[aprssig] APRS Speed Spec? (plan B)

[Robert Bruninga]

> This is not really needed...

The current speed field is limited.  To stay relevant, APRS needs a speed
field that can allow for all speeds on or around earth.

> and it's yet another special case kludge which doesn't work too well....
> Hessu

Putting in a single knee point is a common way to preserve scales that have
a range of very common and frequently used values while allowing expansion
over another magnitude of range that are rarely if ever used.  And the speed
of sound, Mach 1 is a perefect such inflection point.  We should not leave
it as is where there is no way to present these speeds.

And it is trivial to implement as a simple breakpoint at Mach 1.  Takes two
lines of code to decode:

If speed S is Less than or equal to 670 then SPEED = S in Kts
If speed S is greater than 670 then SPEED is 135*S -90000  in Kts

It preserves the better than 0.2% accuracy up to Mach 1 and gives about 1%
resolution at orbital velocity.  We could have used a power-law formula to
preserve accuracy, but the simple linear equation with one multiplication
seemed so much easier.

Here are the examples:

779 = 15,200 knots space station
682 =     2100 knots Mach 3
678 =     1600 knots military jets
677 =     1340 knots Mach 2
675 =     1200 knots Concorde commercial type jets
670 =       670 knots Mach 1

No one is required to implement this, and no one has used any speeds of that
ilk to date anyway.  But this gives us a way forward that does not break
anything, yet allows those that might have the need to at least have a way
to transmit it.

Bob, WB4APR