Changeset 1082 in ntrip for trunk/BNC/bnchelp.html
- Timestamp:
- Aug 30, 2008, 12:30:48 AM (17 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/BNC/bnchelp.html
r1078 r1082 95 95 3.8.3. <a href=#corrport>Port</a><br> 96 96 3.8.4. <a href=#corrwait>Wait for Full Epoch</a><br> 97 3.8.5. <a href=#corrlate>Latency Log</a><br> 97 98 3.9. <a href=#advnote>Monitor</a><br> 98 99 3.9.1. <a href=#obsrate>Observation Rate</a><br> … … 478 479 When feeding a real-time GNSS engine waiting epoch by epoch for synchronized Ephemeris Corrections, BNC drops (only concering IP port output) whatever is received later than 'Wait for full epoch' seconds. A value of 2 to 5 seconds could be an appropriate choice for that, depending on the latency of the incoming Ephemeris Corrections stream and the delay acceptable by your application. A message such as "COCK1: Correction overaged by 5 sec" shows up in BNC's logfile if 'Wait for full epoch' is exceeded. 479 480 </p> 481 <p><a name="corrlate"><h4>3.8.5 Latency Log - optional</h4></p> 482 <p> 483 Message latencies are logged for streams carrying orbit and clock corrections to Broadcast Ephemeris if option 'Latency log' is ticked. 484 </p> 480 485 481 486 <p><a name="advnote"><h4>3.9. Monitor</h4></p> … … 559 564 UTC time provided by BNC's host 560 565 - GPS time of currently processed epoch 561 + Leap seconds between UTC and GPS time 566 + Leap seconds between UTC and GPS time, hard-coded to 14 562 567 -------------- 563 568 = Latency 564 569 </pre> 565 570 <p> 566 BNC can average the latencies per stream over a certain period of GPS time, the 'Performance log' interval. Mean latencies are calculated from the individual latencies of at most one (first incoming) observation or correction to Broadcast Ephemerisper second. Note that computing correct latencies requires the clock of the host computer to be properly synchronized.571 BNC can average the latencies of observations per stream over a certain period of GPS time, the 'Performance log' interval. Mean latencies are calculated from the individual latencies of at most one (first incoming) observation per second. Note that computing correct latencies requires the clock of the host computer to be properly synchronized. 567 572 </p> 568 573 <p> … … 570 575 </p> 571 576 <p> 572 Latencies of observations or corrections to Broadcast Ephemerisand statistical information can be recorded in the Log file/section at the end of each 'Performance log' interval. A typical output from a 1 hour 'Performance log' interval would be:577 Latencies of observations and statistical information can be recorded in the Log file/section at the end of each 'Performance log' interval. A typical output from a 1 hour 'Performance log' interval would be: 573 578 </p> 574 579 <pre>
Note:
See TracChangeset
for help on using the changeset viewer.