Magnus, that may be fully correct. It does look like that it is not ok to look for levels below values of 3 AMU.
At least Trimble does mention this: 'The approximation is accurate from 3 to 25 AMUs.' I found this statement in the TSIP Reference docu (P/N 34462-00 rev. C, April 1999) The complete explanation for AMU / dBc/ Hz is: 'Signal Level The Signal Level parameter (Byte 5-8) (an approximation of C/N0) is stated in antenna amplitude measurement units (AMUs), a Trimble devised unit. An approximate correlation of AMU levels to C/N0 follows: 5 AMUs @ 41 dB Hz 16 AMUs @ 51 dB Hz 26 AMUs @ 55 dB Hz Formula: C/N0(dBHz) @ 20 log(AMU) + 27 Four basic parameters affect C/N0: " Signal strength from the GPS satellite " Receiver/antenna gain " Pre-amplifier noise figure " Receiver noise bandwidth The approximation is accurate from 3 to 25 AMUs.' It took me a while to find it in the net. I found that the given offset value of 27 is too high for my Thunderbolt, fiddling a bit I find the value of 23.5 a good candidate with perhaps some tenths of variance. I could check it between AMU 5 to 18... Perhaps there are some 'unlinearities' still in, but it matches also reasonable with the values I reported in a old mail ... Arnold On Sun, 14 Feb 2010 00:50:57 +0100, Magnus Danielson wrote: >Arnold, >Arnold Tibus wrote: >> Magnus & Mark, >> >> for my Thunderbolt system I found a very good correlation >> between AMU and dBc/ Hz ( observed on my system using >> the ThunderboltMonitor watching over long time stable satellite signals ) >> when I use the Formula: C/N0(dBHz) = 20 log(AMU) + 23.5. >It would work well, but for lower AMU values (in particular 0) it would >not work so well. Something like the above was my initial assumption. >The availability of AMU=0.0 for a 20 dBc/Hz signal suggests a little >more happens. >Cheers, >Magnus >_______________________________________________ >time-nuts mailing list -- [email protected] >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
