Re: [time-nuts] WWVB phase plots
Okay. A little 3586B hacking was required, but here are some wide-band results: http://www.jks.com/wwvb/wwvb.html#wideband Thanks very much. This data shows the full-bandwidth WWVB signal very well. Attached are some plots and an octave script. The first plot shows the demodulated WWVB waveform over one second, averaged across the full 300-second recording, so it's the sum of 300 successive one-second periods. The sharp drop in power at about 45 milliseconds is the main on-the-second marker. Also visible is the mixture of carrier-power increases at 200 ms, 500 ms, and 800 ms after the on-the-second marker. The second plot is a closeup of the on-second marker. The falling edge is quite fast, with a time constant of about 350 microseconds, corresponding to a 3 dB one-sided bandwidth of about 450 Hz. I would guess that this edge might be estimated to within 5% of the time constant, or 20 microseconds (about one carrier cycle), which would be well below other sources of systematic error from propagation. The SNR is just huge, and this is for only five minutes of averaging---an hour, or a day, would be even better. Granted, though, these are good reception conditions. I should pick up one of those wideband USB audio sticks and try it from here in California. I wonder whether the WWVB receiver chips could save power by sampling only near these fast edges (narrow correlator in GPS-speak), going to sleep for the remaining 99% of the time. Unless the local clock is disastrously bad, one would think the device would only need to read the full time code once per month, say, and in between just do occasional trims using the WWVB edges. They seem to be having some difficulty holding the carrier power steady during the low-power intervals. Is that 10-Hz tremolo at the start of the second a power-supply thing? some limitation of the PA? There's some undershoot and overshoot too. I've found a few documents describing the WWVB antenna bandwidth: Page 136 of NIST Special Publication 250-67, showing a scope photo of the waveform: http://tf.nist.gov/general/pdf/1969.pdf Page 5 of this technical report: http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA299080 and another scope photo on page 2 of this magazine article: http://tf.nist.gov/general/pdf/2429.pdf Cheers, Peter attachment: wwvb-averaged-second.pngattachment: wwvb-transient.png# # plot WWVB's transient response # # # convert from .au to 16-bit .wav: # sox ant.wide.5min.15.625.kHz.44.1k.24b.au -b 16 ant.wide.5min.15.625.kHz.44.1k.24b.wav # [y,fs,bps] = wavread(ant.wide.5min.15.625.kHz.44.1k.24b.wav); n = length(y); t = 0:(n-1); freq0 = (15625-0.455)/fs; c = exp(2*pi*i*t*freq0)'; d = y .* c; d = filter([1 1 1],[1],d); # residual carrier phase, one per second (in lieu of a proper PLL) pp = zeros(1,300); for r=0:299, w = d(r*fs+1:(r+1)*fs+1); pp(r+1) = sum(w); endfor ppc = unwrap(arg(pp)); # coherently sum over 300 seconds p = 8*5512.662; a = zeros(1,fs)'; for k=0:299, s = round(8*5100+p*k); a = a + real(d(s+1:s+fs)*exp(-i*ppc(k+1))); endfor # normalize ampl = 15.9; a = a / ampl; # plot plot((1:fs)/fs,a); grid on; pause; r=1900;s=2200;plot((0:(s-r))/fs,a(r:s)); grid on; ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
In message 20120320031431.bf564800...@ip-64-139-1-69.sjc.megapath.net, Hal Mu rray writes: Could somebody please say a bit more about that area. My Shannon level theory is weak. Why does more transitions help anything? Or what does it help? The transitions (where the phase change!) are what you correlate, the more, the better S/N you get. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
The transitions (where the phase change!) are what you correlate, the more, the better S/N you get. Yes---it's too bad that the proposed WWVB changes don't increase the number of transitions at all. Could they not do the low-modulation-index DCF77-like signal on top of the BPSK? That is, put some small, fast phase wiggles on top of the slow 180-degree transitions (or 120-degree transitions if NIST can be convinced to change to that)? But maybe some Loran-like tricks could be tried with an ordinary WWVB signal and a receiver with a few kHz bandwidth. The first part of the exponential decay of the carrier amplitude (at the on-second marks) might be relatively uncorrupted by sky wave, just as with Loran. Considerable averaging would be needed I guess. If the characteristics of the transmitting antenna are known, a model of the pulse decay could be used to estimate the transmit time. The phase transitions happen during the low-power intervals (-17 dB), so they would seem to be less useful than the amplitude transitions: an 11 dB penalty, counting the gain from the antipodal signaling. What is the inherent bandwidth of the DCF77 system, by the way? John, if you're reading this, would your receiver be capable of recording with wider RF bandwidth? Your recordings made during the test period have a bandwidth of about 30 Hz; can it go any wider? I think your web page says you're using an active whip antenna, which is good because the resonant loops would impose their own bandwidth limit. If you could eliminate the narrowband receiver and record the antenna signal directly with the 192 ksa/s ADC, that would be ideal. (I should really cobble up a system of my own, but I'm a fair distance from Colorado.) Cheers, Peter ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] WWVB phase plots
