KH6TY wrote: > The difference between ROS and MFSK16 at idle (i.e. no data input), is > that MFSK16 has repetitive carriers in a pattern, but the ROS idle has > no repetitive pattern and when data is input, the pattern still > appears to be random. Note the additional carriers when I send six > letter "N"'s in MFSK16. It then returns to the repetitive pattern of > an MFSK16 idle. Note that the data (i.e. "N"'s created new carriers > depending upon the data. In this case, the frequency carriers are data > dependent. > > If ROS is just FSK144, then I expected to find a repeating pattern at > idle, but I never see one, even after letting ROS idle for a long time > in transmit.
It's pretty common in modems to randomize the data to prevent carriers when sending all zero's or ones. Phone modems do it, I'm pretty sure P3 does, and other RF modems do. I know of another amateur RF modem that had randomized spectra by design. By this test it would have been considered spreadspectrum, but it was not, it was mfsk with a randomizer. The randomizing algorithm was provided to the FCC, and life was good. This was before SS was allowed at all, and there was not a bit of discussion that it might have been spread-spectrum. If MFSK16 was randomized would it magically become spread-spectrum? All I know is, this is not the spread spectrum everyone is worried is going to ruin the bands! IE: traditional spread spectrum with bandwidth expansion of 100-1000. Have fun, Alan km4ba
