Re: [Freetel-codec2] FreeDV data mode

[David Rowe]

Hi Mark,

Yes I am interpreting it the other way, ie the "spurious emissions
attenuated by ... the lessor of (23,70) == 23dBc for very low power signals.

However it's a bit confusing and you (and Glen) might be right . I
will seek clarification from ACMA.  It may also be different in other
parts if the world.

-/-

Also in response to Glen's earlier email - much of the slow fall off of
the spectrum on the blog post
(http://www.rowetel.com/images/rpitx/rpitx_spectrum_10kbps.png) is due
it being unfiltered FSK (i.e. not really about the Pi).

In busier bands, the bandwidth could be reduced by using a lower bit
rate.  Today we were testing in on 2M at 1000 bit/s and below, and we
could hardly hear or see the carrier on a SDR (let alone the spurious).

In higher bands, there tends to be more space available, so less of a
problem.

Cheers,
David


On 10/8/20 5:07 pm, Mark Jessop wrote:
> I think David is going purely by the spurious emissions limits from the
> Amateur LCD:
> image.png
> 
> So for 10mW TX power, 43+10log(0.01) = 23 dB.
> HOWEVER, it's also worth looking at how the above legislation is
> interpreted.
> When we talk about spurious emissions, we usually talk in dB relative to
> the carrier of the wanted signal (dBc). So the result of the above
> calculation should really be expressed as -23 dBc, with the other limit
> being -50 dBc.
> The letter of those two is -50 dBc. Is this how it should be
> interpreted? I don't know.
> 
> How appropriate are these limits when attempting to play nicely with
> other users of the amateur bands is another issue. 
> I certainly wouldn't want to be amplifying the output of a RPi TX
> without doing something to seriously clean it up, if that's even possible.
> 
> 73
> Mark VK5QI
> 
> On Mon, Aug 10, 2020 at 3:47 PM glen english  > wrote:
> 
> David, looking at that link and plots you sent, that radio will not
> meet
> community and other band user spectral purity expectations.
> 
> At UHF, It needs to be better than -70dBc in an SSB Bandwidth immediate
> adjacent
> 
> At 1MHz away, -110dBc in an SSB BW...
> 
> and proportionally better at lower frequencies
> 
> -40dBc in band you will make alot of other radio amateurs unhappy...
> 
> Here is a worked example from your 144.5 plot.
> 
> +10dBm  - 40  = -30dBc in SSB BW
> 
> Now, how about an SSB users 10km away line of site (-95dB), say 0dBi to
> 0dBi,
> 
> you put -30 - 95 = -125dBm noise in his receiver, an effect desense of
> about 18dB.
> 
> and that is only with +10dBm TX power.
> 
> g
> 
> 
> On 10/08/2020 7:52 am, David Rowe wrote:
> > With the addition of a LPF I believe it meets the spectral purity
> > requirements of the Amateur LCD on 2M:
> >
> >    http://www.rowetel.com/?p=7207
> >
> > and on HF:
> >
> >    http://www.rowetel.com/?p=6317
> >
> > - David
> >
> > On 10/8/20 6:50 am, glen english wrote:
> >> except that the Pi transmitter has unacceptable spectral purity.
> >>
> >> On 9/08/2020 7:18 pm, Adrian Musceac wrote:
> >>> Thanks David for the news.
> >>> One question: how high can the Pi transmitter go in frequency?
> >>> Our 70 cm band is pretty crowded, but the 23 cm band and
> 2305-2400 MHz
> >>> bands are underutilized and we want to make use of them. The Pi/RTL
> >>> system is a good candidate because of price.
> >>>
> >>> 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
> 
> -- 
> Glen English
> RF Communications and Electronics Engineer
> 
> CORTEX RF
> &
> Pacific Media Technologies Pty Ltd
> 
> ABN 40 075 532 008
> 
> PO Box 5231 Lyneham ACT 2602, Australia.
> au mobile : +61 (0)418 975077
> 
> 
> 
> 
> ___
> 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

Re: [Freetel-codec2] FreeDV data mode

[Helmut Oeller]

NO, Pluto as well as LimeSDR perform quite well on all VHF, UHF and SHF
bands. Variable preamp gain and/or total receiver gain of 50 dB and more
guarantee the claimed 2.5 dB NF. Full-duplex capability is a great feature
for satellite communication. Meanwhile approx. 30 percent of QO-100 traffic
uses these small transceivers with big success. AD and Lime microsystems are
no 'hawker's tray' companies and you can trust (or verify) their data sheet.

