[RTTY] RTTY and Skimmer was: Re: Decoder performance on crowded bands

[Lee Sawkins]

At risk of being shot at for self promotion I would like to say that although clusters running CC Cluster software are not using the CT1BOH algorithm , in my opinion what CC Clusters use is better. They are using the Levenshtein algorithm, which is at the heart of the CT1BOH routine. CC Cluster nodes do not give users a quality number for spots, they just drop what I consider to be bad spots. When I am contesting I just want to see good spots. I think this is what 99% of the users want. 

CC Cluster software also detects whether calls are valid for certain countries and drops those which fail. These calls are labelled "not licensed". For US calls, the latest FCC data is used. Calls like NS0RCWT and W1AYD, which are both "busts", are "not licensed". For JA calls, those like "JK1AA" are dropped as the only 2 letter Japanese calls beginning with "J" can be for JAx prefixes or JR6 prefixes. Calls ending with "/1J" are dropped as the country is invalid. Duplicate spots from different networks are treated as duplicates and dropped. 

Although individual Skimmers may only report spots every 10 minutes, during busy contests, Skimmers all get out of sync with each other and almost continuous spotting occurs for stations with big signals such 9A1A. Since N1MM software labels spots within the last 3 minutes as new, CC Cluster software spots the same calls after 3 minutes to keep the "new" tag next to them in the band map. That way you know which ones are more likely to still be there. 

Lee 

----- Original Message -----

From: "Tom Osborne" <w7why at frontier.com> 
To: rtty at contesting.com 
Sent: Thursday, October 1, 2015 5:05:35 PM 
Subject: Re: [RTTY] RTTY and Skimmer was: Re: Decoder performance on crowded bands 

One thing I wondered about was the 'not licensed' stations being spotted. 
They are not legal to work as hams are not allowed to talk to someone who 
is not a ham (other than on a station controlled legally by someone). You 
would think these would just not be posted. 73 
Tom W7WHY 

On Thu, Oct 1, 2015 at 9:25 AM, Michael Adams <mda at n1en.org> wrote: 

> Today, skimmers report once every 10 minutes while a station is detected 
> running. 
> 
> I wonder, if that logic were tweaked to adjust that window to report 
> "every 10 minutes or after another station has been spotted on that same 
> frequency, whichever comes first", could the CT1BOH skimbusted/skimverify 
> logic be tweaked somehow to warn about possible S&P callers. 
> 
> Similarly...does there need to be an adjustment in the skimbusted logic to 
> allow consideration of garble tables for RTTY spots? 
> 
> Finally, elsewhere in this thread, I believe I remember seeing that 
> another node operator was getting feeds from both the RBN and from DL4RCK's 
> system. I have been doing that on my node, but I am discontinuing the 
> DL4RCK feed after my CQWW RTTY experience. On AR v6, duplicate spots from 
> the different networks are being parsed as non-dupes, which impacts the 
> CT1BOH logic. 
> 

RTTY mailing list 
RTTY at contesting.com 
http://lists.contesting.com/mailman/listinfo/rtty