Changeset 11100 in ntrip


Ignore:
Timestamp:
Oct 19, 2026, 10:31:32 AM (4 hours ago)
Author:
stuerze
Message:

minor changeshelp contents updated

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/BNC/src/bnchelp.html

    r11098 r11100  
    82148214    is counted for the minimum of 2 ACs per satellite. A satellite observed by excluded ACs only is combined with
    82158215    the zero-sum condition of all its ACs as before, and if no consistent AC provides corrections for a system, the
    8216     exclusion is not applied. The checks continue: as soon as a check finds no inconsistency any more (or B_IF no
    8217     longer varies noticeably between the satellites, e.g. after the provider has aligned its biases), the AC is
    8218     used for the clock datum again, which is reported in both logfiles. A smaller weight factor of the AC would not
     8216    exclusion is not applied. The checks continue: once 12 checks in a row (1 hour) find no inconsistency any more
     8217    (or B_IF no longer varies noticeably between the satellites, e.g. after the provider has aligned its biases), the
     8218    AC is used for the clock datum again, which is reported in both logfiles. A single passing check is not
     8219    sufficient, because the offsets are reset with the exclusion and not yet converged; a borderline AC would
     8220    otherwise change the clock datum every few minutes. A smaller weight factor of the AC would not
    82198221    help instead: the weights act on the epoch-wise clock variations, while the satellite-specific constant part is
    82208222    fixed by the zero-sum conditions, independent of the weights.
    … …  
    85128514    combined without it at the latest when the newest incoming corrections are 10 seconds (or the sampling interval,
    85138515    if larger) younger than the epoch. Corrections arriving after an epoch was combined are ignored.</p>
     8516  <p>
     8517    The satellite-specific offsets of the ACs in the Kalman filter are modelled as a random walk of 1 cm per 10
     8518    seconds, independent of the sampling interval, so that 5 and 10 seconds give comparable offsets. The combined
     8519    satellite clocks themselves are estimated epoch by epoch.</p>
    85148520
    85158521  <p>
Note: See TracChangeset for help on using the changeset viewer.