73, Helmut, DC6NY

-Ursprüngliche Nachricht-
Von: Steve [mailto:coupaydevi...@gmail.com]
Gesendet: Montag, 10. August 2020 00:00
An: freetel-codec2@lists.sourceforge.net
Betreff: Re: [Freetel-codec2] FreeDV data mode

When the Pluto came out, I had to have one. Playing with it kind of left me
saying "meh" because it didn't seem to have much RX gain. I needed a beam to
focus on the transmitter.

I then tried the LimeSDR-Mini but that thing only lasted about a year and
then failed. Expensive junk. Comparably, the Pluto is much more hobby
oriented with the ability to run Linux. Still running after several years.
It's not cheap at $250 USD, but think of it as a Pi combined with a radio.

On Sun, Aug 9, 2020 at 1:29 AM Adrian Musceac  wrote:
>
> Hi David,
>
> I've watched with interest your recent foray into FreeDV data
transmission.
> There's a new and interesting project called HNAP (https://hnap.de ) which
has the same goals (long distance IP links). The project is well described
in this year's SDRA talk by Lukas O.
> Inspired by both your project and theirs, I wrote a 4FSK data mode which
has the following characteristics:
>
> - 100 kbaud, 1/2 convolutional code
> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region
> 1 allocations)
> - full duplex
> - works with 50 watt FM amplifiers
> - good SNR performance, 20 km point to point access distance tested
> with small non-directional antennas
> - carries layer 2 (Ethernet) frames
>
> Do you have anything planned for VHF as well?
>
> Best regards,
> 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


--
Diese E-Mail wurde von AVG auf Viren geprüft.
http://www.avg.com



___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode NOT ACCEPTABLE SPECTRAL PURITY

[glen english]

David, looking at that link and plots you sent, that radio will not meet 
community and other band user spectral purity expectations.


At UHF, It needs to be better than -70dBc in an SSB Bandwidth immediate 
adjacent


At 1MHz away, -110dBc in an SSB BW...

and proportionally better at lower frequencies

-40dBc in band you will make alot of other radio amateurs unhappy...

Here is a worked example from your 144.5 plot.

+10dBm  - 40  = -30dBc in SSB BW

Now, how about an SSB users 10km away line of site (-95dB), say 0dBi to 
0dBi,


you put -30 - 95 = -125dBm noise in his receiver, an effect desense of 
about 18dB.


and that is only with +10dBm TX power.

g


On 10/08/2020 7:52 am, David Rowe wrote:

With the addition of a LPF I believe it meets the spectral purity
requirements of the Amateur LCD on 2M:

   http://www.rowetel.com/?p=7207

and on HF:

   http://www.rowetel.com/?p=6317

- David

On 10/8/20 6:50 am, glen english wrote:

except that the Pi transmitter has unacceptable spectral purity.

On 9/08/2020 7:18 pm, Adrian Musceac wrote:

Thanks David for the news.
One question: how high can the Pi transmitter go in frequency?
Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz
bands are underutilized and we want to make use of them. The Pi/RTL
system is a good candidate because of price.

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


--
Glen English
RF Communications and Electronics Engineer

CORTEX RF
&
Pacific Media Technologies Pty Ltd

ABN 40 075 532 008

PO Box 5231 Lyneham ACT 2602, Australia.
au mobile : +61 (0)418 975077




___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[glen english]

I doubt it will meet adjacent channel, and in-band noise community 
expectations.


What is the noise level 10kHz  and 100kHz away, in an SSB bandwidth ?

