[time-nuts] WWVB Measurements

[Poul-Henning Kamp]

In message <4D90B859.2080907 at comcast.net>, Greg Broburg writes:

>There is a method to recover a very weak signal
>out of the mud that is fairly easy to build. 

I did this slightly differently:

Take a 1MSPS ADC and average it into 1000 buckets in round
robin form, one after the other.

This can be done very quickly in an interrupt routine or even in hardware.
The Aduc72xx family of ARM can do this running a sweat, any DSP
would have absolutely no trouble doing it.

In Europe use 2000 buckets so you can capture DCF77 also.

By multiplying the buckets with a suitable IQ signal, you can pull
out any signal of integral kHz frequency you care for, and measure
its relative phase and change of phase over time.

Here are some experiements I did many years ago with one million
buckets, so I could demodulate the second pulses:

http://phk.freebsd.dk/loran-c/CW/

I would love to find somebody who can design some proper hardware
for this, so that we can have an "VLF-All-band Time/Phase Receiver",
basically a receiver you can ask, at any time, "What is the phase
of the signal at N kHz, for any value of N you care for.

Poul-Henning

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at 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.