Hi Pedro - interesting data. I have a couple of possible explanations
and one potential, at least partial solution.
Conditions were pretty mixed on Sunday, at least up here. There was
some very rapid QSB. We know that most busts occur when a Morse
character is truncated by one element, or in the case of characters like
E, the whole character is lost. Do the busts of your call seem to fit
this, or any, particular pattern?
The second hypothesis is that we had a very large number of nodes on
during the contest. Since distribution of nodes is uneven (many in
Europe and N America, few in the Southern Hemisphere), it is certainly
possible that you were seeing each node'sinaccuracyof somewhere around
one percent multiplied by many nodes, when you were being heard well in
Europe or the US
Filtering at the node can help. I have been generally impressed by
CT1BOH's algorithm for his skimbusted filter on ARCluster V6, although
it is not infallible - sometimes it will mistake the bust for the real
call. Another approach is simply to use ARCluster V6's unique>x filter,
which doesn't "believe" a spot unless it is made more than x times. I
don't know what VE7CC may use to deal with this.
Finally, let me confirm there was nothing going on with the RBN
infrastructure. Our philosophy is to forward everything that our nodes
copy, and let users filter as they will.
73, Pete N4ZR
Download the new N1MM Logger+ at
<http://N1MM.hamdocs.com>. Check
out the Reverse Beacon Network at
<http://reversebeacon.net>, now
spotting RTTY activity worldwide.
For spots, please use your favorite
"retail" DX cluster.
On 8/11/2015 3:10 PM, Pedro Colla wrote:
Hi,
I'd observed during the contest that quite often my running frequency were
reported as being ocuppied by me (LT7H) and by DT7H who seems to be a seldom
used Italian callsign, but most likely a malformed capture of my callsign
during the contest.
It happened either when I was called by scores of non-European fellows (perhaps looking
for an Italian multiplicator), also when I turned off the "run" mode in N1MM+
and it showed outstanding the DT7H callsign in the same frequency.
Busted calls has always been an unavoidable feature of the RBN, which is a
superb infrastructure resource for all of us, but the % of busted calls were
relatively low and a manageable issue in my experience.
So, after the contest I'd used the raw data info to evaluate how bad was the
problem and the results seems higher than I would have expected.
Sunday numbers were the worse, the distribution of the problem is:
Error(%)
T%
0%
8%
20%
17%
40%
8%
60%
33%
80%
17%
100%
17%
This should be read as 8% of the hours of my Sunday participation (summarized by hour, or 1 hour) the RBN reported 0% busted calls, while 17% of the hours 100% of the reported were busted (this means during two hours of my participation I'd been reported as a busted call only). All in all 67% of the hours I participated on Sunday (8 hours) the busted reports of my call were 60% of the times or higher, about 2 every 3.
Saturday were a little bit at ease as about 1 in 3 times the reports were
busted with a couple of hours were the reports were a busted calls about 3 out
of 4 of the times.
I don't know if something special was going on with the RBN infrastructure
previous or during the WAE, or the conditions were bad, or perhaps there was
other factors.
Some fellow operators took the effort to post me correctly by hand after
working me, saying that what was reported was wrong (deeple appreciated their
kindness).
I wonder whether this has been noticed or experienced by other stations; in WAE
being spotted with an Italian call might not be much of a problem, but in WW
probably would be a bigger issue.
Thanks, Pedro LU7HZ
Dr. Pedro E. Colla
Va.Belgrano-Ciudad de Cordoba
Cordoba- Argentina
"Un objetivo sin un plan es solo un deseo" (Antoine de Saint-Exupéry)
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
|