average power rating of your radio

[David Rowe]

Hello,

I'm looking at ways to reduce the peak to average power ratio of FreeDV
waveforms. However I'm curious about how much of an advantage this will
give us on common SSB radios.

What sort of average power output can your radio tolerate in SSB mode?
For example if your radio is rated at 100W PEP for SSB can it handle 50W
continuous using PSK31?

Thanks,

David

[jdow]

PSK31 has IMD issues. I've very often found that getting the other guy to reduce
his power output decreases the error rate. I think the magic number was about 25
Watts for a 100 Watt transceiver. (ProII and FT-847) And ALC absolutely must be
turned off or the ALC needle should not even be twitching a little. (I use the
latter strategy - and then back off another dB or so.)

{^_^} Joanne/W6MKU

[Graham / KE9H]

David:

Most ham radios are rated "ICAS" which is Intermittent Commercial and Amateur Service.

The spec is full power (100 percent of PEP) as in CW, FM or RTTY for 5 minutes key down,
then 50 percent duty cycle (one minute on, one minute off) indefinitely.

You limitation is going to be the PEP, which is the RF output Wattage rating of the radio,

not the duty cycle.

Typically a 100 Watts ham radio is going to hard clip slightly above 100 Watts.

So, I would not worry about the average power rating.  Just do what ever you can do to
get the average power as high as you can, relative to the peak.

I did some testing with the current FreeDV waveforms.  The peak to average is a very

reproducable 6 dB, without any obvious clipping.  Although theoretically it could more like
12 or 13 dB for a 17 tone system, I never see that high a peak. 

I think the reason is that you always modulate the subcarriers,

even though there may not be any voice modulation occuring, and as a result you never
get the higher peaking, which pretty much requires 17 unmodulated sine wave subcarriers to occur.

I suspect the easiest way to start the peak to average improvement is to deliberately

start raising the average, letting the waveform peaks clip, which will introduce bit errors, and

strengthen the error detection to "heal" across the errors.

After that, you are looking at studying the waveforms to see which bit patterns induce the

highest peak, and modifying your modulation to not use the worst ones.

--- Graham

==

--
You received this message because you are subscribed to the Google Groups "digitalvoice" group.
To unsubscribe from this group and stop receiving emails from it, send an email to digitalvoice+unsubscribe@googlegroups.com.
To post to this group, send email to digita...@googlegroups.com.
Visit this group at http://groups.google.com/group/digitalvoice.
For more options, visit https://groups.google.com/d/optout.

[David Rowe]

Hi Graham,

How did you measure the PAPR of FreeDV? When I measure the PAPR of the
tx waveform in my Octave simulations I get about 12.6dB.

Cheers,

David

> send an email to digitalvoice+unsubscribe@__googlegroups.com
> <mailto:digitalvoice%2Bunsu...@googlegroups.com>.

> To post to this group, send email to digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.
> Visit this group at http://groups.google.com/__group/digitalvoice
> <http://groups.google.com/group/digitalvoice>.
> For more options, visit https://groups.google.com/d/__optout
> <https://groups.google.com/d/optout>.

>
>
> --
> You received this message because you are subscribed to the Google
> Groups "digitalvoice" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to digitalvoice...@googlegroups.com
> <mailto:digitalvoice...@googlegroups.com>.

> To post to this group, send email to digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.

[Graham / KE9H]

To find the peak, I put an oscilloscope on the RF output of the radio under test
(through an appropriate RF power attenuator,) looking at the

RF envelope and sweep the trigger level to find the highest repetitive peak.

By putting out a known RF power CW signal, I can calibrate the scope

voltage versus power for the attenuator set-up.

I depend on an average reading power sensor to determine the average

level.

The highest repetitive peaks are all the same level, and are 6 dB above

the average.  So setting the peaks to 100 Watts on 100 Watt amplifier

gives you distortion free approximate 25 Watt average, which is what I

think you determined empirically.

I expected to see something slightly stronger than  10 log N = 10 log (17) = 12.3 dB

The 10 log N approximation assumes all subcarriers are the same level,
but your sync carrier is 3 dB stronger, so I just guessed it was between

12 and 13 dB.

But, what I measure is 6 dB PAR, with no obvious clipping.

So either it is your continuous subcarrier modulation, or there is some

clipping down inside the signal processing, but before one of the

upsampling filters, which is smoothing everything out again.

--- Graham

==

    <mailto:digitalvoice@googlegroups.com>.

    Visit this group at http://groups.google.com/__group/digitalvoice
    <http://groups.google.com/group/digitalvoice>.
    For more options, visit https://groups.google.com/d/__optout
    <https://groups.google.com/d/optout>.

--
You received this message because you are subscribed to the Google
Groups "digitalvoice" group.
To unsubscribe from this group and stop receiving emails from it, send

an email to digitalvoice+unsubscribe@googlegroups.com
<mailto:digitalvoice+unsub...@googlegroups.com>.

To post to this group, send email to digita...@googlegroups.com

<mailto:digitalvoice@googlegroups.com>.

Visit this group at http://groups.google.com/group/digitalvoice.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "digitalvoice" group.

To unsubscribe from this group and stop receiving emails from it, send an email to digitalvoice+unsubscribe@googlegroups.com.
To post to this group, send email to digita...@googlegroups.com.

[Graham / KE9H]

David:

The scope photos attached.

I don't know if Google groups allows attachment, so I am sending direct, also.

All three are 100 Watt PEP.

FreeDV as set up on the FLEX-6000.

Un processed SSB human speech, typicaly estimated at 12 dB PAR.

Heavily processed SSB human speech, estimated at about 2.5 to 3 dB PAR.

