K3 Questions

[Corboy - Poteet]

I asked this earlier but suppose it was lost in the early frenzy so I
will pose the question again along with a couple more.

The basic question concerns digital voice modulation.  There is no
current standard but AOR is using an open protocol from G4GUO. Unlike
SSB, digital voice will require some sort of standard unless we will
be satisfied talking only to ourselves (not all bad, at least I would
find someone who always agrees with me). I wonder if Elecraft could
push things along by offering digital voice modulation as a built-in
function on the K3?

1) Basic question: could the current K3 DSP handle DVM encode/decode
processing while performing its other tasks?

2) Looking at the K3 photos, there seems to be (a lot of?) room under
the hood.  I wonder if Elecraft has designed the K3 (as they did with
the K2) with plans to add additional functional modules to the
original system?  I am thinking, in this instance, of an independent
DSP module that could do magic such as DVM without burdening the basic
DSP. Heck, with an independent DSP you could even add cell phone
capability to the K3 (might be nice to have out in the boonies).

3) Digital voice modulation question: I assume the basic bandwidth of
a DVM signal is 3 khz.  What happens, on receive, when you narrow the
bandwidth of a DVM signal?  Do you just lose fidelity or do you lose
data (things like syllables or words)?

I am not a big voice mode fan but I do believe that SSB is headed
toward the same ditch currently occupied by AM.  Maybe Elecraft can
help set the direction for future voice modes.


Mike    W5FTD




_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Julian, G4ILO]

On 6/27/07, Corboy-Poteet <[hidden email]> wrote:
>
> I am not a big voice mode fan but I do believe that SSB is headed
> toward the same ditch currently occupied by AM.  Maybe Elecraft can
> help set the direction for future voice modes.

Isn't digital voice just another fad like some of the new data modes?
I don't know much about it, but as a digital mode, I guess that you
get perfect copy until the signal falls below a certain level, after
which you don't hear anything at all. This seems to be counter to the
main ham radio interest of working weak DX stations. I understand
people's interest in experimenting, but don't see any real use for
digital voice on the HF bands. As an experimental mode it seems
inappropriate to support it directly within the K3.
--
Julian, G4ILO
G4ILO's Shack: www.g4ilo.com
K2 s/n: 392  K3 s/n: ???
www.Ham-Directory.com: the best ham resources on the net
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[John GM4SLV]

On Wed, 27 Jun 2007, Julian G4ILO wrote:

> On 6/27/07, Corboy-Poteet <[hidden email]> wrote:
> >
> > I am not a big voice mode fan but I do believe that SSB is headed
> > toward the same ditch currently occupied by AM.  Maybe Elecraft can
> > help set the direction for future voice modes.
>
> Isn't digital voice just another fad like some of the new data modes?
> I don't know much about it, but as a digital mode, I guess that you
> get perfect copy until the signal falls below a certain level, after
> which you don't hear anything at all.

...then you switch to analogue SSB until the signal goes below the
threshold of S/N that makes SSB readable, after which you switch to CW
and give yourself another 17dB to play with...

> This seems to be counter to the main ham radio interest of working
> weak DX stations. I understand people's interest in experimenting, but
> don't see any real use for digital voice on the HF bands.

I think the same about analogue voice modes  ;-)

Cheers,

John GM4SLV

--
G-GRP-Club 2377, QRP-ARCI 12384, SKCC 3214
Member : RSGB, ARRL
Shetland Islands (EU-012) IP90GG
Lerwick Radio Club : http://www.gm3zet.org
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill Steffey NY9H]

At 03:29 AM 6/27/2007, John GM4SLV wrote:
>On Wed, 27 Jun 2007, Julian G4ILO wrote:
>
>...then you switch to analogue SSB until the signal goes below the
>threshold of S/N that makes SSB readable, after which you switch to CW
>and give yourself another 17dB to play with...

when icom demos their dstar digital audio, they suggest that the
digital will work 'deeper' into the noise than the "analog".  my
concern when a fireman or cop is at the fringe,,, which will get thru
the message ??? who cares if it is scratchy if it gets thru...  kinda
like us  . did we make the contact , get the info or not.
Already there is some flak i read about background noise at a fire
scene can confuse and disrupt CURRENT digital audio on public service radios.

