Hello Simon, >So what about the different bandwidth? Assuming SSB is 2.4 kHz and PSK31 is 31 >Hz, is there >an extra advantage here of 800 (19 dB or whatever) as well? The bandwidth is already taken into account. The standard bandwidth for determining S/N is 3 KHz. The small bandwidth of PSK31 is one of the reason why PSK31 is better than SSB. Speed modulation, type of modulation and coding do the S/N.
In general for the same coding, the bandwidth is reversely linear with the S/N. Example: * S/N of PAX: -6 dB * S/N of PAX2 (which twice more rapid that PAX and so has twice more bandwidth): -3 dB. It's true for PSK31 and PSK63 and so on. 73 Patrick ----- Original Message ----- From: Simon Brown To: [email protected] Sent: Thursday, November 23, 2006 4:40 PM Subject: Re: [digitalradio] MFSK 16 Beacon Thanks Patrick, So what about the different bandwidth? Assuming SSB is 2.4 kHz and PSK31 is 31 Hz, is there an extra advantage here of 800 (19 dB or whatever) as well? IMO Digital Modes are ideal for QRP and for those who want to get away from the screaming guys with 1,000m towers and 10kW amplifiers. Simon Brown, HB9DRV ----- Original Message ----- From: Patrick Lindecker Hello Simon, I measured the minimum S/N (in general for about 2 % of errors) for all the Multipsk modes (look at the chapter "RX/TX modes selection and their descriptions" in my help). I did not measure in fact the minimum S/B for a SSB QSO. I suppose that for two Hams speaking in the same native language and with a normal pronunciation, the minimum S/N must be about 0 dB, but I think it is very variable. For aural CW , it is supposed to be -15/-20 dB for very good CWers (and I put this figure in my help). However, I tested myself at 20 wpm (my level must be considered as not very good, see weak, in CW) with a CW signal in gaussian noise. I could decode only to about -5 dB, so the human psychoacoustic dynamic must be very large... 73 Patrick
