Hi David,

Wow Adrian it's really coming along quickly!  Well done :-)  Combined
> with the latest crop of duplex SDRs, I can see this work opening up
> experimentation in the VHF/UHF bands.
>

Thanks! I've been working full time on this in the last month due to a
break at work that might end pretty soon.
I'm aiming at VHF to SHF work, small, portable and field configurable
repeaters (check out the LimeNet-micro and LimeRFE)
and satellite ops. The IP radio modem in qradiolink might also help point
to point IP link connectivity at low bandwidths but it needs more
improvement.
HF is not out of the question but due to QTH conditions can't really use it.
Telemetry modems (even direct sequence spread spectrum for very low
bitrates, like LoRa) and multi-channel repeaters are the next stop
(after release 1.0).


> Some thoughts:
>
> 1/ Codec 2 2400A and 2400B were designed to work on VHF/UHF, and 2400A
> to substantially outperform FM and current digital voice systems at low
> SNRs:
>
>   http://www.rowetel.com/wordpress/?p=5219
>
> They're in the FreeDV API already
>

I'll add 2400A for sure. I don't really see a reason to use 2400B with
SDRs. Maybe for testing purposes only with the FM modulator.
But right now my plate is pretty full trying to test everything already in
and do some packaging.
The 700D modem segfaults in the LDPC encoder with the version of libcodec2
in Debian stable and I didn't even have time to raise a bug with them.
But I'll add it as well once that's figured out. By the way, can I suggest
for libcodec2 Git tags for stable releases? I could not find any tags in
Github and they would be pretty useful as landmarks.


> 2/ Have you done any BER versus SNR (Eb/No) tests to evaluate the
> performance of the modems you are using, e.g. comparing performance to
> theoretical?  We have found many existing VHF/UHF digital voice systems
> have modems that perform poorly.  Gains of 10dB are possible with the
> right modem/waveform design.
>

Yes. All tests are good.
My custom modems are nothing original, they are based on the knowledge of
the GNU radio community, especially Daniel Estevez and Marcus Mueller but
also others.
Some problems occur with some modems like the DQPSK written by me used with
both Opus and Codec2
due to some implementation misconfigures. The Doppler correction used for
them tends to interact badly with the rest.
I'll fix that eventually.
But the FreeDV modems work very well. All numbers check out exactly as you
described in your website.


> 3/ RpiTx would be an interesting option for the TX side:
>
>   https://github.com/F5OEO/rpitx
>

RPiTx is a work of art, but it has some annoyances. It requires a PGA in
front to set the output power (some amplifiers
drive from below 0 dBm). The signals are not exactly clean, but the main
downside is the limited frequency range.
I do SHF (2.4 GHz and 5 GHz) with the PlutoSDR. Options to go all the way
up to 10 GHz with some other MyriadRF hardware.
There's the https://osmocom.org/projects/osmo-fl2k/wiki project which gives
you a very cheap SDR transmitter as well but with the same limitations.

Best regards,
Adrian


> Cheers,
> David
>
> On 17/10/19 10:30 pm, Adrian Musceac wrote:
> > Hi,
> > I made some more progress working on my application[1].
> > First of all, it is now possible to run it headless (maybe even
> > daemonized) and control it remotely using a telnet client.
> > Second, I have added support for mixed mode repeater (Codec2/FM or
> > viceversa or any combination of modes like wideband
> > Opus/Codec2/FreeDV/SSB etc.).
> >
> > The repeater can be connected to a VOIP server using the low latency
> > Mumble voice protocol (known especially for high-performance gaming). So
> > it is possible now to connect multiple repeaters together by putting
> > them on the same VOIP channel (this can be controlled via telnet as
> > well, although it's stil work in progress). They operate in full duplex
> > mode, so a VOIP user can talk at the same time as a radio user and the
> > two audio streams will be mixed together and broadcast. The
> > FreeDV/Codec2 radio frames are transcoded to Opus at very high bitrates
> > (~48 kbit/s) for VOIP transport. This may cause some small delays.
> >
> > I haven't really had time to test the cross-mode repeater so could use
> > some help there from interested parties.
> >
> > [1] http://qradiolink.org
> >
> > Cheers,
> > Adrian
> >
> >
> >
> >
> > _______________________________________________
> > Freetel-codec2 mailing list
> > Freetel-codec2@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/freetel-codec2
> >
>
>
> _______________________________________________
> Freetel-codec2 mailing list
> Freetel-codec2@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/freetel-codec2
>
_______________________________________________
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Reply via email to