Would be cool for Digital Lyle ( eric&wayne) to have a digital audio
mode available for the k3.... esp if we can change out the algorithm
as they improve...

bill

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Stuart Rohre]

Absolutely good points about the fail soft nature of analogue signals.
While the new digital systems usually have a threshold, and fail hard, ie,
the message does not get thru even partially, after a certain loss of signal
point!

This is why analogue ham communications get thru in disasters, and many
Public Safety systems do not, as they depend more and more on go/ no go
digital systems.

-Stuart
K5KVH
retired fireman


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Corboy - Poteet]

Stuart, what bandwidth do the Public Service people use on their
digital systems?  And let me ask you my earlier question: what happens
on digital voice when you decrease the bandwidth on reception? Can you
significantly improve signal-to-noise by narrowing the bandwidth (with
only a loss of fidelity)?  Or does digital voice have no tolerance for
bandwidth reduction; that is, you need basically all of the bits in
order to provide intelligible audio out.

Mike   W5FTD



> Absolutely good points about the fail soft nature of analogue signals.
> While the new digital systems usually have a threshold, and fail hard, ie,
> the message does not get thru even partially, after a certain loss of signal
> point!

> This is why analogue ham communications get thru in disasters, and many
> Public Safety systems do not, as they depend more and more on go/ no go
> digital systems.

> -Stuart
> K5KVH
> retired fireman



_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Stuart Rohre]

In the basics of digitizing the signal you have to clock the conversion at
least twice the highest frequency you want to reproduce in the voice (or
audio) signal.

Now, you can sample a number of ways.  You can use one sample per clock
cycle, or you could do some of the phase shift techniques in encoding,
taking multiple samples per cycle. (Quadrature sampling for example).

Thus, if you narrow the band in which the digital signal is transmitting,
you normally give up frequency response and intelligibility.  Note the AOR
open protocol system developed by the G -ham transmits into your audio input
of a SSB radio, thus the bandwidth is no more than the SSB bandwidth, or at
least that is what I got from an AOR ad.

Another big issue with digital coding is duty cycle.  It might not be within
the finals power envelope to support digital modes other than CW or TTY; as
TTY for example, continuously transmits a signal, and its information is
varied by the shifting of the frequency, while the amplitude stays at a
constant level.  TTY has a high duty cycle compared to on off CW.

In most rigs where you want to run PSK or TTY you back off the power to half
of SSB.  For PSK, it often is enough to do that, and still get the message
for the variants of PSK that are error corrected.

I think hams often want everything in one box, but don't want to pay the
price that would require.  There is no free lunch in RF engineering or other
types.

Until there is a world wide standard agreement on a digital modulation for
voice for ham radio use, there is not much sense in marketing a unique
system in the radio.  In fact, the advantage of ham radio, is that it is
mostly analog, and the signal can still be decoded, (heard) under conditions
of fading, which is not true for digital methods that set a minimal SNR
threshold to work.  Thus in disaster, we can get a message thru when wide
bandwidth, power hungry digital encoding fails.

Most digital systems such as trunking radio, have single points of failure
which are catastrophic.  Ham communications are multi point to multi point,
or redundant. IF one path fails, there is another station to station path
that can work.  The cost of sophisticated digital encoding such as Pactor
III means that not many hams are going to have a spare packet modem box if
the first one fails.  However, you usually have several microphones that can
work on a given radio at lower cost, if the first one fails.  It is also
easier to fix, as the failure modes are mostly mechanical, (connections)
rather than a combination of electronic circuits and mechanical connections.

-Stuart
K5KVH


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Matthew D. Pitts]

Guys,

I'm finally getting around to asking a question about the K3 that will help
a fellow ham decide whether or not he will be buying one; does it currently,
or is it possible for a future firmware revision, for it to support voice
feedback? My friend's reason for wanting to know is because his wife, who is
also a licensed Amateur, is blind, and any HF Radio needs to be something
she can use when he's not around to help her with it.

Matthew Pitts
N8OHU
K2 #5956

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Greg - AB7R]

Matthew,

This is covered well in the FAQ.  But for a short answer, yes.  Withe the optional
DVR installed, a portion of the available space for recording is reserved for audio
feedback for just such cases.  I believe this is still being tested so I am not
sure exactly which control items will give audio feedback, but certainly frequency
and mode.  Probably which XFIL is selected as well.  I'm sure there's more, but I
don't have the exact info yet.  When a final list is available, I will post it on
the FAQ page.

Thanks.

