[Top][All Lists]

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

Re: (python) resources for "offline" signal processing ?

From: Jeff Long
Subject: Re: (python) resources for "offline" signal processing ?
Date: Thu, 12 Aug 2021 08:00:25 -0400

[Back on list - my goof]

On Thu, Aug 12, 2021 at 7:50 AM Kristoff <kristoff@skypro.be> wrote:

Thanks for the very quick reply.
I don't know if you intended to reply just to me, or to the list.

Feel free to forward to the list if you want.

Well, let's take the example of HFDL.
 From what I have found, it's a packet of data that starts with a tone
(to allow the receiver to correct any frequency-offset), and then a
burst of data in BPSK, QPSK or 8PSK @ 1800 baud.

I guess, if you want to decode this, you
- first need to detect the start of the 'frequency-tone'
- measure any frequency offset of that tone
- multiply the rest of the packet with a tone at -x Hz
- then to PSK demodulation.
I am not sure of the PSK order is encoded in a header or so, so I guess
that you just have three decoders running in parallel and see which one
returns a valid bitstream.

It would be nice to do this one manually, so I can really inspect the
But I guess that if you would want to automate this (i.e. write a GR
demodulator block for this), then you need to write a python script that
implements this.

For that, I guess I need to learn what library (numpy, scipy-signal,
commpy, ... ) does what.
(I have the impression that these libraries overlap)

I do not know if there are already HFDL decoders in GR. For me, this is
just an exercise to learn how to do this.

And I think that HFDL is probably one of the less difficult protocols to
decode so it looks like a good place to start.
(just guessing here).


On 12.08.21 13:27, Jeff Long wrote:
> That's a good list. Is there some specific capability you are looking
> for?
> Baudline is an older tool, but is still useful for looking at files in
> a little more detail than Inspectrum.
> You didn't mention URH https://github.com/jopohl/urh
> <https://github.com/jopohl/urh>
> On Thu, Aug 12, 2021 at 7:15 AM Kristoff <kristoff@skypro.be
> <mailto:kristoff@skypro.be>> wrote:
>     Hi all,
>     Does anybody have a list of good python resources for "non-real-time"
>     signal-processing of radio-signals.
>     ("non-real time" as in "an iq-file of prerecorded dataframe", in
>     contrast a continuous stream like a GR flowgraph)
>     Say you start with the IQ-file of a HFDL-packet and you would want to
>     analyse and visualize it, take it apart, demodulate and decode it,
>     etc.
>     I've come across libraries like numpy, scipy-signal and commpy,
>     visualization-tools like matplotlib and seaborn and others.
>     For audio, there is off course audacity.
>     For quick-and-dirty visualization of radio-packets, I have used
>     inspectrum.
>     So there does not seams to be a lack of tool, .. perhaps there are to
>     many of them :-)
>     It would be nice to have a overview of what tools/libraries are
>     available to do what, and what would the good method to build a
>     'flow'
>     to -say- build a GR block for a HFDL packet decoder.
>     I have already found pysdr.org <http://pysdr.org> to be an
>     interesting resource.
>     Anybody an idea of other resources?

reply via email to

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