View Single Post
  #14   Report Post  
Old March 17th 07, 07:10 AM posted to rec.radio.amateur.moderated
Dave Platt Dave Platt is offline
external usenet poster
 
First recorded activity by RadioBanter: Jul 2006
Posts: 464
Default Extension of PSK segment

In article .com,
wrote:

Can a real-time digital voice message be sent in the
width of an SSB voice signal and result in the same
effectiveness? (signal to noise, power requirements, lack of need to
synchronize, tolerance of interference and fading, etc.)?


The review of DRM-based digital voice in this month's QST makes a
point of noting that both the WinDRM and hardware-modem-based systems
require a pretty clean, fade-free propagation path in order to perform
well. On less clean paths, they're prone to drop out... I infer that
as soon as the facing or QRM is severe enough to overcome the forward
error correction coding, you lose an entire packet.

Not the end of the story at all.

Yes, PSK31 is too slow for large amounts of data - because
it wasn't designed for that. PSK31 was designed to be a
keyboard-to-keyboard mode that uses very little bandwidth and has
excellent performance with low S/N ratios. It was
meant as an improvement to FSK RTTY for such QSOs.


I've heard of at least two groups who have been working on a PSK31-based
bulk data transmission system - both systems uses both forward error
correction and an ACK/NAK protocol structure. It's not intended for
megabytes of data, but for semi-unattended transmission of modest
amounts of data during emergencies. For example, basic health&welfare
traffic (queries and "We're OK, are in the shelter" responses) can be
entered via online Web forms, the fields converted to a compact
representation and heavily compressed, and then sent out in big
batches via PSK31 or a similar narrow-bandwidth mode.

The idea isn't to replace SSB voice (or CW net traffic) but to
supplement it, reducing the operators' workload and reducing errors.
It's certainly not intended as a substitute for broadband!

Yup. And that's the "who's going to tie the bell on the cat" question.
Who will come up with that standard?

That's the key question to the whole issue. Who is going to
do all that development work and then give it away free?
G3PLX and a small group did it for PSK31.


Agreed! And it's already been established as a tradition.
PSK-31 is a classic example. Everything about it is wide
open and free-for-the-download.

OTOH, Pactor 2/3 is not free at all.
Implementing it requires buying a
specific hardware modem that is rather pricey. Some may
say that $600 for a modem isn't much in the scheme of
things, but even if that's true, it's the principle of proprietary
methods that goes against the grain.


There was some work going on towards an open-source higher-speed
HF-data protocol a couple of years ago - SCAMP. If I recall correctly
it's based on OFDM (like DRM) with heavy use of forward error
correction. The last I heard of it, it had worked out well under
clean-pathway conditions, but wasn't working all that well under
noisy/fade-prone conditions and wasn't yet considered "ready for prime
time" or (as yet) a serious competitor to Pactor 2/3. Haven't heard
anything more about it in the last year or so - it's possible that
development has stalled.

IOW, watch out for the Law of Unintentional Consequences.


a.k.a. "Oops!" :-)

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!