I’m aware that practical L1 baselines for RTK are limited to around 10 km, but i’m intrested if this increases if using a VRS NTRIP service which creates a virtual reference station by interpolating data from a network of reference stations.
I’m a user of Croatia’s national CROPOS positioning service which is based on a network of refernce stations spaced about 50 km from eachother. Map of reference stations can be seen here. The system produces a VRS on the site of the survey so baseline registered by ReachView is always just a few meters.
My first question would be if somebody has an idea what would be a workable distance from one of the reference stations in this setup? I have achieved good results up to 15 km away from CROPOS reference station but i’m wondering about going further than that. I have an assigment in a few days which will be right in the middle of the two stations, about 25 km from each of them. Is this viable enough to achieve RTK fix?
Second question is regarding the best procedure for PPS surveys conducted with the same corrections source. As a CROPOS user i have access to historical RINEX logs for the whole system (paid by the minute of data). But also, if i turn on the NTRIP corrections in the field ReachRS records the VRS corrections while i work, which is actually cheaper and easier way to get to that data. Can i just happily use the Reach base corrections log with emlid RTKlib or is there any significant advantage in accessing the raw RS data from CROPOS records and doing my own interpolation by methods described by @r.pazus and @TB_RTKhere and here?
Thanks in advance for all thoughts on the subject!
Yes, but be aware of degradation of accuracy and harder to obtain fix with increased baseline.
If you have a second RS unit this can be use as redundancy and a refrencepoint for increased accuracy along with VRS ntrip.
If you have VRS i would prefer that, but sometimes cellular coverage and other circumstances call for a better or stable way. Like initialize your base with VRS and work offline with e.g LoRa from here.
So to recap, if you have VRS service like in your setup above, use it. I think L1 shines best with it.
If you for some reason cant get VRS to work then other options are possible but it highly depends on you setting there and then and what resources you have to pull job.
First question:
With L1 GNSS receiver and using the VRS method by CROPOS service you do not need to check the distance to the nearest reference station. Accuracy will be similar throughout the country. Here it is worth emphasizing that L2 is unnecessary for such short baselines. Here, the stability of the antenna phase center is more decisive than the need for these additional frequencies.
Second question:
In the first part the answer is simple. You have the possibility to download RINEX VRS on the site for the ordered period and your location. For PPS (postprocessed static) you only need to register your session in ‘logging’ without the need for NTRIP. For this session you order RINEX VRS. For your 10-minute session, one hour RINEX VRS (1Hz) is enough.
If you try to download data via NTRIP then it will be data only for the period of your session, nothing more. But here you will need a longer session VRS as a base station for better accuracy.
I was hoping that would be the case. Since CROPOS is suprisingly cheap this makes even a single ReachRS a really good buy in Croatia. Do you, from your experience, have estimation of what accuracy one might expect in normal use, on a national level from ReachRS combined with system such as CROPOS? Would 3 cm in RTK be a reasonable expectation nationwide?
Just to clarify on this since i will need RTK in the field anyway: if i leave the unit running for some time before and after my point collection i should be allright with the NTRIP corrections log recorded by Reach?
