Changeset 11130 in ntrip for trunk/BNC/src/bnchelp.html

Timestamp:

Oct 28, 2026, 9:32:30 AM (7 days ago)

Author:

stuerze

Message:

Some fixes from security point of view

File:

• trunk/BNC/src/bnchelp.html (modified) (4 diffs)

trunk/BNC/src/bnchelp.html

@@ -147 +147 @@
     &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2.8.2 <a href="#corrint">Interval</a><br>
     &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2.8.3 <a href="#corrport">Port</a><br>
-    &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2.8.4 <a href="#ssrqc">QC Directory, QC Streams, QC Reference, QC Interval, QC Bias File</a><br>
+    &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2.8.4 <a href="#ssrqc">QC Directory, QC Streams, QC Reference, QC Interval, QC Bias File, QC ANTEX File</a><br>
     &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2.8.4 <a href="#corrwait">Wait for Full Corr Epoch</a><br>
     &nbsp; &nbsp; &nbsp; 2.9 <a href="#syncout"><b>Feed Engine</b></a><br>
@@ -5097 +5097 @@
 
   <p>
-  <h4 id="ssrqc">2.8.4 QC Directory, QC Streams, QC Reference, QC Interval, QC Bias File - optional</h4>
+  <h4 id="ssrqc">2.8.4 QC Directory, QC Streams, QC Reference, QC Interval, QC Bias File, QC ANTEX File - optional</h4>
   </p>
   <p>
@@ -5106 +5106 @@
     e.g. 'ALL,!SIRGAS200001' for a stream referring to another reference frame than the ITRF. 'QC Reference' selects the reference of the stream comparison (see below). Select
     the 'QC Interval' for the reports (default: 1 hour). A 'QC Bias File' (Bias SINEX, OSB or DSB, e.g. a CAS DCB or
-    a GFZ OSB product) is an optional reference for the code bias check (see below). One daily file
+    a GFZ OSB product) is an optional reference for the code bias check (see below). A 'QC ANTEX File' (e.g. the
+    current igs20.atx) enables the attitude check (see below). One daily file
     '&lt;stream&gt;_S_&lt;YYYYDDD&gt;0000_01D_QC.txt' is written per stream; at the end of each interval and at the
     end of the program, a report per system is appended. Broadcast ephemerides must be received as well, e.g. via
@@ -5241 +5242 @@
     clock 0.689 (STD 0.008), radial 0.020, along 0.036, cross 0.022, SIS (radial - clock) 0.687 (STD 0.008)
   G satellites deviating from the reference, RMS SIS [m]: G02(0.38) G16(0.33) G20(0.91) G26(3.36)
+</pre>
+  <p>
+    <b>Attitude check:</b> With a 'QC ANTEX File', BNC checks for every stream sending the satellite yaw angle whether
+    its clocks are consistent with its yaw angles during noon and midnight turns. In every stream comparison (30
+    seconds), the difference of the yaw angle of BNC's attitude model ('Computed', see section 'Combination') minus the
+    yaw angle of the stream is followed continuously over time; the phase wind-up predicts a clock difference of the
+    stream to an AC following the model of lambda_IF * dpsi / 2pi (lambda_IF: ionosphere-free wavelength of the
+    reference signals, about 10.7 cm for GPS). A turn starts when the yaw difference changes by more than 20 degrees
+    against the samples 5 to 15 minutes before, and ends when it is stable again within 2 degrees for 10 minutes. For
+    each turn, the observed clock change of the stream relative to the consensus is compared with the predicted one:
+    the changes at the end of the turn, the regression slope of the observed on the predicted changes, and the clock
+    RMS during the turn without and with the predicted change. A slope between 0.7 and 1.3 with the RMS reduced by
+    at least half means that the yaw angles in the stream describe the attitude the AC used for its clocks - the
+    condition for selecting the attitude 'SSR' for this AC in the combination. Turns with a predicted change below 3
+    cm are not assessed; a clock change of at least 3 cm is then reported as not explained by the yaw difference,
+    which shows that the other ACs model this turn differently from BNC's model (the check assumes that the consensus
+    follows it), e.g.
+  </p>
+  <pre>
+  G attitude check G14 09:42:30-10:25:30: yaw difference to the model 362 deg, clock change vs. consensus 8.2 cm, predicted 10.7 cm (max 10.7 cm), slope 0.77, clock RMS 6.1 -> 2.0 cm: SSR yaw consistent with the clocks
+  G attitude check G25 07:53:00-08:51:30: yaw difference to the model 39 deg, clock change vs. consensus -9.8 cm, predicted -0.0 cm (max 1.2 cm), slope 1.84, clock RMS 5.9 -> 5.8 cm: clock change not explained by the yaw difference (the other ACs model this turn differently?)
 </pre>
   <p>