[time-nuts] Carrier phase tracking

[Matt Ettus]

On 2/20/07, Peter Vince <pvince at theiet.org> wrote:
>
> >I don't wish to offend, but that is totally wrong.  Even if every
> >satellite were on the exact same frequency, with the exact same
> >doppler, and the exact same direction, you would still have no problem
> >separating the carriers of the multiple satellites.  The reason is
> >that the satellites do not transmit carrier.  Here is how:
> >
> >   Everybody here understands how the PN codes are tracked by
> >despreading.  The despreading operation is multiplication, and when
> >you multiply the PN code by the original signal you get a carrier,
> >only from that one satellite.
>
> Do you?  Or do you just get the modulation?

Yes, you get the 50 bps modulation.  But since you can demod the bits,
you can then multiply the bits into the signal, then you get the
carrier.

In practice, you don't usually get to a time sampled sequence of the
carrier.  You typically track it in a pll whose output is frequency
and phase.

Matt