I am trying to build a Network RTK system.
My specific doubt is - how to coordinate the raw data from all base stations at the central processing station and generate correction data?
I have Skytraq ns-hp modules and I was able to establish traditional RTK fix using 2 ns-hp modules (one configured as base and one as rover) using the inbuilt RTK functionality of ns-hp.
Now, I want to use a network of base RTK stations to get a RTK fix of my rover. I plan to use -
* 4 ns-hp modules as base stations, such that at least one base station is always within ~30 km of the rover
* 1 ns-hp module as RTK rover
* RPI zero W as a central processing station for the network
* RTKLIB on RPI for computation
* wireless communication (haven't decided how but probably 3G/4G)
The rover will send its ns-hp raw data to the central processing station, which will in turn coordinate with the base stations to generate correction data and send it back to the rover.
I wanted to know -
After collecting raw GNSS data from all the base stations and the rover on the central processing station how to compute the correction data?
I believe after I have the correction data, I can send it back to the rover and it can get a RTK fix using this correction data and its own GNSS raw data.
Now, I researched on the topic, and am familiar with a few methods of network RTK (eg - VRS, i-MAX, MAC), but that is not what I want to know.
Everywhere I searched, it was told to use the correction data provided by the local NTRIP provider, but as a RTK network isn't yet available in my country, and I, anyway, am using my own base stations; this information isn't relevant to me.