[CQ-Contest] Understanding the dynamic error free RBN algorithm

[Pete Smith N4ZR]

Bret, I've asked you twice now please to share your data and methods, 
and you haven't even responded to my e-mails.

Sharing information is essential to any solution. The RBN team has 
developed a beta version of the Aggregator that filters out in-band 
image spots.  Several stations using Softrock receiversran the beta in 
the IARU HF contest and sent their spot logs to us for evaluation.  Our 
aggregator guy is working on filter refinements for the next beta.

Meanwhile, CT1BOH did his excellent work, and shared the data and 
methods with anyone who wanted it.  One week later, there are two 
Telnet-server-level implementations being tested in the real world, as 
well as another client-level approach in development that has great 
promise.

So much for ignoring problems.

73, Pete N4ZR
Check out the Reverse Beacon Network at
http://reversebeacon.net,
blog at reversebeacon.blogspot.com.
For spots, please go to your favorite
ARC V6 or VE7CC DX cluster node.

On 7/21/2013 5:28 AM, Brett Graham wrote:
> S56A4ZR said:
>
>> Real life 0IF RX could achieve 60 dB unavoidable image rejection and 
>> this is not enough in contests. Our ham freqs are NOT channelized 
>> although one knows where the image is!
>
> The most popular version of QSD receiver has EVEN HIGHER response at 
> harmonics of its LO than -60 dBc
>
> This means those receivers hear signals FROM OTHER BANDS something 
> around 30 dB LOUDER than the in-band I/Q gain/phase unbalance-caused 
> spurious responses of signals from the band the receiver is SUPPOSEDLY 
> listening to (that S56A4ZR goes on about).
>
> Let's ignore all the mechanisms responsible for wrong-frequency RBN 
> spots, shall we?
>
>> Direct HF sampling ADC RX like QS1R or Perseus with 16 bits has 
>> dynamic range over 100 dB with a single, clean XO and no inherent 
>> images.  This is several orders of magnitude better then single bit QSD.
>
> Too bad that in practice, skimmers using QS1Rs also say signals are on 
> all sorts of frequencies where they are not.
>
> Instead of several orders of magnitude better - I suggest, like I did 
> that the wrong-freq RBN spot rate was some number of %... denied 
> vigorously by N4ZR & now essentially confirmed by CT1BOH who rather 
> than just saying things, looked at & worked with the data - it's 
> roughly as bad as the QSDs.
>
>> We are dealing with low error rates and further processing would 
>> reduce insertion mistakes while increasing omissions.  I like to 
>> catch erronious ones - HI.
>
> Human now making excuses for failure of machine to do better than 
> humans are capable of...
>
> -ex-VR2BG/p.
>
> _______________________________________________
> CQ-Contest mailing list
> CQ-Contest at contesting.com
> http://lists.contesting.com/mailman/listinfo/cq-contest
>