(dont measure in 1Hz BW, measure in the wide band and convert)

g

On 10/08/2020 7:52 am, David Rowe wrote:

With the addition of a LPF I believe it meets the spectral purity
requirements of the Amateur LCD on 2M:

   http://www.rowetel.com/?p=7207

and on HF:

   http://www.rowetel.com/?p=6317

- David

On 10/8/20 6:50 am, glen english wrote:

except that the Pi transmitter has unacceptable spectral purity.

On 9/08/2020 7:18 pm, Adrian Musceac wrote:

Thanks David for the news.
One question: how high can the Pi transmitter go in frequency?
Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz
bands are underutilized and we want to make use of them. The Pi/RTL
system is a good candidate because of price.

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


--
Glen English
RF Communications and Electronics Engineer

CORTEX RF
&
Pacific Media Technologies Pty Ltd

ABN 40 075 532 008

PO Box 5231 Lyneham ACT 2602, Australia.
au mobile : +61 (0)418 975077




___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[Adrian Musceac]

Toko filters are a few dollars retail.

On August 9, 2020 11:14:34 PM UTC, Steve  wrote:
>I may have to download Elsie then and grab a pencil to make some
>coils...
>:-)
>
>http://www.tonnesoftware.com/elsie.html
>
>
>
>On Sun, Aug 9, 2020 at 6:02 PM Adrian Musceac 
>wrote:
>
>> The Pluto receiver is useless without proper filters as it will
>receive
>> everything at the same time. With filters, it works fine, I managed
>more
>> than 800 km on FT8 with it last winter, I talked a little bit about
>it at
>> SDRA.
>>
>> Adrian
>>
>>
___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[Steve]

I may have to download Elsie then and grab a pencil to make some coils...
:-)

http://www.tonnesoftware.com/elsie.html



On Sun, Aug 9, 2020 at 6:02 PM Adrian Musceac  wrote:

> The Pluto receiver is useless without proper filters as it will receive
> everything at the same time. With filters, it works fine, I managed more
> than 800 km on FT8 with it last winter, I talked a little bit about it at
> SDRA.
>
> Adrian
>
>
___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[Adrian Musceac]

The Pluto receiver is useless without proper filters as it will receive 
everything at the same time. With filters, it works fine, I managed more than 
800 km on FT8 with it last winter, I talked a little bit about it at SDRA.

Adrian

On August 9, 2020 9:59:53 PM UTC, Steve  wrote:
>When the Pluto came out, I had to have one. Playing with it kind of
>left me saying "meh" because it didn't seem to have much RX gain. I
>needed a beam to focus on the transmitter.
>
>I then tried the LimeSDR-Mini but that thing only lasted about a year
>and then failed. Expensive junk. Comparably, the Pluto is much more
>hobby oriented with the ability to run Linux. Still running after
>several years. It's not cheap at $250 USD, but think of it as a Pi
>combined with a radio.
>
>On Sun, Aug 9, 2020 at 1:29 AM Adrian Musceac 
>wrote:
>>
>> Hi David,
>>
>> I've watched with interest your recent foray into FreeDV data
>transmission.
>> There's a new and interesting project called HNAP (https://hnap.de )
>which has the same goals (long distance IP links). The project is well
>described in this year's SDRA talk by Lukas O.
>> Inspired by both your project and theirs, I wrote a 4FSK data mode
>which has the following characteristics:
>>
>> - 100 kbaud, 1/2 convolutional code
>> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region
>1 allocations)
>> - full duplex
>> - works with 50 watt FM amplifiers
>> - good SNR performance, 20 km point to point access distance tested
>with small non-directional antennas
>> - carries layer 2 (Ethernet) frames
>>
>> Do you have anything planned for VHF as well?
>>
>> Best regards,
>> 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

Re: [Freetel-codec2] FreeDV data mode

[Steve]

On Sun, Aug 9, 2020 at 5:51 PM David Rowe  wrote:
>
> I wonder if that was a configuration thing Steve, at max analog gain the
> noise figure of the Pluto (AD9363) is specified at just 2.5dB - better
> than many other SDRs and hardware-defined Ham radios.
>
> Gain is pretty cheap in SDRs, once you've set your system noise figure
> just x10 in software if you want 20dB more gain :-)
>
> - David
>

That is a possibility. I was using it with GNU radio and I have to
admit that program gives me fits!


