How the three variances are derived and how the model correction results are applied.

[152286]

I would like to ask how the three variances — ddres, udifres, and uducres — are computed in RTKLIB,could you please provide the relevant code or point me to the specific part of the rtklib for reference? and in which module this calculation is performed. Also, how are the results of the MHM model and SF model applied to the next observation cycle? Are these corrections directly applied to the observation file, or handled during the processing stage?

[NavyyYuan]

Thank you for your comment. The double-differenced (DD), undifferenced and ionospheric-free (UDIF), undifferenced and uncombined (UDUC) residuals can be computed using the RTKLIB software (https://www.rtklib.com/) or some other precise positioning software. Then, the obtained residuals must be encoded to the file format of ddres, udifres, and uducres used in the MAPS. Or, you can modify the source codes (rtkpos.c) concerning the residual output (outsolstat function) in the RTKLIB software to file format of ddres, udifres, and uducres used in the MAPS. Besides, these corrections are handled for the corresponding observations during the processing stage.

[152286]

Thank you for your reply. I noticed that your example residuals file uses a 1-second sampling rate. However, I calculated my residuals file using a 30-second sampling interval based on the observation file. Will this difference significantly affect the model construction?
Best wishes.