|
Ok Mal
Well a data mode was the original idea ,
actually what has evolved (so far) was
never really envisaged , but during testing , the
s/n level of the decodes started to drop and
we had a ring side seat at a race to the noise floor
....
I did try some time ago to get some
interest in adding , arq function to the popular
data modes , after the original set of simple
'shoot out' data tests , the original mfsk
did give the best results , but at wide b/w , but
never managed to gain any ground , other than with
the email handling facility of fldigi
, I think Jim was the first to receive a
radio email from me on 500 using this
method.
The problem with 'All' the soundcard
based data modes is , you need to be able
to transmit, either translated frequency shift and
some times amplitude variation , which drops most
home made kit out of the race , ROS -MF modes
still require frequency translation from
audio to Rf , but runs round the
-22/-27 dB level and can pass via class e/d amps
This is where the Op idea came from , could we use
a cw tx. to send serial data ,
The data coding seems to be a
set length of 6 chr's for Op ,
the original concept , and assumedly still
is , that the Op system would have
modes running , a robust mode
to provide the link /lock and call sign and a
lesser mode to act as the data transfer ....
this is similar to the ROS mode , where the
initial lock sequence is more robust
than the fasted data transfer . I think we have
the lock phase on test at the moment ...
FEC is provided , by way of the data
spreading , not in the classic interleaving /
overlaying stepped in time , but as result of the
numeric spreading , which allow 's for a
random loss of round 55% of the data
to occur , whilst still decoding.
The most simply way to visualise the
process is a jig-saw , where all the pieces of
the puzzle are gathered together , then assembled
to re-create the original data , that
why there is no Rx progress indicator , nothing
is really known until the final decode
phase is entered.
Thanks' to all , makes all the time
Gary and I sat during testing, waiting for a
ding , and yes the song is quite right , 'it don't
mean a thing if it don't go ding' worth while ...
G..
Sent: Thursday, February 02, 2012 9:47 AM
Subject: LF: OPERA QSO
Graham
Opera seems popular, are there any plans ahead for a
basic QSO mode instead of Beacon only.
At present on MF/LF it has popularity in its favour
over the other data modes.
Instead of FEC how about the amtor approach where the
transmission triggers a RX stn who responds and a QSO begins.
The exchange need only be a few key words.
How about modifying Amtor or Pactor for MF use.
mal/g3kev
| 