___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[David Rowe]

> When the Pluto came out, I had to have one. Playing with it kind of
> left me saying "meh" because it didn't seem to have much RX gain. I
> needed a beam to focus on the transmitter.

I wonder if that was a configuration thing Steve, at max analog gain the
noise figure of the Pluto (AD9363) is specified at just 2.5dB - better
than many other SDRs and hardware-defined Ham radios.

Gain is pretty cheap in SDRs, once you've set your system noise figure
just x10 in software if you want 20dB more gain :-)

- David





___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[Steve]

When the Pluto came out, I had to have one. Playing with it kind of
left me saying "meh" because it didn't seem to have much RX gain. I
needed a beam to focus on the transmitter.

I then tried the LimeSDR-Mini but that thing only lasted about a year
and then failed. Expensive junk. Comparably, the Pluto is much more
hobby oriented with the ability to run Linux. Still running after
several years. It's not cheap at $250 USD, but think of it as a Pi
combined with a radio.

On Sun, Aug 9, 2020 at 1:29 AM Adrian Musceac  wrote:
>
> Hi David,
>
> I've watched with interest your recent foray into FreeDV data transmission.
> There's a new and interesting project called HNAP (https://hnap.de ) which 
> has the same goals (long distance IP links). The project is well described in 
> this year's SDRA talk by Lukas O.
> Inspired by both your project and theirs, I wrote a 4FSK data mode which has 
> the following characteristics:
>
> - 100 kbaud, 1/2 convolutional code
> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region 1 
> allocations)
> - full duplex
> - works with 50 watt FM amplifiers
> - good SNR performance, 20 km point to point access distance tested with 
> small non-directional antennas
> - carries layer 2 (Ethernet) frames
>
> Do you have anything planned for VHF as well?
>
> Best regards,
> 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

Re: [Freetel-codec2] FreeDV data mode

[David Rowe]

With the addition of a LPF I believe it meets the spectral purity
requirements of the Amateur LCD on 2M:

  http://www.rowetel.com/?p=7207

and on HF:

  http://www.rowetel.com/?p=6317

- David

On 10/8/20 6:50 am, glen english wrote:
> except that the Pi transmitter has unacceptable spectral purity.
> 
> On 9/08/2020 7:18 pm, Adrian Musceac wrote:
>> Thanks David for the news.
>> One question: how high can the Pi transmitter go in frequency?
>> Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz
>> bands are underutilized and we want to make use of them. The Pi/RTL
>> system is a good candidate because of price.
>>
>> 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

Re: [Freetel-codec2] FreeDV data mode

[glen english]

except that the Pi transmitter has unacceptable spectral purity.

On 9/08/2020 7:18 pm, Adrian Musceac wrote:

Thanks David for the news.
One question: how high can the Pi transmitter go in frequency?
Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz 
bands are underutilized and we want to make use of them. The Pi/RTL 
system is a good candidate because of price.


Adrian



___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[David Rowe]

Hi Adrian,

The RpiTx doc suggest 1.5GHz, but I haven't tried it.  Suspect it gets
"noisier" as freq increases, however as the power is so low it may still
meet spurious requirements.