--- Graham

[David Rowe]

Thank you Graham, very interesting. I'm measuring about 12dB on the tx
wavefile, and about the same for speech files. So bit of a mystery
where the other 6dB has gone .....

I'm currently playing with compression ideas for PAPR reduction on a new
FreeDV waveform. For example clipping or a non-linear function. I can
reduce PAPR about 5dB, but the modem loses about 2dB of performance due
to the distortion. So a net 3dB gain if the PA can handle the higher
average level. A 2dB loss if it can't. Could always make it switch
able I guess.

Cheers,

David

On 08/11/14 11:24, Graham / KE9H wrote:
> David:

> <mailto:da...@rowetel.com <mailto:da...@rowetel.com>>> wrote:
>
> Hello,
>
> I'm looking at ways to reduce the peak to average power
> ratio of
> FreeDV waveforms. However I'm curious about how much of an
> advantage this will give us on common SSB radios.
>
> What sort of average power output can your radio
> tolerate in SSB
> mode? For example if your radio is rated at 100W PEP
> for SSB can it
> handle 50W continuous using PSK31?
>
> Thanks,
>
> David
>
> --
> You received this message because you are subscribed to
> the Google
> Groups "digitalvoice" group.
> To unsubscribe from this group and stop receiving
> emails from it,
> send an email to

> digitalvoice+unsubscribe@__goo__glegroups.com
> <http://googlegroups.com>
> <mailto:digitalvoice%__2Buns...@googlegroups.com
> <mailto:digitalvoice%252Buns...@googlegroups.com>__>.

> To post to this group, send email to
> digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>
> <mailto:digitalvoice@__googlegroups.com
> <mailto:digita...@googlegroups.com>>.
> Visit this group at
> http://groups.google.com/____group/digitalvoice
> <http://groups.google.com/__group/digitalvoice>

> <http://groups.google.com/__group/digitalvoice
> <http://groups.google.com/group/digitalvoice>>.
> For more options, visit

> https://groups.google.com/d/____optout
> <https://groups.google.com/d/__optout>

> <https://groups.google.com/d/__optout
> <https://groups.google.com/d/optout>>.
>
>
> --
> You received this message because you are subscribed to the
> Google
> Groups "digitalvoice" group.
> To unsubscribe from this group and stop receiving emails
> from it, send

> an email to digitalvoice+unsubscribe@__googlegroups.com
> <mailto:digitalvoice%2Bunsu...@googlegroups.com>
> <mailto:digitalvoice+...@googlegroups.com

> <mailto:digitalvoice%2Bunsu...@googlegroups.com>>.
> To post to this group, send email to
> digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>
> <mailto:digitalvoice@__googlegroups.com
> <mailto:digita...@googlegroups.com>>.

> Visit this group at
> http://groups.google.com/__group/digitalvoice
> <http://groups.google.com/group/digitalvoice>.
> For more options, visit https://groups.google.com/d/__optout
> <https://groups.google.com/d/optout>.
>
>
> --
> You received this message because you are subscribed to the
> Google Groups "digitalvoice" group.
> To unsubscribe from this group and stop receiving emails from

> it, send an email to digitalvoice+unsubscribe@__googlegroups.com

> <mailto:digitalvoice%2Bunsu...@googlegroups.com>.
> To post to this group, send email to
> digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.

> Visit this group at
> http://groups.google.com/__group/digitalvoice
> <http://groups.google.com/group/digitalvoice>.
> For more options, visit https://groups.google.com/d/__optout
> <https://groups.google.com/d/optout>.
>
>
>
> --
> You received this message because you are subscribed to the Google
> Groups "digitalvoice" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to digitalvoice...@googlegroups.com
> <mailto:digitalvoice...@googlegroups.com>.

> To post to this group, send email to digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.

[jdow]

17 orthogonal tones will show a peak some 12 dB higher than the average, in
theory. In practice some limiting is not damaging, except to your neighbors on
the band. Modulation should have nothing to do with it unless the modulation
contains an amplitude component as well as a phase component.

{o.o} Joanne/W6MKU

> email to digitalvoice+unsubscribe@__googlegroups.com
> <mailto:digitalvoice%2Bunsu...@googlegroups.com>.

> To post to this group, send email to digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.
> Visit this group at http://groups.google.com/__group/digitalvoice
> <http://groups.google.com/group/digitalvoice>.
> For more options, visit https://groups.google.com/d/__optout
> <https://groups.google.com/d/optout>.

>
>
> --
> You received this message because you are subscribed to the Google Groups
> "digitalvoice" group.
> To unsubscribe from this group and stop receiving emails from it, send an email

> to digitalvoice...@googlegroups.com
> <mailto:digitalvoice...@googlegroups.com>.

> To post to this group, send email to digita...@googlegroups.com

> <mailto:digita...@googlegroups.com>.

[Marciniak, Ed]

‎The Galileo is sort of like a pentium 400 with dp floating point and some tradeoffs of reduced speed for a smaller die. It features Arduino compatible IO headers and a USB host port among other things. (consumes around 3.5W)

It's newer brother, the Edison, has a compute module similar in ‎size to a US priority mail stamp.
‎The compute module is reported to be an underclocked dual core atom running at 500 MHZ, potentially drawing less than a watt depending on Whether Wifi is active. It also happens to have an I2S port as well as serial and some GPIO.

Both run Yocto Linux distributions.

Before reinventing, I thought I'd ask if anyone has tried or contemplated running codec and/or a software modem.

To unsubscribe from this group and stop receiving emails from it, send an email to digitalvoice...@googlegroups.com.
To post to this group, send email to digita...@googlegroups.com.