[Skimmertalk] Monitoring Band Openings

Rick Walker rnwalker at bellsouth.net
Mon Jul 14 20:41:21 EDT 2008


Dave,

I have invested pretty heavily in skimmer technology and have been using it
extensively for several months.  I use six SDR-IQ receivers by RF Space in
my skimmer cluster not Softrocks.  I'm hosting the skimmer cluster software
on a server with dual quad-core 2.5 GHz Intel Xenon processors and a 5 TB
RAID disk farm.  Alex actually recommended the QS1R receiver, unfortunately,
support for that device was not available when I started my project.   I was
in a hurry to get an initial capability up and running to evaluate the
technology so I chose not to wait on support of the QS1R by Skimmer.

I have done some after-action analysis of the data I collected with my
skimmer cluster during the 2008 CQWW WPX.  In case anyone is interested,
here are the overall WPX results:

Recorded data:           430 GB in 48 hours
Total Spots:             67,885 in 34 hours
Unique Call Signs:       12,605 in 34 hours
Unique Call Signs/Band:  17,162 in 34 hours

					Unique Call Signs (Band-by-Band)
Totals								
10M	  	15M	  	20M	  	40M		80M
160M		 TOTAL
351		2,059		9,031		4,745		865
111		17,162

Note:  The CQWW WPX data were collected using my main yagis configured for
general geographical coverage and NO interference from local transmitting.
If you do the math, the average aggregate spot rate was just under 2000
spots per hour and on average my skimmer cluster found 370 unique callsigns
per hour.  Obviously Skimmer has an inherent callsign detection error rate,
but even if the error rate is 50% (it's actually more like 10-20%) that
still represents a significant sample set of potential, valid contest
contacts.

Based on the performance from my skimmer cluster that I observed during
Scott's IARU effort this past weekend, I think the spot numbers were much,
much smaller than in the WPX.  The cluster was using only a small DX
Engineering active RX vertical and Scott was, of course, transmitting the
whole time.  I didn't collect any statistics during this contest because I
didn't want to interfere in any way with Scott's effort.  However, my visual
observation of the spot rates from the skimmer cluster data stream indicated
to me that it was still producing at least as many spots as one generally
sees from a packet node during a major contest. 

Incidentally, I plan to add at least one QS1R into the configuration in the
very near future to run Alex's Faros application once support for the QS1R
receiver is available in Faros.  (I corresponded with Alex about adding
support for the SDR-IQ in Faros, and he indicated his plan for Faros was to
skip the SDR-IQ in favor of the QS1R).  I am interested in collecting
propagation data from the beacon network, which Faros does very well.

I read this prior to hitting the send button and realized my comments were
all over the spectrum and not confined to monitoring band openings...
Hopefully I didn't bore you too much with my ramblings...

73,

Rick
K4TD




More information about the Skimmertalk mailing list