[RTTY] High speed RTTY

[Kok Chen]

On Jan 27, 2010, at 3:57 PM, Bill, W6WRT wrote:

> I rarely use PSK myself, so I have a question:  Since PSK (any speed)
> has a reputation of not working well for multi-hop DX due to phase
> shift, wouldn't PSK125 be even worse there than the traditional PSK31?

Yes, PSK31 is heavily influenced by multipath distortion, to the point that given enough multipath, you cannot punch through no matter how much practical power you use (HAAPR type power not included :-).  With RTTY, you can work through pretty bad flutter if you apply enough juice.

However (there is always a "however" :-), counterintuitive as it might sound, shortening the symbol period can actually reduce the effect of multipath distortion when it comes to phase shift keyed signals.

PSK31 is a form of *differential* phase shift keying.  The value of a bit is not the absolute phase of the signal itself, but the phase difference between the signal at time t and the signal from one symbol period ago, at time t-T.  A symbol is equal to one bit in BPSK31 and one dibit in QPSK31, both have the duration of T = 32 ms.

If the change of relative attenuations of signals from two different paths causes a change of phase by a constant rate, n-degrees per second, then reducing the time between symbols actually reduces the component of demodulation errors that are caused by the multipath phase error.  (The Eb part of Eb/No will still be smaller, of course.  But your question is about phase error when baud rate is increased.)

The symbol period of PSK63 is half (16 ms) that of PSK31 (32 ms), and the symbol period of PSK125 is a quarter (8 ms) of the period of a PSK31 symbol.  

73
Chen, W7AY