|To:||[email protected], [email protected]|
|Subject:||LF: Dual Mode WSPR Beacon now running|
|From:||Andy Talbot <[email protected]>|
|Date:||Mon, 1 Jun 2015 17:56:34 +0100|
|Dkim-signature:||v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=0ONBG2DvDPSplw9KEh4JUPTNa9wiLQ3AqB2/aAe6GnI=; b=i5lUi+TsLgrHx6tTCh6BWG9HWtXCEqHHkWcuC3nZUlFzgTWVsgm9PU37BHQXEXCpQ6 NANEaSN4YMq29WtSs7JfLgEjWnWqkt4Ud/rjI8hczXnCz7HdXQTQsQGiEwllPxpvSiqP N6H3h/6DMLe04VrzisSmXxMxZYRRr2V8YZVPBsw/Px39mvMprko9djibZAT8QGXH9bjN jQ69NMHSxvFb54xq43TXm9wtHd1cYPIcMMur5/Ow5tVjS/A+eTTvg7fxBZ7j4xidvpam 1R0JJEEnFruHqrAXiQ+rcEbn6s5OaGKUkgGidhYf9sJGpTaCU13ZsqGgamIAWOel+0N5 pe2Q==|
I now have a dual mode WSPR beacon running on air. Frequencies are:
Whilst setting it up I had a eureka moment about how to separate out the reports from each on WSPRNet. Having had two callsigns for a long time - for expressly this sort of purpose (and because I can !) the WSPR-15 transmission has been coded with G8IMR whereas the WSPR-2 one carries G4JNT Loc. and Power the same in each message IO90 and 33dBm EIRP respectively. (Might as well make to most of having two callsigns in case the rumours are true and Ofcom may force people to have just then one callsign.)
The transmission pattern is quite convoluted, making use of a pre-stored pseudo random (PR) sequence with two counters pointing into it to determine if a slot will contain a transmission or not. One pointer is incremented at the two minute interval, and counts modulo 29. The other counts at 15 minute intervals modulo 17. However, each count is interrupted if a transmission is underway, so slots will get delayed.
Having two relatively prime intervals (2 and 15 minutes) and prime numbers used for the modulo means the process is almost chaotic (using the term in its mathematical sense) , with the actual sequence generated being determined by the start time, the two modulo counts, and the PR sequence. The same PR sequence is used for both modes, but even they could be made different if deemed necessary
I tested it all out today, using 30 long counts rather than 29 and 17, together with a slightly different PR sequence, and it seemed to give an acceptable spread of transmission slots and types, but with some long gaps. The modified counts and PR sequence should cure that - it'll certainly randomise things!
Hope it all works out on air.
(ps. Completely OT to LFgroups but of general interest, I'm sure. The technique of using two or more PR sequences, with the output of one modifying or delaying the clock for the other(s) forms the basis of encryption used on the GSM mobile phone network, the Keyloq protocol for car key fobs, at least two of the older military secure phones and no doubt many other crypto systems too).
|<Prev in Thread]||Current Thread||[Next in Thread>|
|Previous by Date:||Re: LF: Re: Top loaded vertical spiral antenne, DK7FC|
|Next by Date:||LF: Re: WSPR-X times 2 MF, G3WCB|
|Previous by Thread:||LF: WSPR-X times 2 MF, John Bruce McCreath|
|Next by Thread:||Re: LF: Antennas on low bands : the Siegel and Labus books., riese-k3djc|
|Indexes:||[Date] [Thread] [Top] [All Lists]|