PPP info

Irish Student Looking for Information

It's not my thing, but, some may see the title of this post and be more likely to jump in to respond to your post.

Welcome to this great source of info. We have a lot of fun while sharing a great deal of really worthwhile information. Our backgrounds are so incredibly diverse that we can tackle almost any subject that comes up. I hope that you get some really helpful responses and that you will continue to participate here with us.

PPP is a non differential process, it uses only the satellite data along with precise clocks and orbits to compute a position. Generally long data sets are required.

The SBAS uses corrections and so is more of a differential service much as a RTN network, you are just getting corrections from a space vehicle rather than earth based.

SHG

Read http://gauss.gge.unb.ca/papers.pdf/iongnss2007.leandro.gaps.pdf

Hi Kevin,

SBAS is only a correction delivery method where as PPP is a positioning method. Lets look at it from two perspectives: 1) the network side and 2) the end user.

On the network side:

There are many satellite based augmentation services including:
WAAS
EGNOS
Veripos service
Omnistar
Starfire

In concept they are all generally the same. They all use real-time data collected from a permanent GPS (or GNSS) network to provide an enhanced positioning service by sending corrections for satellite orbits, clocks and in the case of WAAS/EGNOS the ionosphere. In terms of accuracy there is quite a large difference. This is more by design then by anything else.

WAAS/EGNOS etc. were designed for meter or sub-meter level positioning mainly for aircraft guidance using L1 (single frequency) receivers. So the precision of the corrections and update interval of these services are less/slower. Also they must provide ionosphere corrections since it was intended to be used with single frequency receivers. Also these are regional services as the base stations only cover a region of the earth.

Veripos, Omnistar and StarFire were designed for cm - decimeter level positioning and therefore have faster update rates and provide higher precision corrections. The reference networks tend to be global so they offer global coverage. These corrections were intended to be used with dual frequency receivers so no ionosphere model is generally provided.

There are also many internet based augmentation services where by the same corrections that are broadcast via the satellites are sent through the internet. Several companies offer this including Trimble, Starfire, Nexteq, the IGS etc. The advantage to this is that the hardware is less expensive, you only need an internet connection to receive them. But in areas of poor internet coverage this can be a drawback.

On the user side:
PPP is a positioning method. However, it is a vague description. Lets look at each term individually.

Precise - this implies that the accuracy/precision is better than the standard positioning service achieved with GPS (2-5 meters). But there is no definition to how much better. For PPP the precise part is achieved through two means:
1) using a network of GNSS receivers to provide improved orbit and clock corrections (like those sent from WAAS,StarFire etc.)

2) using carrier phase observations in addition to simply pseudorange observations.

Point - this implies that we are not performing differential positioning (as is the case for RTK). Another way to say this is that we are using un-differenced (as opposed to double differenced) observations in our positioning filter.

So basically to do PPP you need three things:
1) improved orbits
2) improved clocks
3) use un-differenced carrier phase observations along with un-differenced pseudorange observations.

Coming back to the various correction services that we described above, you can do PPP with any other them. If I choose WAAS as my orbit and clock provider, my accuracy will not be as good as if I choose one of the commercial services but they all meet the three criteria listed above - precise orbits, precise clocks and un-differenced carrier phase.

For carrying out some tests, the online tools available for PPP are quite good.

In addition to the ones you listed you can also try:

GAPS - http://gaps.gge.unb.ca/
MAGIC GNSS - http://magicgnss.gmv.com/

There was recently a comparison done on GPS world for some PPP software:
http://gpsworld.com/7-free-alternatives-opus-post-processing-in-government-shutdown/

For real time I would suggest using the IGS real-time stream:
http://rts.igs.org/

and a free demo kit that they offer (not for the faint of heart):
http://rts.igs.org/access/

Hope this helps!

Best of luck,

Trah

Idea

Hey Trah,

Thanks very much for taking your time to go through this for me, much appreciated. It is becoming clearer to me now, i will do a bit more reading about it before i come up with a final project idea. In college we just learn the basic GNSS, have only used Trimble Business Center for processing and only carried out Network RTK surveys so i wanted to try something new.

I was thinking of setting up a control point and taking a static measurement on this point for a several hours, then setting up over the same point to carry out the PPP method (i assume that i just gather GNSS data over that point as usual) and post process it using the free online services.

I don't know if this will work.

Cheers again

Idea

"I was thinking of setting up a control point and taking a static measurement on this point for a several hours, then setting up over the same point to carry out the PPP method (i assume that i just gather GNSS data over that point as usual) and post process it using the free online services."

This sounds reasonable. You could save yourself some time by only doing a single setup and running that same data through Trimble Business center and then submit the same file to the online PPP tools.

Or if you wanted to you could download RINEX files from the IGS: (ftp://cddis.gsfc.nasa.gov/pub/gps/data/)

or NOAA (once the US government work stoppage is over).

Most of the online PPP software require RINEX format. Manufacturers have some tools to convert from their format to RINEX.

If you collect data in the field be sure that the antenna type is properly set in the RINEX header, especially if you are interested in height. It can cause ~10 cm errors in some cases.

Good luck!

Idea

Howdy,

If your interest is in the differences resulting from the different approaches to processing GPS data, by all means submit the same data to the different tools. This will insure that the differences examined are the result of the difference in tools and not tracking, SV geometry, etc.

You should refer to the thread on this site dealing with alternatives to OPUS. The GPS World article linked in the thread addresses some of the issues.

I also want to emphasize the importance of understanding the issue of reference frames and reference epochs. Be sure to do your comparisons of positions and heights at the same epoch and reference frame. I would look at Cartesian coordinates myself.

On the matter of heights, Trah correctly emphasizes the importance of using the correct antenna model and height. I have been quite pleased to see the close agreement between the results from the different tools. Each tool uses a different processing engine and there are different approaches to solving for neutral atmosphere effects.

Using a commercial package like the Trimble to process yourself may require some additional preparation. I would be sure to use precise ephemerides and update your reference station's position to that at the mid-point in date and time if your observation session.

There are numerous other threads on this site discussing these issues.

You may find some additional details in the analysis PDF for Lab 6 at:
http://geodesyattamucc.pbworks.com/w/page/13931102/FrontPage#Lab6UsingOPUSRSandcomparingresultswithOPUSdue25March2013

There is also a good discussion of the various processing tools in a 2006 GPS Solutions journal article at:

http://folk.uio.no/treiken/GEO4530/Online_GPS_posisitoning_services.pdf

Good luck,

DMM

Idea

Cheers Guys,

Great information provided.

I think i have my head around it now, i am planning to go out next week and do a trail run to get some data. I will do my static and post process it in TBC with free RINEX data from Irelands CORS network (OSi) and compare it with the online PPP services.

Ireland has 2 fully operational CORS networks (Ordinance Survey Ireland (OSi) CORS network - 23 stations and Trimbles VRSNow network - 22 stations) with 3 different NRTK systems Leicas - SmartNet, Trimbles - VRSNow and Topcons - TopNET. We also have the Commissioners of Irish Lights providing a free DGPS corrections service that has full national coverage. So as a small country we have a very good survey system in place.

Do ye think there would be any need for using PPP in Ireland? I guess it would be only used for setting out control.
I read that several hours of data is needed for PPP post processing for cm accuracies.

In a big country like USA it would provide much more consistency over large areas and be very useful where control networks are unavailable.

Thanks again

Hi Kevin

You could get in touch with the Ordnance Survey of Ireland, they might have some ideas and point you in the direction of doing something that might yield useful results for the surveying community.

It's also worth looking over the (free) published papers on the FIG web pages, lots of comparative studies and so on.

Artie