On Dec 26, 2013, at 1:27 PM, Kai wrote:
THAT is why we need a limit! Ham radios, especially SDR units, will be able to
handle higher BWs - let's get the limit in place NOW.
Kai,
I think you are preaching to the choir.
I don't think anyone on this reflector is opposed to a bandwidth limit. The
only disagreement is *what* that limit should be. We hear anything from 500 Hz
to 6 kHz. So, pick your poison :-).
By the way, there is one bandwidth "limit" when you design HF modems -- and
that is the HF channel simulator. When Watterson et al wrote their seminal paper on how
the ionosphere behaves, they mentioned that the model is only accurate to 12 kHz and that
is only under quiet conditions. Under poor conditions, the model is limited to a
narrower bandwidth.
That does not mean that you cannot design HF modems for 20 kHz bandwidths, but
just that you need some other way to evaluate the modem than by using the ITU
published standard for HF modem testing.
One thing that you will notice from the SCS links that I posted earlier is that
SCS uses a 4 kHz equivalent noise bandwidth (they are probably using a
commercial HF channel simulator).
Because of Johan Forrer KC7WW back in the early 1990s, hams have been using 3
kHz equivalent noise bandwidths and that is what PathSim defaults to.
cocoaPath allows all industry standards (1.24 kHz, 3 kHz and 4 kHz). So, be
sure to apply a 1.25 dB adjustment when you compare SNR numbers from SCS (4
kHz) with the amateur measurements (3 kHz).
Speaking of the Watterson ionospheric model, C. C. Watterson left us in April
of this year:
http://obits.dignitymemorial.com/dignity-memorial/obituary.aspx?n=Clark-Watterson&lc=2558&pid=164435418&mid=5509056
73
Chen, W7AY