Articles

Untangling Networks - Part 2

Category: Articles

Written by Shawn "Shag" Greer

Created on Friday, 02 April 2010 22:34

In my last article I discussed the different types of networks that are available. I would now like to focus on the most popular type available. The IP internet based network.  These networks started popping up in the late 90's. Early networks were generally private and developed with subscriptions in mind. Now there are many "Free" or public networks available as well as many subscription based networks.

Taking a minute to understand how networks work can help you and your company make a wise decision when it comes to your next machine control purchase. Most that are using network based correction services will tell you that they don't know how they did it before they had this solution. Networks provide many benefits to contractors that include:

No Base Station Required

Radio interference is eliminated

Ability to run multiple jobs and pieces of equipment

Many are discovering the hidden benefits of networks that include the wireless data access that is required when working with internet based networks. With the Internet in the machine or on the pole this creates a new way to transmit and receive data between the office and the job. Access to email and FTP makes communicating changes and receiving as built data a snap.

Now that we know all of the productivity gains that we can get by using a RTN for correction let's spend a few minutes talking about how they work. There are a lot of buzz words being tossed about by network manufacture including: VRS, Modeled Solutions, Single Base Line, Clusters, TCPIP and NTRIP.

Taking a look at all of the technical information from a contractor’s stance can be a rather daunting task. There are really only a couple of key factors to look out for when choosing what type of RTN to use. First and foremost make sure that the base stations the RTN provider has in service are tracking the same set of signals that you purchased with your equipment. This means that if you bought a Glonass capable receiver and get on an RTN that does not have reference stations that support Glonass than you will not get any benefit from that feature on the machine/rover end. The same can be said for L2C signals or any future signals on your receivers. Most of the larger network providers do support these features but there are a few out there that are working with older technology.

When you break a network down to the raw infrastructure the only thing they do is act as a base station for you. This means that you are receiving the same types of corrections that you would if you setup a radio based receiver on your site. Keeping this in mind there are a few small differences. In most cases the big problem created by a network based solutions is the distance from the closest base to the working end of the receiver. In some cases these distances can get upwards of 25mi/40km. The problem that is created by these long distances is found in some assumptions that are in RTK calculations. One of the variables solved for in these calculations has to do with atmospheric conditions that interrupt or distort ranging signals from satellites. This phenomenon is known as Ionospheric and Tropospheric disturbance.  In most cases this error is assumed to be the same or relative at the base and the rover.

While making this assumption is less problematic at 12mi/20km than at 25mi/40km. This is where some of the buzz words start to be used. There are many philosophies on how to make these errors negligible at large distances. Having not tested all of them I cannot speak intelligently about what type of results to expect. But one thing I can tell you is that as you develop a relationship with your RTN provider they should be willing to work with you to test which service they provide will be the best fit for your needs.

Networks work under a very simple process that involves knowing the location of the user.  Knowing the location that you are expecting to use the reference stations allows the network to choose the most suitable station for you.  There are many variations on what types of services are available after you connect.  These services can mean the difference in the performance you get from your receivers.  Some of the more common services are Single Base Line RTK, some type of modeled solution, or a simple DGPS correction.

Single base line RTK is in the simplest form of RTK and exactly what we were using when we had base stations onsite.  The only thing that is different about this service when it comes from a network provider is the method of delivery.  Internet instead of radio.  Single base line simply means that you are working with one base station.  This service will satisfy most of your requirements on the job depending on the location of the nearest station relative to your work area.

Modeled solutions were developed as a way to minimize the number of stations that are required in a given geographic area.  Minimizing stations means increasing the distances between stations and potential job sites.  Increasing these distances raises the error probability in one or both dimensions (Horizontal/Vertical) of the position solution.  The theory is a relatively simple concept.  As I discussed earlier the assumption of Iono and Tropo errors in the RTK solution is assumed at the rover and base.  Having multiple base stations allows this error to be measured at all of the stations in the network.  Knowing this a mathematical model of the correction can be created.  Basically creating a contour map of the error you can map the location of the rover or machine and make a best guess of the error at that location.  Now having said that mouthful it is at this point that different manufactures do different things.  Some create virtual base stations some include different messages in the standard RTK messages.  Bottom line here is that if you intend to use a modeled solution make sure your gear can support the messages and philosophy coming from your RTN provider.