Yes---it's too bad that the proposed WWVB changes don't increase the number of transitions at all. Could they not do the low-modulation-index DCF77-like signal on top of the BPSK? That is, put some small, fast phase wiggles on top of the slow 180-degree transitions (or 120-degree transitions if NIST can be convinced to change to that)? The problem is that if you superimpose a wider bandwidth modulation over the narrow one you have distributed the available sideband power over the broader bandwidth. You have less power transmitted over the narrow bandwidth than before. This reduces the range of the station for narrow and wide bandwidth users. I changed the batteries in our cheap clock here at home and it took several days for it to finally get a signal strong enough to set the time. The signals are weak for many everyday users since their clocks have relative small antennas and they have lots of local interference from compact florescent bulbs and switching power supplies in computers and TV sets. 73 Bill wa4lav ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
Jeeze the answers simple DCF is metric and wwvb in english. They never work correctly together. ;-) Regards Paul WB8TSL On Tue, Mar 20, 2012 at 6:18 AM, Peter Monta pmo...@gmail.com wrote: The transitions (where the phase change!) are what you correlate, the more, the better S/N you get. Yes---it's too bad that the proposed WWVB changes don't increase the number of transitions at all. Could they not do the low-modulation-index DCF77-like signal on top of the BPSK? That is, put some small, fast phase wiggles on top of the slow 180-degree transitions (or 120-degree transitions if NIST can be convinced to change to that)? But maybe some Loran-like tricks could be tried with an ordinary WWVB signal and a receiver with a few kHz bandwidth. The first part of the exponential decay of the carrier amplitude (at the on-second marks) might be relatively uncorrupted by sky wave, just as with Loran. Considerable averaging would be needed I guess. If the characteristics of the transmitting antenna are known, a model of the pulse decay could be used to estimate the transmit time. The phase transitions happen during the low-power intervals (-17 dB), so they would seem to be less useful than the amplitude transitions: an 11 dB penalty, counting the gain from the antipodal signaling. What is the inherent bandwidth of the DCF77 system, by the way? John, if you're reading this, would your receiver be capable of recording with wider RF bandwidth? Your recordings made during the test period have a bandwidth of about 30 Hz; can it go any wider? I think your web page says you're using an active whip antenna, which is good because the resonant loops would impose their own bandwidth limit. If you could eliminate the narrowband receiver and record the antenna signal directly with the 192 ksa/s ADC, that would be ideal. (I should really cobble up a system of my own, but I'm a fair distance from Colorado.) Cheers, Peter ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
On Mar 20, 2012, at 4:18 AM, Peter Monta wrote: John, if you're reading this, would your receiver be capable of recording with wider RF bandwidth? Your recordings made during the test period have a bandwidth of about 30 Hz; can it go any wider? I think your web page says you're using an active whip antenna, which is good because the resonant loops would impose their own bandwidth limit. If you could eliminate the narrowband receiver and record the antenna signal directly with the 192 ksa/s ADC, that would be ideal. So the goal here is to measure the bandwidth of their antenna system? (or what they limit the transmitted bandwidth to be to make sure no power is wasted by the bandwidth limitations of the antenna). The first IF of the 3586B is 50 MHz and is filtered to 10 kHz BW before mixing to 15.625 kHz. So recording the second IF is one possibility. The other as you say is to connect the active antenna directly to the sound card, crank up the input gain and hope for the best. No reason I couldn't playback the recording and run it through the SA and 3586 again to see what it looks like. Let me work on that. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
On 18 Mar, 2012, at 10:52 , John Seamons wrote: They do talk about using the 11-bit Barker code for autocorrelation. But the sync bits transmitted only match the Barker code if you interpret them a little bit out-of-order. The part of the paper that talked about the Barker code confused me somewhat since I couldn't quite figure out how it was relevant. The autocorrelation property of the Barker code is only interesting if the Barker code is the only thing being sent (over and over), but in this case the concerns are more about spurious correlations with the variable data, something for which no solution seems to be possible. It is the case, however, that (non-circular) autocorrelations of the fixed sequence are relevant at small offsets. In your data the fixed sequence seems to be -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, -1 which, ignoring the contribution of the variable data (which increases with increasing offset), gives this basic result for offsets from 0 to 13 seconds: 14, 1, 2, 1, -6, 