Re: help for a sdr-based sound-art installation

[Fabian Schwartau]

Hi Stefano,

don't worry asking such questions here on the list. That's its purpose I
guess ;)
First of all: I was able to run both of your flowgraphs on my PC with a
HackRF One, so the problem lies somewhere else.
A small google search reavealed no perfect solution, but it indicated
that something may be wrong with your installation. Like gnuradio was
build against a certain library version but you are using a different one.
How did you install gnuradio on the Pi? What Pi is it exactly? An
original Pi 1 (A/B)? You may have trouble feeding 10 Msps though that
old hardware. But it should start anyway.
What OS are you using? Which version?
I would recommend installing gnuradio from the distro repositories using
apt. If you did so, a very simple solution may be to try a different
distribution (or version), like Raspi OS (Raspbian), Ubuntu, Arch, ...
there are many for the Pi. This will likely be a straight forward
workaround for your problem.
Nevertheless, I just tried it on one of my Pi 3 with a fairly clean and
up-to-date Raspi OS. I installed the following stuff:
$ sudo apt install gnuradio gr-osmosdr hackrf
And both of your flow graphs worked like a charm with that.

If you have any further questions, feel free to ask them :)

Hope I could help,
Fabian


Am 13.02.21 um 15:04 schrieb Stefano Zorzanello:
> Dear list members,
> 
> I’m sorry to bother you with the classic newbie stuff, but I’m stuck -
> since some days (the solutions found on the web didn't work for me) - to
> this point:
> 
> - gnuradio-companion RaspberryPi 3b + hackRF one with a very simple
> blocks flow doesn’t display any signal window;
> and the error message I get is the following one:
> 
> - - - -
> 
> Warning: failed toXInitThreads()linux; GNU C++version 6.2.0 20161010;
> Boost_106100; UHD_003.009.005-0-unknowngr-osmosdr 0.1.4(0.1.4) gnuradio
> 3.7.10built-in sourcetypes: file osmosdr fcd rtl rtl_tcp uhd miri hackrf
> bladerf rfspaceairspy soapy redpitaya *** Error in`/usr/bin/python':
> corrupted double-linked list: 0x01a81c08 ***
> 
> - - - -
> 
> I'm trying to follow some very nice youtube tutorial (Ossman, Hackaday
> etc. ) but I cannot proceed if I can't neither make grc working with a
> simple sketch.
> 
> Any help is greatly appreciated.
> 
> More, if you have any suggestion or advice, here it follows a general
> description of the process I want to develop for my sound-art / sdr
> based installation.
> 
> The set up should be as follows:
> 
> - a small robot(flor hover type) moves randomly in a restricted area
> (few sq meters)of a big room.
> On this device, an RPI(“A”) + [DVB-T+DAB+FM] usb dongle,
> battery-powered, is mounted to detect some RF activity in some rf range.
> Some values should be extracted (intensity over time of a particular
> freq // and intensity of shifting detected frequencies). The movements
> of the robot in the room should cause different detected values.
> 
> - these values should be shared or sent in real-time to an external
> software (PureData) that sends these values via OSC to another RPI(“B”),
> which is steady, that will use the incoming values as a control signal
> for some audio DSP-treatments (sampling and granulation). The raw audio
> material for dsp audio treatments will be just RF noise.- The audio
> treatments will be played inside the room via loudspeakers.
> 
> - through a microphone and dsp, the live audio stream will be analyzed
> in PureData (intensity and spectrum-centroid) in real-time by the steady
> RPI. The values extracted from the audio-domain should be sent to
> GNURADIO-COMPANION to transmit some signals via HackRF one in the very
> same frequency range which is detected by RPI(“A”).
> 
> - the idea is to build a close circuit of information generation, which
> should be modulated over time and ever-changing, due to the conversion
> of values between the domains radio/audio, and to the implication of the
> room acoustic response analyzed and encoded to interfere in the
> radiowaves domain.
> 
> The suggestion of this idea comes from the works of Agostino Di Scipio
> “Audible Ecosystemics” (2000-2005) which develop very deeply feedback
> network processes based in the pure audio/acoustic domain. My purpose is
> to try to do something similar in a cross-domain set (Radio/Audio
> through Sonification and Dsp treatments)
> thank you very much for your kind attention and help,
> 
> best
> 
> Stefano Zorzanello
>