discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Tutorial on complex signals and receiving SSB no


From: Marcus Müller
Subject: Re: [Discuss-gnuradio] Tutorial on complex signals and receiving SSB no audio
Date: Fri, 13 Sep 2019 13:47:52 -0500
User-agent: Evolution 3.32.4 (3.32.4-1.fc30)

Hey Chris,

I think you're on the right way! I'll address a couple of things
regarding your flow graph below, but I think the thing you need is less
of a detailed critical (grumpy) review of your flow graph, but some
learning material.

I think complex signals, you might want to go and enjoy Mossmann's
tutorials: https://greatscottgadgets.com/sdr/

Best regards,
Marcus
 

Now, my grumpy mutterings :)

Throttle is wrong here
----------------------

GNU Radio companion will very loudly complain about you having a
Throttle in the same flow graph as a hardware sink (in the console
output in the lower left) – your soundcard is already defining the rate
at which it's consuming samples, the throttle  doesn't *do* anything to
the signal, it just slows down copying its input to its output: you
should *never* have a Throttle in a flowgraph, unless you are in a
simulation only where you, for visualization reasons, want to slow down
the movement of samples. Again, this block doesn't mathematically do
anything to your signal.

Sampling Rates don't look sensible
----------------------------------

Going backwards through your flow graph:
The audio sink wants 44.1 kS/s. You resample by 4/3, so that means you
need to supply a signal sampled at 58.8 kS/s to the input of the
resampler. That input is running at 1/8 of the input of the Freq
X'lating FIR filter, which hence would need input sampled at 470.4
kS/s. I'm positively sure you don't have that. From the file name
alone, my _guess_ (nobody can tell, that info is not part of the file!)
that the signal in that file was sampled at 256 kS/s.

You've got frequency translation in there, so you do bad frequency
translation atop of that
------------------------------------------------------------------

The purpose of the frequency xlating FIR is to shift in frequency and
filer your signal. After that, you decompose the complex values into
real and imaginary part and multiply them with sine and cosine – that's
just exactly what complex multiplication does. I'm not quite sure how
you came up with this way – it's interesting to see (I've used
something similar in educational settings to show that complex
multiplication with complex sinusoids shifts signals), but it's not
useful.


On Fri, 2019-09-13 at 11:48 -0400, Chris Gorman wrote:
> Hello All,
> 
> I'm trying S. Katz's tutorial
> (
> http://www.csun.edu/~skatz/katzpage/sdr_project/sdr/grc_tutorial4.pdf
> )
> on SSB and can't seem to get any audio out of the gnuradio-companion
> flow graph I've generated.  I'm sure the error is mine, but can't
> figure out where I made it.
> 
> I'm using gnuradio 3.7.13.5 under Linux and have attached my grc
> file.
> If anyone has the time to try it out, you will need to download
> ssb_lsb_256k_complex2.dat from
> http://www.csun.edu/~skatz/katzpage/sdr_project/sdr/ssb_lsb_256k_complex2.dat.zip
> and unzip it, then change the file source in my tutorial4.grc file to
> point to the unzipped file.
> 
> Thanks in advance,
> 
> Chris
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio




reply via email to

[Prev in Thread] Current Thread [Next in Thread]