Don't normally work in areas with cell phone coverage but last couple days I have and have been using Cell phone RTK for work...for first time.. pretty cool..just jump out of truck with a rover and go to work while checking Facebook status at same time.
But the last part of the day I discovered a false fixed position on two control points I RTK'd after I set up a total station on one of them and laser shot a distance. Control pair didn't jive by 3 feet horiz and vert.
Anybody else use this type setup... do you see anomalies like this?...false fixes?
shooting elevations with a prismless gun from a two pair rtk control to get heights of a bunch of telecom towers and antennas for an FAA cert.
The accuracy standards are quite low...20ft horz by 3 ft vertical.
I use a Trimble iBase (RTK over cellphone) system daily.
It is pretty much glitch free. In the past three years I can recall only two examples of a "bad" point.
The accuracy does degrade with distance from the base - we try to keep around 6 - 10km. Further than that we go for standard RTK
For control points:
We ALWAYS hit at least three existing control marks. This should show up a bad init.
To reduce plumbing errors we measure a second time with a 180 degree rotation of the rod.
For cadastral or other high accuracy work we measure them again after a re-initialisation and at least thirty minutes.
I don't know how using a cell phone for the data pipe would cause errors. the actual over the air protocol is the same (CMR, CMRS+, RTCM, etc), and I would assume that cell data is error corrected by some type of protocol. The issue mentioned by jim.cox due to distance could happen to any system...if you could get a radio link that far.
I've had this issue maybe 3 times ever. The best thing you can do is to observe your control more than once and average it. If you have time you should wait a few hours in between observations so you have a completely different constellation and solution. If you get the same result you pretty much know it's good at that point. Failing that, you can intentionally lose initialization and gain it back then re-observe. (turn the rod upside down and then right side up again)
I have had that happen only a few times in the past, coincidence maybe, one of times I was near a tall object such as your tower.
I posted a thread under the heading of RTK Experiment last summer in which I documented a bad fix on a point (0.6 feet). That was done using network RTN and cell phone modem. I'm not sure if that had anything to do with it or not.
3' in both the Hz and V almost sounds to me like a datum shift - you have to make sure that the corrections you're receiving are in NAD83, not IGS.
I agree. Never had one of these, at least with my Leica.
Been in some sketchy places too. (Aren't they all?)
You might of got a bad fix, BUT I highly doubt it is due to using a cell phone for the RTK datastream vs a radio. Agree with Mr. Hamilton.
SHG
This has happened to me while using both the internal modem and a cell phone. Has happened 5-6 times for me. I can say that on all occurrence I've been out in the Texas hill country, probably right near the network limit. I've had errors from .30 to just over 5 feet.
It's a scary feeling when this happens. Makes you think about all other times this could have happened but you didn't catch it. I've learned to just bust out my base station as I've never had this issue with that setup.
