I'm looking for an explanation of something that was probably considered during the design of the 1600 mode but is implemented differently from other systems that use pilot tones.
Looking at the spectrum and the scatter diagram I see that the pilot tone amplitude is much larger than the QPSK carriers and I wondered why. The phase of these pilot tones (BPSK?) is the same as 2 of the QPSK points, in other systems I'm familiar with the BPSK modulated pilots are at a 45 degree offset from the QPSK points and have a lower amplitude because there is both phase and amplitude difference from the QPSK symbols to assist demodulation. This allows the use of a lower PAPR in the modulation, which is beneficial because it allows a larger amplitude for the QPSK bits and a less linear transmitter can be used. No criticism is intended, I'm just trying to understand the trade offs, I suppose that there are only 4 symbol phases to worry about whereas with offset phase BPSK there are then 6, which really means 8. No rush to get an answer, this is just an itch I wanted to scratch when I was testing with the recorded files that come with freedv and what I saw got me thinking. Cheers -- Brian G8SEZ _______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
