FPGA Central

there has been a thread that discussed de-rotation of received I,
samples, that the reason is freq synchorization.

Can someone explain in more details how this process is going? Suppose
received I, Q samples, I should first estimate the freq offset? then rotat
the IQ along the offset?

thanks
patrick

On May 15, 10:09*am, "pzou" <patrick_...@hotmail.com> wrote:
> there has been a thread that discussed de-rotation of received I, Q
> samples, that the reason is freq synchorization.
>
> Can someone explain in more details how this process is going? Suppose I
> received I, Q samples, I should first estimate the freq offset? then rotate
> the IQ along the offset?
>
> thanks
> patrick

That's one way to do it. Or, estimate the phase offset from each
symbol
and "track" it using an appropriate method, typically based on
feedback
filtering (the so-called loop filter). Or, use a PLL or Costas
loop.

If all of this sounds alien to you, let's start with something very
basic.
What are you trying to do and what is your background?

Julius

If you are talking about GMSK (as mentioned in the subject line), the
de-rotation is not for removing frequency offset. The purpose is
convert GMSK moduloation (which is a form of MSK) into BPSK. The
reason we do this is simply because it is much easier to build a
receiver for detecting BPSK. How/Why this equivalence is achieved
requires another short explanation and some deep thinking....

Regards
Piyush

On 2009-05-15, pzou <[email protected]> wrote:
> there has been a thread that discussed de-rotation of received I, Q
> samples, that the reason is freq synchorization.
>
> Can someone explain in more details how this process is going? Suppose I
> received I, Q samples, I should first estimate the freq offset? then rotate
> the IQ along the offset?

You might want to read:

http://en.wikipedia.org/wiki/Differential_coding

--
Ben Jackson AD7GD
<[email protected]>
http://www.ben.com/