Message-ID: <D93051AC437B4E1AB0A34D366CA964BA@AGB>
From: "Graham" <g8fzk@g8fzk.fsnet.co.uk>
To: <rsgb_lf_group@blacksheep.org>
References: <CAK59VFNOneVPmCEjVc1kK3FpvfmCX5qp_gZo5-AFy58Hmiq-6g@mail.gmail.com> <500FF064.1030109@iup.uni-heidelberg.de> <DEF0879101734958AE61D0D590E070DF@AGB> <500FFDDD.6070204@iup.uni-heidelberg.de> <16CA55FCF4C44B789AA114D2CCB51D1B@AGB> <004f01cd6a8c$7566a7c0$0401a8c0@xphd97xgq27nyf>
In-Reply-To: <004f01cd6a8c$7566a7c0$0401a8c0@xphd97xgq27nyf>
Date: Wed, 25 Jul 2012 23:57:17 +0100
MIME-Version: 1.0
Importance: Normal
Subject: Re: LF: 4X OP32 report
Content-Type: text/plain;
	format=flowed;
	charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


The thing  is  Mal , that  its  CW but  not as  'you  know  it  , to  miss 
quote   CapJTK  , Yes the  slow  qrss  'did'   paint  CW on the  screen  or 
paper  tape  , but  with the  narrow  frequency   shift  keying  , with 
equal  mark  / space  ratio , with a  vari-code  ie  Morse , which  needs to 
be  interpreted   by the   operator , its  neither  CW or  Data transmission

Seeing  traces  , is the  $64,000  question  , yes  we can see  , and yes we 
can  Guess , but  if you where a  RO working  on BP tankers , with 
everything  sent  in 5  figure  code  groups , or  other  users  of this 
system, 1 wrong  letter and  you  could  loose the whole   message ..  which 
takes  the  story full  circle .. unless  there  is a  full  trace,  q5 copy 
or  a  decode  , the message  has not  been  successfully   transmitted

With  time  hopping  , Ala FEC , data  modes are  able to  decode , after 
loosing a  substantial  % of the  signal  , upto 50% in the  case of  Opera 
, the  fec  and  data  coding  are designed to  provide maximum  efficiency 
in qrm    , static  impulses  are  met  by a non linear routine, that  seeks 
to  reduce the  dynamic  range  , also  the  noise   impulse  /   Tx   'ON' 
time ratio  is  very  low , adding  a  second layer of  'limiting' , ( there 
is  no  need  to  pre-process the  audio  signal)

73-G..
--------------------------------------------------
From: "mal hamilton" <g3kevmal@talktalk.net>
Sent: Wednesday, July 25, 2012 6:39 PM
To: <rsgb_lf_group@blacksheep.org>
Subject: Re: LF: 4X OP32 report

> Graham es Co
> While Opera es other data modes are an alternative to CW for a change they
> need another level of demodulation and cannot really compete with the 
> basic
> CW mode or QRSS CW.
> I often see traces of all data modes that could easily be identified in 
> QRSS
> but they do not decode therefore the purpose of the transmission is lost.
> de mal/g3kev
>
> ----- Original Message -----
> From: "Graham" <g8fzk@g8fzk.fsnet.co.uk>
> To: <rsgb_lf_group@blacksheep.org>
> Sent: Wednesday, July 25, 2012 4:07 PM
> Subject: Re: LF: 4X OP32 report
>
>
>> schrieb   Stefan, :)
>>
>> Why do you think so?
>>
>> Its linked to the  DSP  engine  in Op , there is a  timing  issue  , if
>> other  applications  lock the  sound  card at a  different   rate , due 
>> to
>> the  % tolerances  , low level  decodes  can be  lost ,   noting, Op is a
>> 'data  mode  , so the  overheads in the  transmission  are  much  higher 
>> ,
>> as are the  DSP  engine  requirements , as can be seen  with  'older' 
>> pc's
>> cpu loading  can be a  issue , as  is a  one-pass  system , all  the
>> processing  is  taking place in real  time
>>
>> But  on with the  show ,
>>
>> 73 -G..
>>
>> -76 dB  in 28  mHz  b/w    , what  is  that  when  translated  to  2.5Khz
>> B/w ?
>>
>>
>>
>>
>> --------------------------------------------------
>> From: "Stefan Schäfer" <Stefan.Schaefer@iup.uni-heidelberg.de>
>> Sent: Wednesday, July 25, 2012 3:08 PM
>> To: <rsgb_lf_group@blacksheep.org>
>> Subject: Re: LF: 4X OP32 report
>>
>> >
>> > Am 25.07.2012 15:51, schrieb Graham:
>> >>
>> >> Pity  nothing decoded  in the  US last night , but  signal 's in 1500
>> >> /2000 mile  range where  quite  substantial ...
>> >> from this , it seems  that propagation  on 136  is  really  linked to
> the
>> >> seasons ?
>> >
>> > Yes (of course) and no. W1VD received me quite well a few weeks ago, 
>> > see
>> > https://dl.dropbox.com/u/19882028/LF/07_July2012.jpg
>> > Not to forget W1TAG:
>> > https://dl.dropbox.com/u/19882028/LF/DK7FC_Juli2012.jpg
>> >
>> > So there is some propagation in summer but of course it depends on the
>> > solar activity. Don't know how the condx have been yesterday.
>> >
>> >> One  point  from previous  trials , If  Spec-lab  is  used  with 
>> >> Opera
>> >> , then  SL must  be  set to  48Khz  sample  rate , or  OP can loose
> low
>> >> level decode sensitivity.
>> > Why do you think so? As long as the frequency range of a SSB bandwidth
> at
>> > audio levels passes through the SL filter, then it should be fine. So 8
>> > kHz SR should be fine. Of course it depends on which shift frequency 
>> > you
>> > have to use. I have to use 11 kHz shift and so i choosed a SR of 30 
>> > kHz.
>> >
>> >
>> > 73, Stefan
>> >
>>
>
>