Re: LF: RE: [rsgb_lf_group] Re: LF: slow WSPR?

[Markus Vester <markusvester@aol.com>]

Hi Graham, LF,

 

wow such a flood of incoming mails...

 

> Well yes Joe (K) is right, BPSK is better, but needs a linear system to transmit .... Joe (EA) has stated, he could extract another 6 dB if the modulation system was changed ......

 

BPSK vs. ASK? Unfiltered PSK _can_ be sent by a nonlinear transmitter, it's only very unfriendly to others due to the spectral sidebands from the sharp transitions. But so is unshaped ASK!

 

In simple words, ASK steps from 1 to 0 wheras BPSK transitions go from +1 to -1. So with BPSK you get twice the sensitivity, along with twice the keyclicks. If you compare ASK at a given peak power to BPSK sent at -6 dB, you end up at same sensitivity and same clicks. Only average power for PSK would be half (25% instead of 50%).

 

Phase-continuous FSK as used in WSPR is much more gentle in that respect. There are no steps in the waveform, thus the click spectrum falls off much more rapidly.

 

> WLOF is already coded and makes use of multi pass to gain s/n, but is psk and needs a liner system ...and is not a one-pass decode system , when the s/n is low .but at -41 dB, by what ever scale, OP32 is well into the noise in single pass

 

The advantage of such a "multipass" system is that it can be adaptive to SNR, ie. a strong signal decodes fast, and only for a weak one you have to wait longer.

 

> We didn't set out to produce a low level beacon mode, it was supposed to be a replacement for the CW key ..

 

Apparently Opera is functioning as a beaconing system, and nothing else. You basically transmit one information ("I'm there"), and you get a reply by the internet ("I see you"). Am I missing something here?

 

Sooner or later, someone here will surely ask that question: On the other hand, if you are aiming for two way communication, and there is no SNR advantage, then why would you want to replace the Morse key in the first place?

> The Op structure allows for up to 50% loss of signal randomly along the time line, ie first 50%, last 50% or randomly distributed

 

Yes with Opera's distributed and redundant coding you can chop off half of the signal time. WSPR can do the very same stunt. Both will need more SNR during the remaining half, at least 3 dB, probably a bit more. Even DFCW could do it if you had sent two repetitions at double speed ;-)

 

> and over a real path , the Op system is able to make use of deep variations in fading and is immure to phase and Doppler distortion.

 

Yes of course, on the air there are other factors than "AWGN" white noise. Spherics and impulsive QRM have to be dealt by appropriate (preferably wide-band) noise blanking strategies. Fading and Doppler (which is only milliHz on LF) may have to be dealt with. But it remains to be proven that under these conditions Op is so much superior that it can make up for the 6 dB shortfall under lab conditions. 

 

> But, 'the eating of the pudding is in the proof of the making'? Stefan, last night reaching ua0aet over land, with 7 dB left in the system, taking some big bites out of the distance records on 136

 

Yes, a very nice result! I can state without envy that Stefan has a good signal, and it is going further than others. But does that really make a point for Opera, versus any other mode?

 

Graham, I'm in no way against Opera mode per se. But I have to say that I dislike the bragging.

 

> The best thing about the Op system is 'The number of reason's it cannot work' :)

 

Hey, that's what all those perpetuum mobile inventors keep claiming ;-)

Best 73,
Markus (DF6NM)

 
Re: LF: RE: [rsgb_lf_group] Re: LF: slow WSPR?
Von: Graham <g8fzk@g8fzk.fsnet.co.uk>
Datum: Mi, 12 Sept 2012 12:24 pm

 

Re: LF: Fwd: Re: Ideas for a slower WSPR for the 137 khz band
Sent: Wednesday, September 12, 2012 1:08 PM

 

Re: LF: RE: [rsgb_lf_group] Re: LF: slow WSPR?
Datum: Mi, 12 Sept 2012 4:16 pm

 

Re: LF: Fw: LOST TRACK
Datum: Mi, 12 Sept 2012 5:18 pm
...

-----Ursprüngliche Mitteilung----- 
Von: Graham <g8fzk@g8fzk.fsnet.co.uk>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Mi, 12 Sept 2012 4:16 pm
Betreff: Re: LF: RE: [rsgb_lf_group] Re: LF: slow WSPR?

Well  its mostly a silly  argument  as  the  systems  are totally  different  

, in terms of  data  processing  and the  operation of the  decoder  and the  

technical  level  of the  equipment needed to tx/rx  the mode 

 

We didn't  set out to produce a  low level  beacon  mode , it was supposed  

to  be a replacement  for the  CW key ..  the longer times  where as  a  

result of studying the  various    EU/VK  qrsss  plots  and  reasoning that  

over 50% of the  32  min cycle could  be  above the  decode level ... the  

rest  (will be) history :) 

 

DSP and  associated  'Numeric Processing'   facts and  fictions  are  very  

difficult to separate  , not  helped by the  ongoing  pie fight  out to  

'our'  west , however... so far  so good ! 

 

The  best  thing  about the  Op system is  'The  number of reason's  it  

cannot  work'  :) 

 

G..