At FDIM George was talking about Coherent CW. Now, my views on CW have been made quite clear, but I am going to deliberately avoid snide comments about how Coherent CW might be a contradiction in terms. Thank you. I think this is something that Pete N6QW and I can agree on.
Nevertheless, this is all technically interesting. This reminds me of what happens with WSPR. But I wasn't quite sure what George was getting at with his discussion of the phase of the incoming signal -- I can see the need to get the receivers' narrow filter exactly on the transmitting station's frequency, but why the phase?
Here is Bob Crane's interview with George:
http://soldersmoke.com/N2APB23.mp3
And here are a couple of things with background info on CCW.
https://www.sigidwiki.com/wiki/Coherent_CW
http://www.arrl.org/files/file/Technology/tis/info/pdf/7509026.pdf
Thanks George, thanks Pete, and thanks Bob.
This is really cool, also interpreted BASIC instead of Arduino C ? Sounds GOOD !
ReplyDeleteChuck, WB9KZY
Interpreted BASIC is indeed really cool, and it's come a long ways since the days when we used to play with it on the IBM360 and even the Commodore 128. Besides having fast cores of the Pi to work on, the MMBasic version being used has numerous built-in libraries that are super convenient for today's homebrewers ... and even FFTs, windowing and other DSP functions. Now having said all this about BASIC, I will be taking the CCW project over to a Teensy 4.0 processor (and using C++) for reasons of advanced processing speed and power ... but will still have the Teensy plug into the K1SWL PMK baseboard so the CCW shields being developed will still fit atop. :)
ReplyDeleteThanks, the CCW project sounds really promising. With so many of the original CCW articles it felt like being handed a block of wood and a knife and being told there is a boat in there somewhere, carve it out for yourself :)
DeleteChuck, WB9KZY
To help explain the need for phasing as one of the three factors comprising "coherency", consider receiving a perfect 10 Hz square wave from a sender (ie, a 12 wpm dit stream) and the Rx processor is expecting to decode that signal by comparing against its perfect 100ms time windows (ie, also 10 Hz). Even though the Tx and Rx are sync'd to GPS (ie, the 100ms windows are "very close") there still will be some delays that shift the incoming signal with respect to the reference window. The Rx side would want to maximize the overlap of its window with that of the incoming bit stream to ensure that the FFT filtering is done across *all* of that signal window, not just part of it. Lots of words here, but if one is able to shift that 10 Hz Rx timing frame to best match that of the 10 Hz incoming signal, a tighter filter and a higher SNR can be achieved.
ReplyDeleteGeorge, N2APB
ReplyDeleteGreetings! Yes, CCW is an interesting mode for the experimenter, especially one who has a decently stable transmitter for starters. The symbol timing is important, but I think one item that gets overlooked is the "inter-symbol phase stability", for lack of a better term. The transmitter keying should not disrupt the phase in between symbols.
I did a bit of CCW work back around 2000~2006, wrote a Labview Program "CWLabxx" for Windows. It supported both CW and CCW. QSO'd with G3IRM , Peter Lumb in BurySt.Edmunds on 20M, and still have the screenshot of Peter's signal on CWLAB02.
I like the idea of running CCW on smaller more portable devices- ------
Could this be CCW's third calling?
First: on Hardware-Based Boat Anchors
Second: on PC's and Laptops
This time: Pi Pico, Teensy's,what's next? Wearables?
73!WN2A
I tried downloading the Coherent CW software for MS-DOS, to run under FREE-DOS but the code I downloaded seemed to be infected with something unpleasant. I would like to try again if I could find the right software.
ReplyDeleteG7VFY
Hi Stephen,
ReplyDeleteThe CCW program posted on the web page (part of the PMK code "PMK_v2.0_(24-Mar-2023).txt"), is a text file and is virus-free - no infections. I don't know about FREE-DOS but perhaps you could try running it with the MMBASIC interpreter program for Windows/DOS ... https://geoffg.net/WindowsMMBasic.html.
Let me know if you still have any problems.
73, George N2APB