LF: RE: WSPR S/N Tests

["Lee Hudson" <hudsonville@btinternet.com>]

Hi Andy, Jim, LF Group,

 

I did a similar test back in January and had similar results using a 20Hz bandwidth on WE2XGR/6 and managed 20dB or so improvement.

The WSPR spectrogram showed nicely the ripples in the filter used, and appeared to be a limiting factor as to how far you could go.

Also WSPR spent more time trying to recover something from these ripples, in one case nearly the full 2 mins.

 

Test was triggered also by too good to be true signal reports from some stations, as Andy has spotted.

Interestingly these stations would fail to report other stations that should clearly be readable given the S/N ratios reported.

 

With careful inspection of the reports it was sort of possible to get a feel for what bandwidth was actually being monitored. But not conclusive enough to be sure. A transmitting station that drifted through the WSPR band was very handy for highlighting this.

Looking at the spread of S/N reports you would also find a sudden cut off point way above -30dB norm were reports would not be given and this could be an indicator of the gain achievable by over filtering that should tie in with your findings.

Rather than spend hours analysing logs, and contemplating how my station could be improved further, it was much easier to discount these reports all together as being unreliable and of no use.

 

73,

Lee M0LMH.

 

-----Original Message-----
From: owner-rsgb_lf_group@blacksheep.org [mailto:owner-rsgb_lf_group@blacksheep.org] On Behalf Of Andy Talbot
Sent: 07 February 2010 18:21
To: rsgb_lf_group@blacksheep.org
Cc: jfell@tesco.net
Subject: LF: WSPR S/N Tests

To try to get to grips with the falsely high values of  S/N reported by some stations recently, I made a set of controlled measurements over a direct link using a range of Rx bandwidths and input signal levels.

 

The results were dramatic.  With receive bandwidths less than 300Hz, the reported S/N increased considerably.   It was only possible to use bandwidths in steps of 500Hz so 100,150,200 etc with the SDR-IQ but these were more than enough to show the effect.

 

Full details can be found in www.g4jnt.com/WSPR_SNR_Test.pdf  but the results can be summarised as :

 

-------------   Don't use a bandwidth below 300Hz unless you want artificially high SNR reports    -----------------------

 

300Hz or anything higher that allows the decoding software to see the full noise band is OK.

Andy
www.g4jnt.com

No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 9.0.733 / Virus Database: 271.1.1/2674 - Release Date: 02/07/10 19:35:00