-----------------
73,
Greg - AB7R
Whidbey Island WA
NA-065

On Wed Jun 27 17:14 , "Matthew D. Pitts"  sent:

>Guys,
>
>I'm finally getting around to asking a question about the K3 that will help
>a fellow ham decide whether or not he will be buying one; does it currently,
>or is it possible for a future firmware revision, for it to support voice
>feedback? My friend's reason for wanting to know is because his wife, who is
>also a licensed Amateur, is blind, and any HF Radio needs to be something
>she can use when he's not around to help her with it.
>
>Matthew Pitts
>N8OHU
>K2 #5956
>
>_______________________________________________
>Elecraft mailing list
>Post to: [hidden email]
>You must be a subscriber to post to the list.
>Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft   
>
>Help: http://mailman.qth.net/subscribers.htm
>Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[David Woolley (E.L)]

Stuart Rohre wrote:
> In the basics of digitizing the signal you have to clock the conversion at
> least twice the highest frequency you want to reproduce in the voice (or
> audio) signal.

That's only at the input to the encoding chain (and technically it is
twice the bandwidth, not twice the highest frequency).

However, I would imagine all digital modes that would be used for
communications, rather than broadcasting (and for that matter, also
those used in modern broadcasting systems) don't send time domain data.
  One way or another they send frequency domain data, often in the form
of just the formant frequencies used in a model of vocal tract resonances.

The critical rate for these systems is the syllable rate, not the
frequency of the highest component.  I seem to remember that the
military were using 2400 bits per second codecs maybe a couple of
decades ago.    I suspect that was partly do do with how fast they could
encrypt.

Mobile phone codecs tend to be vocal tract model based, although they
use more than 2400 bps so that the voice sounds reasonably natural (but
try them on modem tones or even music!).

To get much better than 2400 bps, I think you would probably have to
recognize phonemes, which is basically the continuous speech voice
recognition problem.  I think that might get you down to about 300
bits per second.

--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Corboy - Poteet]

David. let me pose a question: if you converted the digitized voice
signal to mp3 or ogg format, is this similar to (or the same as) a
freq domain data stream?  The data compression from CD wav file to mp3
is around 10 to 1; you are certainly not storing a digital copy of the
sound wave.  Voice compression may not be so good but still should
result in the need to transfer considerably less data in order to
assure a useful voice recovery on the receiving end.

Mike  W5FTD





> Stuart Rohre wrote:
>> In the basics of digitizing the signal you have to clock the conversion at
>> least twice the highest frequency you want to reproduce in the voice (or
>> audio) signal.

> That's only at the input to the encoding chain (and technically it is
> twice the bandwidth, not twice the highest frequency).

> However, I would imagine all digital modes that would be used for
> communications, rather than broadcasting (and for that matter, also
> those used in modern broadcasting systems) don't send time domain data.
>   One way or another they send frequency domain data, often in the form
> of just the formant frequencies used in a model of vocal tract resonances.

> The critical rate for these systems is the syllable rate, not the
> frequency of the highest component.  I seem to remember that the
> military were using 2400 bits per second codecs maybe a couple of
> decades ago.    I suspect that was partly do do with how fast they could
> encrypt.

> Mobile phone codecs tend to be vocal tract model based, although they
> use more than 2400 bps so that the voice sounds reasonably natural (but
> try them on modem tones or even music!).

> To get much better than 2400 bps, I think you would probably have to
> recognize phonemes, which is basically the continuous speech voice
> recognition problem.  I think that might get you down to about 300
> bits per second.


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[David Woolley (E.L)]

Corboy-Poteet wrote:
> David. let me pose a question: if you converted the digitized voice
> signal to mp3 or ogg format, is this similar to (or the same as) a

I don't know the fine details of ogg, but mp3 is certainly frequency
domain, and I think ogg will be the same.

mp3 is fairly complicated, and designed for music, which is more
demanding that speech.  It basically works by noting that the ear cannot
detect weak sounds close in frequency to strong ones (the K2 is a bad
ear!) and therefore doesn't bother encoding those weak frequencies, but
in the end, everything is coded as the spectrum not the time domain
signal.  (This is very simplified, because I cannot remember the fine
details).

> freq domain data stream?  The data compression from CD wav file to mp3
> is around 10 to 1; you are certainly not storing a digital copy of the

Actually, that's the compression used to try and match CD quality.  For
speech, you can get away with about 16kbs (good enough for mono language
practice; if I remember correctly, the codec will go down to 8kbps).


--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com