Cheers,
David

On 9/8/20 6:48 pm, Adrian Musceac wrote:
> Thanks David for the news.
> One question: how high can the Pi transmitter go in frequency?
> Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz
> bands are underutilized and we want to make use of them. The Pi/RTL
> system is a good candidate because of price.
> 
> Adrian
> 
> On August 9, 2020 8:14:52 AM UTC, David Rowe  wrote:
> 
> Hi Adrian,
> 
> You data mode (and HNAP) sound pretty cool.  Thanks, I enjoyed the SDRA
> HNAP talk by Lukas, that's quite a sophisticated system.
> 
> -/-
> 
> A couple of recent developments in this area by Bill, VK5DSP, and myself:
> 
> 1/ We have ported the 4FSK/LLR/LDPC code work to C in this PR
> 
>   https://github.com/drowe67/codec2/pull/136
> 
> Some command line examples down the bottom, e.g. 5 bits/s with a MDS of
> -160dBm (!), and 100 bit/s across-town links on 25uW (!!):
> 
>   http://www.rowetel.com/?p=7273
> 
> You can scale to any bit rate you like of course, if you have the link
> margin.
> 
> The 4FSK/LLR/OFDM work has been carefully developed to hit the
> theoretical rate limits, which I understand are about Eb/No=7dB
> (information bits) for 4FSK.  Pls correct me if I'm not explaining this
> correctly Bill!
> 
> 2/ The RPi/RTLSDR based system is moving along, yesterday we managed to
> sent a few packets over a 10km non-LOS (135dBm loss) link on 145.5 MHz
> at 5000 bits/s, using 13dBm Tx power.  Current code/status here:
> 
>   https://github.com/drowe67/pirip
> 
> 3/ Steve and I have also been work on the HF data modes.  We have
> initial versions of 3 new OFDM/LDPC data modes running in C, between 10
> and -5dB SNR.  Next step is some tuning and QAM16.  That modem might
> also run OK at higher rates on VHF and above, but we haven't tried it.
> 
> Cheers,
> David
> 
> 
> 
> On 9/8/20 3:57 pm, Adrian Musceac wrote:
> 
> Hi David,
> 
> I've watched with interest your recent foray into FreeDV data
> transmission.
> There's a new and interesting project called HNAP (https://hnap.de )
> which has the same goals (long distance IP links). The project
> is well
> described in this year's SDRA talk by Lukas O.
> Inspired by both your project and theirs, I wrote a 4FSK data
> mode which
> has the following characteristics:
> 
> - 100 kbaud, 1/2 convolutional code
> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU
> region 1
> allocations)
> - full duplex
> - works with 50 watt FM amplifiers
> - good SNR performance, 20 km point to point access distance
> tested with
> small non-directional antennas
> - carries layer 2 (Ethernet) frames
> 
> Do you have anything planned for VHF as well?
> 
> Best regards,
> 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
> 


___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2

Re: [Freetel-codec2] FreeDV data mode

[Adrian Musceac]

Thanks David for the news.
One question: how high can the Pi transmitter go in frequency?
Our 70 cm band is pretty crowded, but the 23 cm band and 2305-2400 MHz bands 
are underutilized and we want to make use of them. The Pi/RTL system is a good 
candidate because of price.

Adrian

On August 9, 2020 8:14:52 AM UTC, David Rowe  wrote:
>Hi Adrian,
>
>You data mode (and HNAP) sound pretty cool.  Thanks, I enjoyed the SDRA
>HNAP talk by Lukas, that's quite a sophisticated system.
>
>-/-
>
>A couple of recent developments in this area by Bill, VK5DSP, and
>myself:
>
>1/ We have ported the 4FSK/LLR/LDPC code work to C in this PR
>
>  https://github.com/drowe67/codec2/pull/136
>
>Some command line examples down the bottom, e.g. 5 bits/s with a MDS of
>-160dBm (!), and 100 bit/s across-town links on 25uW (!!):
>
>  http://www.rowetel.com/?p=7273
>
>You can scale to any bit rate you like of course, if you have the link
>margin.
>
>The 4FSK/LLR/OFDM work has been carefully developed to hit the
>theoretical rate limits, which I understand are about Eb/No=7dB
>(information bits) for 4FSK.  Pls correct me if I'm not explaining this
>correctly Bill!
>
>2/ The RPi/RTLSDR based system is moving along, yesterday we managed to
>sent a few packets over a 10km non-LOS (135dBm loss) link on 145.5 MHz
>at 5000 bits/s, using 13dBm Tx power.  Current code/status here:
>
>  https://github.com/drowe67/pirip
>
>3/ Steve and I have also been work on the HF data modes.  We have
>initial versions of 3 new OFDM/LDPC data modes running in C, between 10
>and -5dB SNR.  Next step is some tuning and QAM16.  That modem might
>also run OK at higher rates on VHF and above, but we haven't tried it.
>
>Cheers,
>David
>
>
>
>On 9/8/20 3:57 pm, Adrian Musceac wrote:
>> Hi David,
>> 
>> I've watched with interest your recent foray into FreeDV data
>transmission.
>> There's a new and interesting project called HNAP (https://hnap.de )
>> which has the same goals (long distance IP links). The project is
>well
>> described in this year's SDRA talk by Lukas O.
>> Inspired by both your project and theirs, I wrote a 4FSK data mode
>which
>> has the following characteristics:
>> 
>> - 100 kbaud, 1/2 convolutional code
>> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region
>1
>> allocations)
>> - full duplex
>> - works with 50 watt FM amplifiers
>> - good SNR performance, 20 km point to point access distance tested
>with
>> small non-directional antennas
>> - carries layer 2 (Ethernet) frames
>> 
>> Do you have anything planned for VHF as well?
>> 
>> Best regards,
>> 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

