Topband: Modeling the proverbial "vertical on a beach"

[paul_group at greenrover.demon.co.uk]

I don't believe it waits 45 seconds before reporting, Skimmer is posting 
spots within a few seconds of starting.  I think the reference was to an 
assessment of band noise up to and including the previous 45 seconds used 
in computation.

With QS1R here the measured and reported snr compares well with signals 
generated using autotims which is a setup used commercially for 
transmission imparement.

Of course band noise varies and its not and never was meant as an absolute 
measurement but I can say with my setup, increasing noise by 3db degrades 
the snr report when enough samples are taken. Also I can confirm that again 
statistics show reduction of both signal and noise does not change snr -  I 
validated this over some 10000 spot sample over many months and I'm 
confident enough in my setup that for amateur purposes its good.

Try degrading snr by introducing noise to the test signal and compare say 
100 samples and see if it shows anything interesting.

Alex is very competent and always the issues I have are due to my lack of 
understanding :-)

Regards Paul.

------- Original message -------
> From: Juan EA5RS <ea5rs at ono.com>
> To: paul_group at greenrover.demon.co.uk
> Cc: Topband at Contesting.com
> Sent: 17.8.'14,  11:46
>
> Thank you for the reference Paul.
>
> Regarding the quote by N4ZR stating:
> "I had earlier understood that SNR was calculated in a relatively short 
> interval just before the spot was sent to the Telnet server.  Turns out 
> that the calculation is based on data collected in the 50-Hz decoder 
> channel for about 45 seconds up to the moment when the spot is validated 
> and forwarded to the Telnet server"
>
> I just rechecked with a copy of Skimmer v1.8 and a recorded audio snippet 
> with my call and found that is not the case with my setup, I cannot speak 
> for others.
> With a clean Skimmer program start, it sent out a spot after two 
> instances of my callsign sent at 34 wpm, that is after less than three 
> seconds of signal sampling.
> I repeated this same experiment five times, same behavior every time. 
> Reported SNRs varied from 19 dB to 22 dB. The input to the soundcard is 
> the same recorded audio signal.
>
> 73
>
> -----Mensaje original-----
> De: paul_group at greenrover.demon.co.uk 
> [mailto:paul_group at greenrover.demon.co.uk]
> Enviado el: domingo, 17 de agosto de 2014 9:10
> Para: n7rt at cox.net
> CC: ea5rs at ono.com; hsvdds at juno.com; Topband at Contesting.com; w8ji at w8ji.com
> Asunto: Re: Topband: Modeling the proverbial "vertical on a beach"
>
>
>
> ------- Original message -------
>> From: Hardy Landskov <n7rt at cox.net>
>> To: ea5rs at ono.com, hsvdds at juno.com
>> Cc: Topband at Contesting.com, w8ji at w8ji.com
>> Sent: 17.8.'14,  3:47
>>
>> Juan,
>> Am I wrong to assume that skimmers are not calibrated? They should be
>> calibrated to S-9=50 uV into 50 ohms at least to provide some kind of
>> uniformity band-to-band and skimmer-to-kimmer. I don't know you guys
>> tell me what's going on because I see these reports scrolling across
>> my screen that make no sense......
>> Hardy N7RT
>>
>
> Please take time to read the following page to assist your understanding 
> of the RBN snr report
>
> http://reversebeacon.blogspot.co.uk/2014_03_01_archive.html
>
>
> 73 de Paul LA/GW8IZR
>
>
>> ----- Original Message ---
>