Our group has settled on the specific modes we use through trial and
error under the exact conditions and bands we would find ourselves using
in an emergency. By running 3 nets a week for well over a year now,
we've found that on 80m and 10m, Olivia 8/500 is the best compromise for
speed, accuracy, tuning forgiveness and bandwidth for shorter messages
and keyboard chat. On occasion we have to switch to the much slower
16/500 olivia, but that's rare.
For larger messages and spreadsheets on 80m and 10m we use MT63 1k long.
We do a broadcast transmission with a WRAP file and then go around the
table with amazingly good results with all but the stations reporting
the weakest rx signals.
On FM we use MT63 2k long. Again, it seems to be the best compromise
for speed, accuracy and tuning forgiveness.
Interestingly enough, there are a ton of new digital sound card emcomm
and traffic nets popping up on 80m, many of which tried MFSK16, Domino
and others, but they all seem to land on 8/500 olivia just like we did.
And I've heard nothing but good things from those using MT63 2k long on
FM. We have tried the PSK modes (including R) as well as the very
wide/fast MFSK and Olivia modes on FM, but nothing seems to out perform
MT63 2k long.
Anyway, my point is that on paper and s/n ratio tests, many modes come
out as looking better than Olivia and MT63, but there's no substitute
for running MANY nets with more than one other station under varying
conditions.
We're always willing to try new modes, but we always wind up with Olivia
and MT63 being the best overall fit for our needs here in WPA.
-Dave, KB3FXI
Brian Lloyd wrote:
On Sat, Apr 17, 2010 at 1:43 PM, Stelios Bounanos <[email protected]> wrote:
It is really Really *REALLY* important to understand what is going on
with
these protocols in the presence of propagation anomalies in order to be
able
to make decisions about which modes work best under what conditions. My
own
recent experiments in monitoring and graphing the phase/frequency changes
of
the WWV carrier over time (20-30 minute samples typically) have convinced
me
that sequential testing of protocols, i.e. transmitting a message and
then
immediately transmitting the same message using the other protocol, is
not
likely to produce valid results unless repeated numerous times and then
the
results averaged with outliers discarded.
True, and I seriously doubt that anyone has bothered to do that.
Well, Chen has taken a first pass using HF channel simulator software which
tells us something. The CCIR evidently did a fair amount of work to quantify
the characteristics of an HF channel. Chen used that.
(I am finding that I am coming at this from behind several people who have
done significant work but it doesn't seem to me that anyone has taken it
beyond building a couple of modems.)
I have been pestering Dave to do a test version of fldigi that would
let me transmit the same data in two modes at the same time in the
same passband so that both experience *exactly* the same propagation
conditions concurrently.
This is somewhat more complex than it sounds...
Oh, I figured that.
what would work right
now would be to generate two audio files (with signals at suitable
frequencies of course) and mix them using another program. I'm sure sox
can do that in one short command line.
<dope slap to the forehead> Duh! Of course! And that would be very simple.
Of course, that is what Chen did for his testing too.
In fact if fldigi had xmlrpc methods to start/stop audio file
generation, it would be easy to write a script to do an audio file for
each mode (for a particular input text), and then you could mix them as
needed for your tests. I guess I'll need to write those xmlrpc methods
first :-)
[snip]
Time to take the next step and build on Chen's work by getting real,
controlled data from live propagation.
Some comparisons have been posted to the digitalradio yahoo group.
There is one by K2MO, and another by WZ7I. URL for the latter:
http://mysite.verizon.net/wz7i/digitalmodes.html
The tests done by K2MO are in the files area of digitalradio, and he has
posted them to the group a few times too.
I can't say how scientific those experiments were, but I know that the
path simulator used (pathsim) is GPL'ed so at least that aspect of the
tests can be verified by anyone so inclined.
The problem as I see it is that we have these similar-but-different fixed
modes; e.g. olivia, contestia, MFSKx, DominoEX, and then the various flavors
of PSK; that make multiple changes but don't allow us to evaluate the
protocols while only changing one parameter at a time. That is certainly
useful for hams wondering which mode to try now but isn't helpful in
selecting which mode is likely to perform best. Where *is* the crossover
between IFK, MFSK, and PSK?
Hmm, wouldn't it be nice if the transmission header contained a descriptor
of the MODEM/CODEC so that the receiver could put together the correct
pieces.
ES1HJ also has published some test results but I don't recall if they
were on-air or simulation tests.
OK, there is a lot more research than I was aware of. I am trying to soak
all this up (and feeling like I am drinking from a fire hose) with the idea
of eventually building something to automate the process, i.e. evaluate the
channel and then configure the best operating mode. Thank you for all the
pointers Stelios.
------------------------------------------------------------------------
_______________________________________________
fldigi-alpha mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/fldigi-alpha
_______________________________________________
fldigi-alpha mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/fldigi-alpha
