if it doesn't then it can't be called GPS... or has that changed now?
notation: aGPS, AGPS or A-GPS, which is it? all? why are all used in the article?
This article is too ambiguous