Hi all,

I've been asked a number of questions about fundamentals concerning
the new "Fast JT9" submodes in WSJT-X.  This seems like a good time to
review the basics of *all* WSJT modes, including their particular
strengths, limitations, optimum propagation types, etc.

A useful comparison among all the WSJT modes is presented in a graph
posted here:
http://physics.princeton.edu/pulsar/K1JT/wsjt_modes.pdf

The graph shows the bandwidths of signals in each WSJT mode, plotted
as a function of message duration -- the time required to transmit the
message once at the keying rate used for that mode.

The slow modes fall on the right side of the figure.  The
transmissions last for about 48 s (JT4, JT9, and JT65) or 110 s
(WSPR), and message information (including encoded redundancy for
error correction) is sent just once per transmission.  These modes are
optimized for propagation types that yield more-or-less steady signals
over a full transmission.

The fast modes (toward upper left of the figure) use faster keying
rates.  Message durations are much less than the T/R sequence length,
so information is repeated many times in a single transmission.  These
modes are designed to take advantage of short signal enhancements such
as meteor pings or scatter maxima.

JT4, JT9, and JT65 use structured messages compressed into exactly 72
bits of user information.  In contrast, JTMS, FSK441, ISCAT, and JT6M
permit arbitrary message lengths up to 28 characters.  For comparison
purposes in the plotted figure, message lengths were taken as 19
characters: two 6-character callsigns, a four-character grid or
report, and the required spaces separating them.

Fast keying requires relatively wide bandwidth.  Slow modes can get by
with smaller bandwidths or can use many more tones, as in JT65, which
brings other advantages.  Slow keying with wide tone spacing is a good
choice for propagation types such as 10 GHz EME, where inevitable
Doppler spreading renders signals much wider than their natural
modulation-induced widths.

The WSJT modes with structured messages and strong forward error
correction have become very popular for a variety of applications
including EME at VHF and above and weak-signal DXing on the MF and HF
bands.  The fast JT9 modes seem likely to become popular for scatter
propagation; JT9H works fairly well also for meteor scatter on 6
meters.  It would be nice if the JT9H message duration (0.425 s) could
be made smaller, for meteor-scatter use.  Unfortunately, this would
require more bandwidth, which is impossible with traditional ham SSB
transceivers.  For this reason JTMS and FSK441 remain the best choices
for meteor scatter at 144 MHz and above, and probably also at 50 MHz
at power levels 100 W and less.

It would be nice to have a mode with SSB-compatible bandwidth, message
duration less than 0.2 s, and strong forward error correction.  A
(presently hypothetical) mode labeled JTMS-FEC in the figure
represents a possible way to address this need.  Experiments with
such a mode are planned for the near future.

      -- 73, Joe, K1JT

------------------------------------------------------------------------------
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Reply via email to