Re: [amsat-bb] FM vs SSB

["John Stephensen" <kd6ozh@xxxxxxx>]

The articulation index (AI) depends on the SNR at the receiver and the audio
bandwidth. For FM, a 97% AI requires a 12 kHz IF bandwidth, 200-6000 Hz
audio bandwidth and 64 dB-Hz carrier-to-noise ratio. With a 24 kHz IF
bandwidth, 59 dB-Hz gives a 97% AI.

When the audio bandwidth is reduced to 200-3000 Hz the maximum AI is 75%. To
acheive this in a 6 kHz IF bandwidth requires 63 dB-Hz. For a 12 kHz IF
bandwidth 57 dB-Hz is required. You can see why hams that used the 32 kHz
wide IF 5000 kHz audio FM systems in the 1960s preferred them.

A 75% AI with SSB requires 60 dB-Hz average power or 67 dB-Hz PEP. All of
these numbers assume preemphasized and clipped speech.

60% AI requires much less power and may be a more realistic goal for a
satellite-based system. I've taken four graphs from the book for a 300-3000
Hz audio bandwidth and converted them to tables:

Average RF Power & Unprocessed Speech

        DSB     12kHz   6kHz
AI      SSB     FM      FM      AM
20%     33      44      43      49
30%     38      46      45      54
40%     42      49      50      59
50%     48      51      58      -
60%     52      56      -       -
70%     60      -       -       -

Average RF Power & Processed Speech

        DSB     12kHz   6kHz
AI      SSB     FM      FM      AM
20%     31      40      39      39
30%     35      42      42      43
40%     39      44      44      47
50%     43      45      45      51
60%     47      46      48      55
70%     52      51      53      60

These numbers may be off by a few dB as the vertical and horizontal scale
divisions were 20% and 10 dB. However you can see that a 3dB power increase
for 12 kHz wide FM increases AI from 30% to 40% for unprocessed speech. For
processed speech the increase is from 30% to 50%.

73,

John
KD6OZH

----- Original Message -----
From: "Bob Bruninga" <bruninga@usna.edu>
To: "John Stephensen" <kd6ozh@AMSAT.Org>
Cc: <amsat-bb@AMSAT.Org>
Sent: Thursday, July 31, 2003 03:05
Subject: Re: [amsat-bb] FM vs SSB


> On Fri, 25 Jul 2003, John Stephensen wrote:
>
> > ...This table compares the modulation types by the received signal level
> > required for an articulation index of 30%.
> >
> > Unprocessed     Processed       RF
> > PEP     Avg.    PEP    Avg.    PEP     BW     Mod.
> > 39      53      35      42       3      SSB
> > 46      46      42      42      12     FM
> >
> > The articulation index is the percentage of spoken words received
correctly.
> > The audio bandwidth in all cases is 300-3000 Hz.
>
> Ah, this is really weak signal stuff if only 30% of the words are
> understood at the receive end.  And of course FM is much worse at the weak
> signal part of the knee..  THus this presentation ONLY applies to digging
> out the weak ones over a very poor very noisey link...
>
> For serious two-way non-contest communicatinos link from GEO, I dont want
> to listen to weak-signal very noisy painful SSB.  There are many
> applications in the Amateur Satellite Service that could use such a clean
> signal as FM.
>
> In comparing SSB to FM, remember, that above the FM threshold affect, the
> dB improvement for FM is far greater than 1:1 where as for SSB, it always
> remains 1:1.  THus, I'd like to see the numbers after only a 3 dB increase
> in power for both.  My guess is that FM would improve 10 dB and SSB on ly
> 3.  FM in this case would have a much higer articulation index in this
> case...  So, is there a table in that book for an articulation level of
> 97%?  If we want to pass meaningful traffic via satellite (not just chase
> DX) then we need a better signal...
>
> Thanks
> Bob, WB4APR



----
Sent via amsat-bb@amsat.org. Opinions expressed are those of the author.
Not an AMSAT member? Join now to support the amateur satellite program!
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org