1, -2, -3, 0, -3, 0, 1, 0, 1 So there is a quite large autocorrelation at 4 seconds offset. If I weight the search pattern by the fixed pulse widths (there are 3 0.2 second pulses and 3 0.8 second pulses in the fixed sequence; I gave the rest a weight of 0.5) that gets a little better, i.e. 7.0, 1.4, 0.4, 0.5, -2.1, -0.4, -0.4, -0.9, -0.3, -0.9, 0.0, 0.5, 0.0, 0.2 if I did that correctly, though at the apparent cost of making the autocorrelation at a 1 second offset a bit worse. In any case, if this is the pattern they selected I really would have liked to have seen a discussion of the tradeoffs involved in picking it, along with the assumptions they made about how it would be detected. And I kind of hope I don't have to read about that in someone's patent since technical descriptions written by lawyers are really boring. In any case, I think the paper left out the good parts. Dennis Ferguson ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
What makes me wonder: Why don't they adapt the DCF77 implementation? Is it the not invented here syndrome? At it looks like they never heard of Kasami sequences. - Henry Dennis Ferguson schrieb: On 18 Mar, 2012, at 10:52 , John Seamons wrote: They do talk about using the 11-bit Barker code for autocorrelation. But the sync bits transmitted only match the Barker code if you interpret them a little bit out-of-order. The part of the paper that talked about the Barker code confused me somewhat since I couldn't quite figure out how it was relevant. The autocorrelation property of the Barker code is only interesting if the Barker code is the only thing being sent (over and over), but in this case the concerns are more about spurious correlations with the variable data, something for which no solution seems to be possible. -- ehydra.dyndns.info ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
Hi Henry: There are millions of WWVB clocks in use and the new signal must be fully compatible with them. Have Fun, Brooke Clarke http://www.PRC68.com http://www.end2partygovernment.com/Brooke4Congress.html ehydra wrote: What makes me wonder: Why don't they adapt the DCF77 implementation? Is it the not invented here syndrome? At it looks like they never heard of Kasami sequences. - Henry Dennis Ferguson schrieb: On 18 Mar, 2012, at 10:52 , John Seamons wrote: They do talk about using the 11-bit Barker code for autocorrelation. But the sync bits transmitted only match the Barker code if you interpret them a little bit out-of-order. The part of the paper that talked about the Barker code confused me somewhat since I couldn't quite figure out how it was relevant. The autocorrelation property of the Barker code is only interesting if the Barker code is the only thing being sent (over and over), but in this case the concerns are more about spurious correlations with the variable data, something for which no solution seems to be possible. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
Hm. I had a quick look at http://en.wikipedia.org/wiki/WWVB I cannot see why it won't work with the DCF77 scheme. The carrier is always on-air. Do I miss something? To low bandwidth of the transmitting antenna? Sorry, I didn't followed the thread in whole. - Henry Brooke Clarke schrieb: Hi Henry: There are millions of WWVB clocks in use and the new signal must be fully compatible with them. -- ehydra.dyndns.info ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
DCF77's AM modulation is a much better fit for what they did, and a much better design in general. All the useful phase modulation needs to be carried by the carrier at full power. DCF77's AM modulation drops the carrier power for only 100 ms or 200 ms at the beginning of the second, which gives them a full 0.8 seconds in every second at full power (if I'm remembering right the minute marker has no carrier reduction, so the very longest carrier reduction is only 0.2 seconds). Their chip sequence is just under 0.8 seconds long and sits in the full power part of each second. WWVB is not nearly so convenient. The carrier reductions for WWVB are deeper than DCF77, making it even more imperative that the information be carried in the high power segments only, but WWVB's carrier drops are 0.2, 0.5 or 0.8 seconds long, so in many seconds they only have 0.5 seconds of high power and in 7 seconds per minute there is only 0.2 high power seconds. I think there's no good way to make DCF77's silk purse out of the WWVB sow's ear. It is also the case the DCF77's phase modulation probably isn't as good as it could be if the goal is to find it in the noise since it only swings +/- 15 degrees rather than +/- 90. Its big advantage might be that it is high speed, with lots of transitions, so you can probably measure phase alignment pretty accurately with that. As a national time service, however, it only needs to serve a fairly compact country relative to WWVB's intended coverage area, so that plus WWVB's crappy AM format probably pushed them to forget about trying to match DCF77 and to just concentration on doing the best they could to improve coverage. That would be my guess, anyway. Dennis Ferguson On 19 Mar, 2012, at 19:47 , ehydra wrote: Hm. I had a quick look at http://en.wikipedia.org/wiki/WWVB I cannot see why it won't work with the DCF77 scheme. The carrier is always on-air. Do I miss something? To low bandwidth of the transmitting antenna? Sorry, I didn't followed the thread in whole. - Henry Brooke Clarke schrieb: Hi Henry: There are millions of WWVB clocks in use and the new signal must be fully compatible with them. -- ehydra.dyndns.info ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
