[RTTY] ARRL issues Official Reply: Re 2.8KHz HF digital BW

[Kai]

Hi Chen
The Henning Harmuth story was really interesting. I've run into his works during 
my tenure with Ultra Wideband and IEEE802 standards work. And folks worry here 
about 2.8 kHz BW signals!  Thanks for the nice recollections.

We must remember that when radio started it was all wide band (spark) and was 
called "wireless". Eventually spark was abolished, and wireless became narrow 
band and was called "radio". Then came Armstrong who challenged "the narrowband 
mantra" to give us the audio quality of wide band FM. Later, Qualcomm introduced 
wide band digital spread spectrum, challenging the narrow band mantra once more. 
Well, we have wide band (up to 40 MHz BW) CDMA now (in dedicated bands), and 
"radio" has again become "wireless"!

Henning was right: wide and narrow don't mix very well, but remember that 2.8 
kHz ain't really wide! Many already legally do around 2 kHz digital at HF with 
currently authorized emissions (Pactor, Amtor) at HF. The proposal just ups that 
to 2.8, and more importantly, gets rid of the baud rate limitations.

My CW will always get through, and we'll always have 170shift 45.45 baud ham-RTTY.

Thanks again, and 73
Kai, KE4PT

On 7/25/2013 4:20 PM, Kok Chen wrote:
>> Does this comfort you?
> As comforting to a CW op feels when I unleash 2.8 kHz wide digital signals down at 14.025 MHz, where I am authorized by the FCC to do.
>
> Wide signals and narrow signals just don't mix (I still remember a quote by Henning Harmuth at an IEEE conference back in the 1970s regarding the use of Walsh Functions as radio carriers).
>
> Keep 2.8 kHz signals above 14.125 MHz and it might make sense.
>
> Otherwise change the existing symbol rate rules to limit bandwidth to 500 Hz.  Not 2.8 kHz.
>
> Re: Harmuth.  Henning Harmuth had back in the 1960s proposed a different orthogonal basis instead of sine waves, and had developed an entire system (including how to phase antennas for Walsh carriers).  His orthogonal basis?  The Walsh Function.  And instead of Fourier Transforms and spectrum, you have Hadamard Transforms and Hadamard spectrum.
>
> At one conference, someone pointedly asked (I paraphrase): "Prof. Harmuth, your system would just splatter all over our spectrum of carrier based signals, making the existing systems useless."  Harmuth's reply: "No, it is *your* carrier based systems that are creating wide splatters in *my* Hadamard spectrum and rendering my system useless."
>
> Now imagine that the Hadamard stuff extends over 2.8 kHz.
>
> Ivory Tower? Look up Walsh Functions and Hadamard transforms on the web and you might find that some of your favorite digital modes actually use them (but over a narrower bandwidth).
>
> I still have Harmuth's "Non-sinusoidal Waves for Radar and Radio Communication" (1981, ISBN 0-12-014575-8) on my book shelf.  Fascinating read (stuff like how to construct bandpass filters for Walsh functions) if you like thinking out of the box.  There is even a section on "Bandwidth Required for Teletype and Data Links."
>
> 73
> Chen, W7AY
>
> _______________________________________________
> RTTY mailing list
> RTTY at contesting.com
> http://lists.contesting.com/mailman/listinfo/rtty
>