Hi Phil, Sure help with the waveform documentation would be great. We can get you a password for the wiki and the word processor doc source if you're keen.
If you want a challenging modem project I'm looking for help on the cohpsk modem, to support the 700C FreeDV mode I am developing. Much more fun than re-implementing an existing waveform. Cheers, David On 23/08/16 06:06, Phil Frost wrote: > Thanks David, that explains so much. I was really scratching my head for > a while trying to figure out why it wouldn't sync :) > > If I wanted to update the description after I hash out the details and > get my implementation working, what would be the best way to do that? > > On Wed, Aug 17, 2016 at 8:25 PM David Rowe <[email protected] > <mailto:[email protected]>> wrote: > > Hello Phil, > > Good question, and actually took me a while to remember why the two sync > carrier lines are just 25 Hz apart. > > FreeDV 1600 uses a symbol rate of Rs=50 symbols/s for each of the 17 > carriers, including the central DBPSK sync carrier. Note that we are > using Differential PSK (DPSK) - not PSK. > > The start of a 40ms codec frame is indicated by a change in phase in the > DPSK sync carrier, so for example this sequence of phases (in radians) > every 20ms: > > ... pi pi -pi -pi pi pi -pi -pi ..... > > Note the _change_ in phase every 2 symbols (40ms), indicating codec > frame sync. > > The cosine of the sequence of phases above is ....1 1 -1 -1 1 1 -1 -1 > .... which (when up sampled and filtered) approximates a sampled sine > wave of frequency Rs/4 (takes 4 symbols to repeat). So using this Rs/4 > frequency sequence to modulate the center carrier at frequency fc we get > > s(t) = cos(2*pi*t*fc)*cos(2*pi*t*Rs/4) > s(t) = 0.5*cos(2*pi*t*(fc+Rs/4) + 0.5*cos(2*pi*t(fc-Rs/4) > > which is two sine waves Rs/2 = 25Hz apart. > > Cheers, > > David > > > > > The BPSK sync carrier sends alternating ...101010...., this a modulating > signal that takes two symbol periods to repeat. So it's frequency is > Rs/2 or 25Hz. > > So when you multiply a Rs/2 signal by the central carrier: > > cos(2*pi*fcentre*t)*cos(2*pi*t*Rs/2) > = 0.5*cos(2*pi*t(fcentre+Rs/2) + 0.5*cos(2*pi*t(fcentre-Rs/2) > > which should look like two sine waves Rs apart > > On 18/08/16 00:00, Phil Frost wrote: > > I've been working on an independent implementation of the FDMDV > modem as > > a personal exercise. As I understood the description*, the central > BPSK > > subcarrier should be at a symbol rate of 50 per second, but looking at > > the tone spacing it's too narrow: but a 25 symbols per second seems to > > line up. Am I overlooking something or is the sync subcarrier > really at > > 25 sym/sec? > > > > *: > > http://freedv.org/tiki-index.php?page=FreeDV+Specification > > http://n1su.com/fdmdv/FDMDV_Docs_Rel_1_4b.pdf > > http://www.arrl.org/files/file/FDMDV.pdf > > > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > > Freetel-codec2 mailing list > > [email protected] > <mailto:[email protected]> > > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 > > > > > ------------------------------------------------------------------------------ > _______________________________________________ > Freetel-codec2 mailing list > [email protected] > <mailto:[email protected]> > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 > > > > ------------------------------------------------------------------------------ > > > > _______________________________________________ > Freetel-codec2 mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 > ------------------------------------------------------------------------------ _______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