It is also the case the DCF77's phase modulation probably isn't as good as it could be if the goal is to find it in the noise since it only swings +/- 15 degrees rather than +/- 90. Its big advantage might be that it is high speed, with lots of transitions, so you can probably measure phase alignment pretty accurately with that. As a national time service, however, it only needs to serve a fairly compact country relative to WWVB's intended coverage area, so that plus WWVB's crappy AM format probably pushed them to forget about trying to match DCF77 and to just concentration on doing the best they could to improve coverage. Could somebody please say a bit more about that area. My Shannon level theory is weak. Why does more transitions help anything? Or what does it help? I can see how it might make it easier/faster to get synced up, but if the goal is to accurately measure frequency or phase, I'd expect that you are already locked and looking for the next layer of detail. In that case, I'd expect better results with fewer transitions. Fewer transitions means lower bandwidth so you can use a narrower bandwidth filter and get rid of more noise. -- These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB phase plots
On Mar 16, 2012, at 1:32 AM, Peter Monta wrote: Attached are some more renderings of John Seamons' WWVB data. This is what one might expect from a receiver that knows when the phase reversals happen and takes them out noiselessly---re-reversing the out-of-phase bursts to recover an approximation of the usual WWVB signal. Thanks for the additional analysis Peter. Odd that there is significant phase jitter. I've added to my website the only other significant recording I made: ten minutes in the dead of night (2:27 AM MST, 9:27 UT). http://www.jks.com/wwvb/wwvb.html#10-min The phase data I extracted is the same as the two minutes of data I captured earlier. A constant pattern that repeats every minute. Certainly not the full protocol as described in the NIST paper. So maybe this test was to simply evaluate the phase modulation effects on receiving equipment (in which case it's shame we didn't find out earlier so we could do more boat-anchor compatibility testing). In an earlier message Dennis Ferguson points out that the paper doesn't fully specify the 11/14-bit minute-sync and 60-bit hour-sync codes. So it's not clear what they were actually transmitting. They do talk about using the 11-bit Barker code for autocorrelation. But the sync bits transmitted only match the Barker code if you interpret them a little bit out-of-order. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] WWVB phase plots
Attached are some more renderings of John Seamons' WWVB data. This is
what one might expect from a receiver that knows when the phase
reversals happen and takes them out noiselessly---re-reversing the
out-of-phase bursts to recover an approximation of the usual WWVB
signal.
The first plot shows the entire interval of the recording, about 210
seconds. The Y axis is phase in radians. Whenever the phase was
greater than 90 degrees or less than -90 degrees, that sample was
assumed to be part of a phase-reversed segment, and 180 degrees was
added to bring it into alignment.
The spikes are the phase reversals. The phase-modulation test ends at
around 138 seconds, and there is an interval of no signal, during
which the phase is random. Finally the old WWVB picks up at around
158 seconds.
The interesting thing here is the phase jitter during the test, which
seems significantly worse than the normal WWVB signal. The short
spikes can be ignored---it's the performance during the bulk of any
given one-second interval that's important. The phase is jumping
around on a second-by-second basis by 0.1 radians or so (about 0.3
microseconds). The normal WWVB signal, by contrast, is nice and
smooth.
The second plot shows a closeup. Around the 85-second mark, for
example, the phase is clearly jumping around.
Also attached is the Octave code that produced the plots (Octave is a
free-software system similar to Matlab).
Cheers,
Peter
attachment: wwvb-phase1.pngattachment: wwvb-phase2.pngy = wavread('wwvb.post.process.8k.16b.wav'); # read in the WAV file
x = y(:,1); # extract the first channel (containing the raw samples)
n = length(x); # total number of samples
t = 0:(n-1)';# time vector
fs = 8000; # sample rate
phase0 = 0.1;# carrier parameters
freq0 = 0.23121151;
drift0 = 2.0e-13;
p = phase0 + t.*(freq0+t*drift0);# vector of carrier phases
c = exp(2*pi*i*p)'; # synthesize the carrier
z = c.*x;# multiply by carrier to downconvert to complex baseband
zf = filter(ones(1,15),[1],z); # simple boxcar lowpass filter
zf = zf/1.67;# rescale so the peak is near 1.0
#plot(real(zf(277000:365000))); # plot a segment of the waveform
#grid on;
#plot(zf(1:1e6));# plot a constellation diagram of the PM part of the test
q = arg(zf); # extract the phase
q(qpi/2) -= pi; # remove 180-degree phase reversals
q(q-pi/2) += pi;
plot(t/fs,q);# plot the result
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
