[time-nuts] BPSK decoder for WWVB

Attila Kinali attila at kinali.ch
Thu Jul 4 18:08:18 UTC 2013


On Thu, 04 Jul 2013 07:59:59 -0500
briana <alsopb at nc.rr.com> wrote:

> Please read the WWVB article on this. There is a whole section devoted 
> on how it can  provide higher immunity to MSF.   I don't think it a 
> false advertising claim.
> 
> www.jks.com/*wwvb*.pdf
> 
> Read the Fundamentals of the new protocol section.

I'm not an expert in signal theory, but i call their claims wrong.

(Disclaimer: What i write here is based on what i've learned on signal
theory ages ago, and forgot most of it. There will be inaccuracies and
mistakes in it)


Yes, if you argue with the phase state diagram, then you have a larger
difference between the old AM and the new BPSK modulation. 
But: They argue about decoding the bits of the signal, which is something
totally different than using WWVB as a frequency or time reference.
In the bit decoding case, yes, you get a higher immunity to white noise
or burst noise. I am not convinced that a narrow band jamming signal can be
correctly substracted from the BPSK signal. Fortunately, they leave all the
important details out, how this would be acheived or what the theoretical
limits would be. Using WWVB as a frequency reference, you have to substract
a previously unknown phase modulation from the signal, hence you lose a bit
of noise resilience (it's not much because the bitrate is low). As for timing
reference, you gain nothing at all, because the edge where the second
begins has still the same properties as with the pure AM modulation:
there is only one edge and it is degraded by any noise.

Their claim that the DCF77 645Hz spread signal has not the same jamming
resiliance because it uses only a +/-13° phase modulation is plain wrong.
The jamming and noise resilence is indeed degraded by the low modulation
index compared to what could have been acheived using a 180° modulation.
But using only +/-13° modulation ensures that people who are phase locking to
the DCF77 carrier do not run into the troubles that are well documented on
this mailinglist.  Also the 645Hz modulation spreads the signal over a wider
bandwidth. Instead of the very narrow tone you get a BW of ~1000Hz of the
first lobe. And the gain in BW is proportional in gain of SNR you get
(roughly and with a lot of simplifications). Compare that to ~1Hz spread
you get for the new WWVB signal. And unlike the WWVB signal, where you
don't know what the phase modulation signal will be, with DCF77 you exactly
know it (pseudo random bit string), hence you can use this knowledge to
remove the phase modulation and gain the above mentioned dB's in SNR
for better frequency/phase lock (or jamming/noise resilience). When using
DCF77 as a timing reference, you get a few hundred additional edges at known
positions to figure out when the second exactly began. By using these
additonal edges you can average out a lot of noise and hence get a better
noise resilience as timing source as well. 

If they would really have cared about resilience to MSF interfering
with WWVB reception, then they would have employed something similar
to what DCF77 does, which works the same way as a very well known
and tested same frequency transmitter de-jamming tool: CDMA


Oh.. and Brian. Dont worry about being shot. My laser gun is ineffective
for anything at a distance of more than one meter ;-)


			Attila Kinali
-- 
The people on 4chan are like brilliant psychologists
who also happen to be insane and gross.
		-- unknown



More information about the Time-nuts_lists.febo.com mailing list