Hello Joe, all, I understand that there are better ways to achieve a lower S/N decoding than going back to JT9-2. But awaiting this I would anyway want to do some tests with longer JT9 transmission on 630m. Since I am afraid that it is out of my league to dive in to the WSJT-X code and modify it I thought of an alternative route: Generate a "normal" JT9 signal, but transmit it at half speed. This transmission will take 98 seconds and be at half the original audio frequency (also half bandwidth). At the RX side the audio is recorded as a WAV file, speed is doubled again (just by changing the sample rate in the WAV file header) and this file is fed to WSJT-X. I know it sounds is a a bit of an odd thing to do, but at least feasible for me. Would this work (and improve SNR), or am I just trying to be too smart?
73, Rik ON7YD - OR7T ________________________________________ Van: Joe Taylor <[email protected]> Verzonden: donderdag 6 september 2018 16:51 Aan: WSJT software development Onderwerp: Re: [wsjt-devel] JT9-2 mode? Hi Rik, We are well aware that propagation-induced signal distortions at LF and MF are very different (and generally much smaller) than at HF. The extremely narrow bandwidths used by discontinued sub-modes JT9-2, JT9-5, JT9-10, and JT9-30 took advantage of these differences and provided sensitivities not otherwise achievable. However, the data rate of these sub-modes was proportionally smaller, as well. A minimal standard QSO takes 12 minutes in JT2 and 3 hours in JT9-30. As a result none of these submodes ever gained any significant use. We now consider JT9 to be essentially a "legacy mode", and we have no plans to bring back JT9-2, etc. In the meantime we have developed better ways of achieving lower S/N decoding thresholds, with new coding and modulation schemes. At present these ideas and tentative plans are on the "back burner", but we may return to them in due course. Of course, all of our past code can be recovered from the open-source archive. With dedication, someone with enough time and interest could resurrect what's needed to make a working platform for JT9-2, JT9-5, etc. I do not recommend this approach, because better schemes are possible. -- 73, Joe, K1JT On 9/6/2018 2:59 AM, Rik Strobbe wrote: > ​Hello, > > I hope I am at the right place with my question (if not so, please let > me know where I need to be). > > I am using JT9 on the 630 meter band (472-479kHz), it is an excellent > mode to work distant stations. But even with JT9 is is very difficult to > cover distances of over 5000km. > > At these wavelength it is a hard job to get some signal radiated > (antennas are very small compared to then wavelength), so most stations > have an ERP of less than 1W. > > In addition, depending on the countries, there is a ERP limitation > between 1 and 5 W EIRP. So increasing the radiated signal is often not > an option. > > In an early version of WSJT-X several "flavours" of JT9 were available: > JT9-1 (1 min sequence), JT9-2 (2 minutes sequence) and JT9-5 (5 minutes > sequence). > > JT9-2 would give an extra 3dB (more or less), very often the difference > between success and failure on 630m DX. > > I believe JT9-2 and JT9-5 were abandoned because there were issues with > the doppler shift and multi-path propagation on HF, but this is less > critical on lower frequencies and on 630m at least JT9-2 was very useful. > > Question 1: Is the early version of WSJT-X (that supports JT9-2 and > JT9-5) still available somewhere? > > Question 2: Would it be feasible to implement JT9-2 and eventually JT9-5 > in a new version of WSJT-X? This would also take advantage of > the decoding improvements made the last years. > > 73, > > Rik, ON7YD - OR7T ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ wsjt-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/wsjt-devel _______________________________________________ wsjt-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/wsjt-devel
