5
\$\begingroup\$

A few years ago the Germans added a pseudo-random noise signal (from a LFSR) to their time reference DCF77. I heard this should allow you to get a more accurate signal at the receiver side. But the signal delay due to the distance (up to 2000 km) and due to the electronics is the same for the basic AM signal as for the pseudo-random noise, isn't it?
How can I use the added pseudo-random noise?

\$\endgroup\$
2
\$\begingroup\$

According to Meinberg:

The carrier frequency of 77.5 kHz is amplitude-modulated (AM) with time marks at the beginning of each second. Radio clocks decode those time marks in order to yield the current date, time, and status. Since the AM signal is often superimposed by interfering noise, there received signal must be filtered with low bandwidth, causing a skew and a jitter of the demodulated time code in the range of 10±3 ms.

This accuracy is often not suitable for modern applications, so the PTB introduced an additional modulation which allows for increased accuracy using correlation techniques for the receiver.

In addition to the AM, the carrier signal is phase-modulated with a pseudo-random noise sequence (PZF, PRN). The PRN sequence contains 512 bits which are transmitted by phase modulation between the AM time marks, including the same number of logical '0' and logical '1' bits to keep the average phase of the carrier signal unchanged. The maximum phase deviation is ±10° and the length of one bit corresponds to 120 periods of the 77.5 kHz carrier frequency, i.e. 1.55 ms. The complete PRN sequence is transmitted once every second, starting 200 ms after the beginning of a new second and ending shortly before the next one.

The phase modulated signal can be received with higher bandwith than the AM signal. The radio clock generates the same PRN sequence locally and uses correlation techniques to derive the time frames with a dispersion of a few microseconds only. Additionally, correlation technique provides a high degree of immunity against interference and electrical noise.

(my emphasis)

More details of the PRN algorithm etc are at the usual place

Even more details in PTB paper 1988

\$\endgroup\$
  • \$\begingroup\$ If you're 300 km from Mainflingen you have a delay of minimum 1 ms (more added by electronic components). How relevant is microsecond dispersion then? OP seems to ask how or if the pseudo-random noise will solve this. \$\endgroup\$ – Johan.A Oct 15 '13 at 14:21
  • \$\begingroup\$ Delay isn't the problem, as it's easily corrected by adding a constant. The problem, as the first paragraph says, is the low-pass filtering. The phase modulated signal can be received with higher bandwith than the AM signal, thus the jitter is less. \$\endgroup\$ – Phil Frost Oct 15 '13 at 15:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.