Re: [Discuss-gnuradio] FFSK/MSK Demodulation in Gnuradio

[Michael Kramer]

Am 01.12.2015 um 18:12 schrieb Michael Ossmann:
> On Tue, Dec 01, 2015 at 02:14:59PM +0100, Michael Kramer wrote:
> > I have the following problem. I've got a signal and I know it's FFSK
> > modulated with a bitrate of 2400. I've tried to read some tutorials or
> > find other information about FFSK (which is the same as MSK
> > apparently), but I found very little.
> I've never heard of FFSK, but it looks to me like you have a secondary
> frequency modulation.  In other words, you have a frequency modulated
> AFSK signal; the data bits are frequency modulated (FSK) onto a carrier
> in the audio range, and then that signal is frequency modulated onto an
> RF carrier.  It doesn't look like MSK.
>
> Try using a second frequency demodulation in your flowgraph.  One way
> would be to insert Hilbert->Quadrature Demod after your second LPF.
> Scope the output of the second Quadrature Demod, and you should see a
> squarish signal where data zeros are high and data ones are low (or vice
> versa).  You should be able to verify the symbol rate by analyzing this
> output and then pass it to a clock recovery block.  You'll probably need
> an additional step to center the signal around zero before clock
> recovery.
>
> Mike
First of thanks for your answer! I've seen every one if your tutorials
and it's an honor to get an answer from you^^

It's interesting that you've never heard of FFSK. I haven't found much
myself in the internet. Only that it's Fast Frequency Key Shifting and
should be the same as MSK.

I've tried your suggestion but again I'm a little bit confused by the
results. The attached Wave2 is produced if I decimate by 100 in my Low
Pass Filter right before the Hilbert -> Quadrature Demod. If I don't
before, but after the second Quadrature Demod I get Wave3. So I guess
I'm doing right by decimating within the Low Pass Filter.

I can even more or less see my data inside Wave2 (even though it seems
to be inverted. it looks like its 00000111111111001110110 instead of
111111000000000110001001), but I'm not really able to center it since
the upper spikes are crossing zero sometimes.

Also I wanted to ask if my calculation is correct: I have sampled with
2MHz. The bitrate (should be) 2400. So if I want 4 or 5 samples per
symbol for the Clock recovery I need to resample to 4*2400=9600 or
5*2400=12000. I've not found much on the Clock Recovery Block as well.
It seems for most it's a magical block which usually does what they
expect...

Greetings,
Michael


---
Diese E-Mail wurde von Avast Antivirus-Software auf Viren geprüft.
https://www.avast.com/antivirus

Wave2.PNG
Description: PNG image

Wave3.PNG
Description: PNG image