Locked Re: Modifying Preferred Frequencies on WSPR for Band Hopping?

Bill Somerville

On 13/07/2020 03:21, Ted Wood wrote:
I get that WSPR is mostly popular on HF frequencies, however, myself and some others in my immediate area are using WSPR to track and try to predict openings on 6m/2m/70cm around the SE US. I was reading through the documentation and noticed this excerpt:

Band-switching occurs after each 2-minute interval. Preferred bands are identified with time slots in a repeating 20-minute cycle, according to the following table


If the preferred band is not active according to your band-hopping schedule, a band will be selected at random from among the active bands.

Now, it's possible (likely?) that I'm misreading this but it sounds like if I have all of the yellow, preferred, bands selected on a given time window, there will not be any time window in which a non-preferred band will be selected during band hopping. Hopefully this is not the case the case and someone can clarify my understanding of how band hopping works.

On the off-chance I am interpreting the documentation correctly, how would one correctly run all HF bands + 6m/2m/70cm with band hopping? It seems to me like it's not possible. Utilizing the second statement in the docs "if a preferred band is not active..." by deselecting 160m and substituting 6m seems like a viable option. In practice, however, I've observed that 6m is very infrequently selected in place on 160m. Would it be possible to modify which frequencies are preferred or more fairly balance the time spent on the "non-preferred" frequency?

Hi Ted,

you are correct. It should be obvious that all the slots for the band hopping schedule are full so no changes are possible. If you want to add other bands then you must disable some of the scheduled bands from your schedule. As there are no set times for the bands outside of the globally scheduled ones the application uses a random selection criteria. That should be effective but clearly not with the "herd gathering" benefit of the globally scheduled bands.

The issue with your substitution proposal is that for it to work every station using the substitution(s) would have to do exactly the same substitution(s) to be coordinated in synchronicity. I think the randomized solution is the best we can offer. I suggest you use a higher Tx percentage than you might think necessary to make it a little more likely that you transmit on an uncoordinated band. This of course has a disadvantage in that you will be transmitting more often on the coordinated bands but does increase the probability that another station substituting one of the same non-coordinated bands will see you transmissions.

I would say that for best performance for non-coordinated bands you would be served best by not having *any* coordinated bands enabled and let the scheduling pseudo-randomized algorithm pick you selected bands without bias.


Join main@WSJTX.groups.io to automatically receive all group messages.