Re: [Freetel-codec2] FreeDV data mode

[David Rowe]

Hi Adrian,

You data mode (and HNAP) sound pretty cool.  Thanks, I enjoyed the SDRA
HNAP talk by Lukas, that's quite a sophisticated system.

-/-

A couple of recent developments in this area by Bill, VK5DSP, and myself:

1/ We have ported the 4FSK/LLR/LDPC code work to C in this PR

  https://github.com/drowe67/codec2/pull/136

Some command line examples down the bottom, e.g. 5 bits/s with a MDS of
-160dBm (!), and 100 bit/s across-town links on 25uW (!!):

  http://www.rowetel.com/?p=7273

You can scale to any bit rate you like of course, if you have the link
margin.

The 4FSK/LLR/OFDM work has been carefully developed to hit the
theoretical rate limits, which I understand are about Eb/No=7dB
(information bits) for 4FSK.  Pls correct me if I'm not explaining this
correctly Bill!

2/ The RPi/RTLSDR based system is moving along, yesterday we managed to
sent a few packets over a 10km non-LOS (135dBm loss) link on 145.5 MHz
at 5000 bits/s, using 13dBm Tx power.  Current code/status here:

  https://github.com/drowe67/pirip

3/ Steve and I have also been work on the HF data modes.  We have
initial versions of 3 new OFDM/LDPC data modes running in C, between 10
and -5dB SNR.  Next step is some tuning and QAM16.  That modem might
also run OK at higher rates on VHF and above, but we haven't tried it.

Cheers,
David



On 9/8/20 3:57 pm, Adrian Musceac wrote:
> Hi David,
> 
> I've watched with interest your recent foray into FreeDV data transmission.
> There's a new and interesting project called HNAP (https://hnap.de )
> which has the same goals (long distance IP links). The project is well
> described in this year's SDRA talk by Lukas O.
> Inspired by both your project and theirs, I wrote a 4FSK data mode which
> has the following characteristics:
> 
> - 100 kbaud, 1/2 convolutional code
> - fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region 1
> allocations)
> - full duplex
> - works with 50 watt FM amplifiers
> - good SNR performance, 20 km point to point access distance tested with
> small non-directional antennas
> - carries layer 2 (Ethernet) frames
> 
> Do you have anything planned for VHF as well?
> 
> Best regards,
> 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] FreeDV data mode

[Adrian Musceac]

Hi David,

I've watched with interest your recent foray into FreeDV data transmission.
There's a new and interesting project called HNAP (https://hnap.de ) which has 
the same goals (long distance IP links). The project is well described in this 
year's SDRA talk by Lukas O.
Inspired by both your project and theirs, I wrote a 4FSK data mode which has 
the following characteristics:

- 100 kbaud, 1/2 convolutional code
- fits in a 200 kHz channel on the 70 cm and 23 cm bands (IARU region 1 
allocations)
- full duplex
- works with 50 watt FM amplifiers
- good SNR performance, 20 km point to point access distance tested with small 
non-directional antennas
- carries layer 2 (Ethernet) frames

Do you have anything planned for VHF as well?

Best regards,
Adrian___
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2