| |
PA2AGA > HDDIG 20.09.00 23:34l 190 Lines 7067 Bytes #999 (0) @ EU
BID : HD_2000_252I
Read: DC1TMA GUEST
Subj: HamDigitalDigest 2000/252I
Path: DB0AAB<DB0RGB<DB0MRW<DB0ERF<DB0BRI<DB0HAG<DB0ACH<DB0OVN<PI8JOP<PI8ZAA<
PI8HGL
Sent: 000920/2013Z @:PI8HGL.#ZH1.NLD.EU #:16298 [Den Haag] FBB $:HD_2000_252I
From: PA2AGA@PI8HGL.#ZH1.NLD.EU
To : HDDIG@EU
Date: Sun, 17 Sep 00 16:38:26 MET
Message-Id: <hd_2000_252I>
From: pa2aga@pe1mvx.ampr.org
To: hd_broadcast@pa2aga.ampr.org
X-BBS-Msg-Type: B
So a PSK31 RST might be 09 15 23 meaning 90% correct copy, a SNR
of 15 dB, and an IMD of -23 dB (minus assumed in the reports). This
gives the sending station real information. He can tell whether he needs
to increase (or decrease) power, and whether he needs to work on the
linearity of his system to improve signal purity. The software could offer
a "hot key" to insert this information automatically where a signal report
is required.
Gary
Gary Coffman KE4ZV | You make it |mail to ke4zv@bellsouth.net
534 Shannon Way | We break it |
Lawrenceville, GA | Guaranteed |
------------------------------
Date: Fri, 15 Sep 2000 01:14:15 GMT
From: "Hank Oredson" <horedson@att.net>
Subject: Psk Signal Reports
Just a few small comments below ...
--
... Hank
http://horedson.home.att.net
"Gary Coffman" <ke4zv@bellsouth.net> wrote in message
news:imm2ssg21d1m5h88nr7jk0civh38qo1iu8@4ax.com...
> On Wed, 13 Sep 2000 13:16:48 +1200, "Gerard Welford"
<gerard.welford@xtra.co.nz> wrote:
>
> >I have been reading a number of postings to the PSK31 group in
> >www.egroups.com and there seems to be a concensus of opinion that the RST
> >form of signal reports is not suitable for this mode. Not only is the
signal
> >strength misleading due to the S meter reading within the bandpass of the
> >receiver so that the loudest station on the band is recorded, but the tone
> >part of the report is really only applicable to cw transmissions.
> >It would appear that most favour the QRK type of report and I think if
> >enough PSK operators adopt this type of report it will quickly become
> >standard practice.
> >73......Gerry ZL1AWW
>
> The problem with the QRK reporting system is that it doesn't indicate
> *why* the signals are bad or excellent. So the operator isn't given any
> indication of what steps need to be taken to improve the link. Basically,
> it is just the 'R' of the RST system.
>
> Adding the S and T information is an improvement over the QRK report
> because it gives the operator potentially valuable information on the
> quantity and quality of the signal. That's why RST was instituted. What
> we need to do is to refine the S and T reports to make them meaningful
> for PSK31 signals.
>
> The first thing we need to do is divorce the notion of S meter readings
> from the 'S' report. Historically there is no parallel between them. The
> S report merely uses an arbitrary scale that goes from very weak (1)
> to very strong (9). There is no direct correlation with an S meter, which
> often did not even exist on receivers in the era in which RST was
> developed.
>
> S meters on current amateur equipment generally do not follow the
> somewhat mythic 6 dB per S unit anyway. They are in practice very
> nonlinear from one S unit to another, and typically range from 2 dB
> to 5 dB per S unit over their range. So the use of S units for reports
> is very misleading. (This nonlinearity is intentional to a large extent,
> manufacturers have found that amateurs tend to prefer receivers
> with a "lively" S meter.)
>
> For the purposes of PSK31 (and other sorts of signals too), the more
> interesting S value is the signal to noise ratio. That information is
usually
> conveyed by color on the waterfall display, but it should be trivial for the
> software writers to give us a number for the selected signal in the same
> fashion as they now give us an IMD number.
Yup, this is what CLOVER does (reports SNR).
> The 'T' report is signal quality, and that's very important for PSK31. We
> already have a good measure of it. It is the IMD number reported by the
> software.
CLOVER reports phase dispersion as well.
> We shouldn't limit ourselves to single digits for each of 'R', 'S', and 'T'.
> That may have been adequate in CW days where "by ear" estimates
> were the rule, but our software can do better than that. I'd suggest
> that the 'R' number should report percent correct copy to the nearest
> 10 percent, IE 00 to 10. SNR should be in dB, a 2 digit range should be
> sufficient. T should be the IMD value, also a 2 digit value.
CLOVER reports information about sub-block reception.
> So a PSK31 RST might be 09 15 23 meaning 90% correct copy, a SNR
> of 15 dB, and an IMD of -23 dB (minus assumed in the reports). This
> gives the sending station real information. He can tell whether he needs
> to increase (or decrease) power, and whether he needs to work on the
> linearity of his system to improve signal purity. The software could offer
> a "hot key" to insert this information automatically where a signal report
> is required.
Then we add some FEC ... and ARQ ... and ... there we are!
I think this is actually a *really* good idea.
With the protocol in open source, folks can play with it and improve it.
Of course this would be a lot different than PSK31 ("Just like RTTY, except
it sounds a little different.")
> Gary
> Gary Coffman KE4ZV | You make it |mail to ke4zv@bellsouth.net
> 534 Shannon Way | We break it |
> Lawrenceville, GA | Guaranteed |
------------------------------
Date: Thu, 14 Sep 2000 22:36:02 -0400
From: "Bob Lewis" <rlewis@staffnet.com>
Subject: Psk Signal Reports
> Then we add some FEC ... and ARQ ... and ... there we are!
Right back where we started from :-) I though the idea of PSK31 was
to provide a friendly keyboard mode that didn't have the latency of
error correction and ARQ. Somehow we seem to want to change it so it's
just like Pactor and Clover - but then it won't be a friendly keyboard
mode any longer.
I do think Gary's reporting system makes sense.
------------------------------
Date: Fri, 15 Sep 2000 05:48:36 GMT
From: "Hank Oredson" <horedson@att.net>
Subject: Psk Signal Reports
Wouldn't be the same protocol.
What I was hinting at was that it might be fun to experiment
with various different protocols of intermediate "flavor" between
PSK31 / RTTY and PACTOR / CLOVER.
CLOVER has some of that idea with it's FEC mode.
One could use that mode much like one uses PSK31, and get
a bit of error correction as well.
--
... Hank
http://horedson.home.att.net
"Bob Lewis" <rlewis@staffnet.com> wrote in message
news:LOfw5.287$Cq3.1519@newsfeed.slurp.net...
> > Then we add some FEC ... and ARQ ... and ... there we are!
>
> Right back where we started from :-) I though the idea of PSK31 was
> to provide a friendly keyboard mode that didn't have the latency of
> error correction and ARQ. Somehow we seem to want to change it so it's
> just like Pactor and Clover - but then it won't be a friendly keyboard
> mode any longer.
>
> I do think Gary's reporting system makes sense.
>
>
>
------------------------------
End of Ham-Digital Digest V2000 #252
******************************
You can send in your contribution to this digest by
sending an e-mail to: hd-group@pa2aga.ampr.org
or (via BBS-net) to: hdaga@pi8vnw.#zh2.nld.eu
Read previous mail | Read next mail
| |
