Data Scrubber, part 2

I am now testing the "wrong band reported" piece of the data scrubber. A quick check seems to indicate that it's correctly identifying all the cases reported to the forum over the last few weeks, and well as quite a few others.

Some quick statistics on the data filtering based on the last two weeks of data:

Bad calls (bogus decodes): 0.1% of all spots
Bad timestamp: 0.1%
Duplicate reports: 1.7%
Wrong band: 0.6%

So, about 2.5% of spots are being filtered. Duplicate reports include multiple reports of same station in same band by same reporter at same timestamp, so if you are gettting mutliple decodes of same station in the passband because of heterodyning in the rx or tx, this filter will keep only the strongest report.

Since the wrong band percentage is low (<1%), I'm inclined to just delete the incorrect band reports rather than try to fix them in place. The data is always there to do that later if it is somehow important.

Note for stations running simultaneous multiple tx or multiple rx in one band:

The information transmitted is callsign, grid, and power. If you are txing on multiple bands simultaneously (possibly in same time slot), please make at least one of those three things different. Easiest is to use different power if possible. If you are txing the same exact thing on multiple bands at the same time, it's hard to figure out what are correct band reports.

Likewise, if you are running multiple RX simultaneously in same band, please make either your reported call or your reported grid different between the two stations if you want both copies of the data to be kept.