|
It would be nice to start , with
the opening statement 'all things being
equal' but as 'we' all know that not
quite the case , one of the problems being , not
all stations can send all the modes and then not
all can translate from audio .. so even if
say WSPR or Ros MF or EME where to
be more efficient , not all can join in .
From a communicating efficiency
point of view, MFSK will offer a
higher efficiency compared to single tone,
but , as can be observed , LF operations tend to
take place in extended time scale so gain from time
integration has tended to be the path taken
..
QRSS Visual observation of
course is similar to the final wspr
'overlook' but with the ability to re-construct
what's perhaps more likely , but in contrast,
the amount of data sent via wspr is
limited and cannot be used as a qso mode , it
was developed as a beacon and as Jim notes , is capable
of decoding with only partial reception of
the transmitted 'pulse' I think there are 18 frames
?, the arguments continue as to beacon or data
mode ..
There is of course the
other mode ....
In on air tests , I
have observed the ROS-MF-1 system operating
some 2 to 3 db lower than the cut
off point of wspr , whilst running at 36 characters/minute, the MF modes (and
EME) modes use only an algorithmic approach to
data processing , there being -no- randomisation
deployed in the MF or EME modes , a simple
test , observing a beacon, shows an identical pattern
for each transmission.
The modulation being phase
continues mfsk avails itself to non liner systems,
one trick developed with Gary G4WGT , is
to double the drive frequency , then present to
the logic drive of a class e/d amp ,
resulting in the correct tone spacing , this
should work with WSPR , I think Gary has
tested this as well (Not required if
you are using a complimentary out put stage ! )
MF and the EME mode are true
data modes , unfortunately the link to the
MF- DDS 'project' has been removed in later versions , as
that did present a way of generating at LF,
perhaps with the advance of the SDR , the tx
side may follow .. ? as winter approaches ,its not
beyond possibility that with a pool of interest and a
couple of emails ... progress could be made !
73 -G..
Sent: Wednesday, August 24, 2011 6:06 PM
Subject: Re: LF: WSPR or QRSS: which is better?
I know what you're referring to. In cases of impulsive noise, there
is a finite probability of something eventually getting through
the decoder and being flagged as valid. The very nature of heavy source
coding, means that the resulting random output will look like a valid
callsign. However, WE then apply the next level of error detection,
by knowing the combination must be rubbish.
'jnt [and there is another example of source coding]
On 24 August 2011 17:47, Graham <[email protected]>
wrote:
>> and gives absolutely guaranteed error free decoding, or
nothing at all. <<
Are you sure ?
G..
Sent: Wednesday, August 24, 2011 5:12 PM
Subject: Re: LF: WSPR or QRSS: which is
better?
WSPR works in a 1.46Hz signal bandwidth and because of its very high
level of error correction and soft-decision decoding, means that it will
work at a S/N of about 3dB in this bandwidth, and sometimes a bit lower
still (Normally, FSK with no correction at all needs about 10 -
12dB S/N for near error-free performance)
QRSS has to show about 6 - 10dB in its signal bandwidth to be able
to discern fully what is sent, although a slightly lower S/N may be useable
when you 'know' what you should be receiving. (A form of forward error
correction is now in use here as well perhaps :-) So lets say 5dB
S/N is a working value..
So take 3dB in 1.46Hz as a starting point and derive the
bandwidth for QRSS needed to get 5dB S/N with the same signal. This will
have to be narrower to get a 2dB higher S/N and works out as 1.46 /
10^(2/10) = 0.92Hz
So QRSS used with a 0.9Hz bandwidth - which I think means about a 2 - 3s
dot period ought to be decoded at the same S/N as a WSPR signal.
Which is probably the info you wanted.
But now compare source coding efficiencies. WSPR fits a
callsign, locator and power level into a 110 second transmission - and gives
absolutely guaranteed error free decoding, or nothing at all. About
12 characters in actuality, but that is being a bit unfair as the coding
forces certain callsign and locator formatting. So in all
probablility, more like 7 or 8 effective characters (I'm being a bit empirical
here)
Assuming standard QRSS - not DFCW - , which if like standard Morse, then
5 characters takes about 50 dot symbols to send (12WPM = 60 chars in 1 minute,
= 1 char / second, or about 10 dot periods / second. Dot
speed = WPM / 1.2) If we have 2s dots, that is 5 characters can be
sent in the time for a WSPR transmission.
So as a quick estimate, WSPR wins by roughly 2dB in S/N terms for a
given dot period / noise bandwidth. And at similar S/N values, WSPR is
about 1.5 times faster
Andy
|
