Re: incorrect values of snr difference on wspr.live grafana page ? Possible bug-report


ON5KQ
 

Hi Arne,
excellent - I realized, that you are already working on implement a map with real SNR difference numbers!

However we must be careful, when calculating!
A special case, which makes problems:
- What if first receiver don't hear a station at all and the reference receiver's mean receive SNR of that station is -20dbm.
In that case there should be no difference calculated, because the 'difference' then is 100%....hi

However if we follow the calculation as it is implemented now, we get a lot of false reports - let's take the example of above
receiver 1 don't hear the station at all.....---> Does it mean this field is 0  (zero?!)
Then reference receiver hears this station with -20dbm  --->   0dbm minus (-20dBm) = +20dBm

So:  An orange point is shown and receiver 1 as if the receiver 1 would hear the station 20db better - instead it don't hear it at all !

We can indeed see lots of high numbers - when we compare totally different  receive locations, where the chance is high, that one station don't hear the same stations as the reference receiver... could you check it please ?!

But this map will be great, once you have found the time for some tuning.... (as with the case mentioned above)

Many thanks for the great work !

Ulli, ON5KQ

Join wsprdaemon@groups.io to automatically receive all group messages.