[RTTY] Fw: ARLB006 NTIA: No Objection to Additional Data Modes on60 Meters

Kok Chen chen at mac.com
Thu Mar 29 18:56:53 PDT 2012


On Mar 29, 2012, at 6:45 PM, Frank wrote:

> Name four.

I have already given one.  The next biggest problem with FSK transmission is baud rate.  And it is something most FSK ops are clueless about.

In the old days, like on my ST-8000, real UARTS were used.  Unfortunately, the last USB device I used that was capable of 45.45 baud is the Belkin F5U103.

Today, you have the choice of (1) bit bangers such as what you get from MMTTY's FSK operation, or (2) the wrong baud rate from the MicroKeyer family (I bet you didn't know that either, did you?), or (3) tone/AFSK converters.

(1) The bit bangers produce bit jitter but not accumulated error (each bit clock has the same amount of zero mean error at the receive end).  It not only produces more decoding errors, but additional keying sidebands in the form of a raised noise floor (asssuming a rectangular probability density function).  OM7ZZ has measured pretty scary jitters on typical Windows machines (causing even more problems with 75 baud).

(2) The MicroKeyer family's baud rate is determined by a division ratio.  However, there is no division ratio that produces 45.45 baud.  You can get 45.00 baud or somewhere around 45.8 baud.  The good part is that this does not raise the keying noise floor (so you are not adding QRM to an adjacent station) and you can also get preceise 75 baud.  Just not 45.45 baud.  However, the bit error rate is no longer uniform.  The most significant Baudot bits suffer more error than the least significant bit.  It does not affect transmitted signals, but causes a fraction of a dB loss in SNR at the receiving end. 

(3) Even if you use on-off-keyed FSK schemes such as the OOK mode in the DigiKeyer II, or the K4DSP FSKit, there will be clock jitter.  They are smaller than the MMTTY FSK, generating a uniform probability density function that extends for 0.5 ms if you start with Mark and Space tones around 2000 Hz.  Again, the problem is greater when you yse 75 baud.

With the USB audio class, data is transmitted isochronously in addition to having stable clocks for the sampling rate.  Thus the AFSK signal is generated to any baud rate precision as you wish.

If you want more examples of AFSK advantages, look up things like agile RTTY tuning, and never having to worry about inverted RTTY signals.

With AFSK, you also will never put out an unmodulated carrier (you hear that often in RTTY pileups :-).  You either get an honest RTTY signal or no signal at all (and you should be able to see the latter from your RF power meter).

73
Chen, W7AY







More information about the RTTY mailing list