[Yaesu] DSP-based radios (was: changes in rigs when CW dropped?)

Tue, 25 Nov 2003 11:22:55 -0800

Hi Brian,

It is interesting that since the mid-1990's, all HF equipment purchased by
military and governmental users, as well as larger commercial HF systems,
have been 100% DSP-based. They must sound OK to the acceptance-test
engineers and contract administrators; they certainly do the job they were
designed to do - "get the message through".

Please have a look at the Rohde & Schwarz, Rockwell-Collins, Harris and
Tales (Racal) websites.

Cheers for now, 73,
Adam VA7OJ/AB4OJ

-----Original Message-----
From: [email protected] [mailto:[email protected]]On
Behalf Of Brian Carling
Sent: Tuesday, November 25, 2003 05:37
To: [email protected]
Subject: Re: [Yaesu] changes in rigs when CW dropped?

Interesting... but with ALL of these improvements they will
never SOUND as good or as clean as a good old tube receiver
like a Collins, for example.

Computers can try all they want but they will still sound very dirty and
abrasive to
the ears.

I had not realized how BAD the solid state rigs ALL sound until I got hold
of a
couple of old Collins 51J receivers lately.
SO SWEET! And they hear everything better than all of these
$1000 - 3000 rice boxes with their DSP / APF / ANF / preamp /
mumbo-jumbo add-ons will ever do! It cracks me up!

With the filters in this WW2 military tube rig I can dig out things
that the new rigs can't hear because of their inherent digital
noise floor.

On 23 Nov 2003 at 14:06, David Willmore wrote:

>
> Hey, Jim.
>
> > When the code is dropped will all the rigs come without all the nice
> > CW features?
>
> My take on its is yes and no. :)  I say 'yes' because I do feel that
> we're in for a large shift in the way rigs are designed and used.  I
> say 'no' because it's not going to happen *exactly* then and it will
> not be because of the relaxation of the CW requirement for HF access.
>
> Rigs with the controls on the front pannel suffer for a number of
> limitations. One is that there is only so much space on the front of
> the pannel.  The pannel is mechanical in nature, so is not flexable.
> There are only so many controls and displays that can be fitted to
> this resource.  They can only be 'shared' amongst limited variety of
> functions before the user just forgets what all a knob does and when.
>
> I think we've seen some rigs like this.  I know I've heard the
> complaint that people were sticking with an older rig because the
> newer ones just had too many darn buttons and "der blinkin' lights".
> Several things have been tried with some success, but not enough to
> solve the problem--large LCD display, 'soft' buttons, etc..
>
> Also, the processing that occurs in a radio to acheive the level of
> performance that is becoming standard these days takes a lot of CPU
> ability.  It is true that using dedicated DSPs helps, but the flip
> side is the lack of flexability of that processor--it can only do
> signal processing, not your taxes or read your email. :)
>
> What I see us going to are simpler looking and, in a sense, simpler
> designed radios.  I hate to mention them, but they're ahead in this
> area *minutely*, but Icom has the PCR-100/1000 receivers which are
> more along this line.  But they don't go far enough to take advantage
> of the benefits of the new scheme.
>
> I see future radios as a 'block' that takes in/puts out RF energy on
> one side and communicates with a computer on the other.  Between the
> computer and the radio is just a digital connection--USB 2.0 would be
> good.
>
> Here are the bentfits:
> 1) simpler receiver design--no need to take the signal down to an
> analog baseband signal. 2) No big chunk of signal processing necessary
> 3) No display and forest of knobs 4) easier connection to the
> shack--three connections: power, RF, and computer.
>
> So, radios get cheaper and performance gets better.
>
> Since most people have/will have a computer, the most expensive chunk
> of this system can be considered a 'given'.  Sure, some people won't
> have one and some people will operate where one is not practical.
> That's one of the reasons that I say this won't happen instantly.  A
> replacement for the FT-817 using this type of design would be at least
> 5 years behind the FT-1000 version.
>
> With the use of a computer as the processing and control element in
> the system, you gain a large chunk of flexability in user interface
> and signal format. I'm sure to start with, vendors would ship software
> that displayed a 'virtual faceplace' type of design--like the Kenwood
> TS-2000 does.  But, that's just the first generation to 'bridge the
> gap'.  Don't need too many changes all at once or we'll lose everyone.
> :)  But, in the future things could change and take advantage of the
> advances that are being made in user interface design-- all those
> computer scientists coming through the colleges and universities are
> bound to come up with some interesting stuff so it would do us good to
> be in a position to take advantage of it--which would be hard to do
> with knobs and buttons.
>
> For the signal processing, well, one of the big limits now on the
> digital modes that are becoming so popular at HF is the bandwidth that
> one can get off and on the air.  It's not as much of a problem at HF
> as we're used to limiting ourselves to SSB bandwidths, but at VHF and
> above, it's signifigant. With the new scheme, we're not bringing in a
> little 3KHz chunk of analog baseband signal.  We could bring in a
> 100KHz chunk or larger (without all the baseband limitations like IM
> products, etc.  Anyone see the IM problems of a strong station in
> PSK31 smearing out the rest of the band?) and do the rest of the
> 'selection' and demodulation in software.  Adding support for new
> modes would be much easier as the hardware would not need to change.
>
> Oh, and you could still do CW with it.  We might need to add another
> 'port' to the radio 'brick' to plug a key into.  Or someone could hack
> up a little key to USB adapter.  Hmmmm, that sounds like a fun
> project....
>
> So, to summarize, yes and no. :)  I could go into more detail, but I'm
> very sure I've gone on long enough already.
>
> Before anyone goes off and says this is impossible, I'll mention that
> all of the fundimental chunks of this system are already done or are
> late in the design stages--from various groups.  GNURADIO is working
> on a software degined radio schema to allow flexable handling of big
> chunks of digitized RF.  They're also working on an RF to USB2.0
> converter (with digital downmixing and filtering).  Linrad is working
> on software for handling large bandwidths of signals and getting
> better performance in weak signal work than top of the line
> conventional radios.  So, this *is* were the hobby is going.  The
> concern is if our vendors (hey, Yaesu, you listening?) are going to
> lead, follow or get lost.  Well, we could say the same for our
> members, but every generation of radios in the past has always stirred
> up a cry of "this is as good as it gets, this is the last radio I'll
> ever buy."  Hi hi. Yeah, right. :)
>
> Cheers,
> David N0YMV
>
> Moderator: Ray Brown, KB�STN
> _______________________________________________
> Yaesu mailing list
> [email protected]
> http://mailman.qth.net/mailman/listinfo/yaesu

Moderator: Ray Brown, KB�STN
_______________________________________________
Yaesu mailing list
[email protected]
http://mailman.qth.net/mailman/listinfo/yaesu