Fwd: Re: [digitalradio] A closer look at ROS]]

Mike O'Dell mo at ccr.org
Mon Feb 22 06:29:17 CST 2010

so where in the regs is the definition of "bandwidth-expansion 
modulation" ??? that would seem to be the crux of this. it's
also damn tricky.

if one wishes to be extremely pedantic, which I don't recommend,
*any* modulation of a raw carrier is an example of "bandwidth-expansion
modulation". there is no other way to modulate - any impression of
modulation on a carrier results in a wider signal, and this includes CW.
It simply follows from Shannon - no increase in bandwidth,
no information added.

so pursuing that argument results in *all* HF operations being illegal.

as a good friend says
	"The problem with reductio ad absurdum is knowing when to stop!"

Viewed this way, HF-RTTY is FSK, and FSK is simply hopping
between two frequencies with the hopping pattern carrying the 
information. I also note that the air interface of "direct FSK"
is absolutely identical to doing FSK at audio and then applying
that input to an SSB transmitter. The spectra of the two are
identical if the AFSK tones are chosen correctly. The signal
can be synthesized both ways equally well.

And take human voice applied to an SSB transmitter. The air
interface spectrum consists of a 3KHz wide block of very narrow
"subchannels", the amplitude of each being adjusted to mirror
the amplitude of the same subchannel in the input voice spectrum.
Given the structure of the human voice, there will be many
subchannels whose amplitude can be considered to be zero at any
instant in time, so again, the pattern of how the energy is
distributed versus time is what carries the information.

Also note that this is precisely now some voice compressors work,
and is also how the Telebit Trailblazer modem got 19.2Kbps down
a barbed-wire fence. Call it "convolution modulation",
or "lattice-coded multi-tone", or "fruit-salad", it's still
the same thing. if one wishes to be extremely precise, the
voice-on-SSB waveform results when one takes the limit as the
size of the subchannel goes to zero - ie, it becomes continuous.
just so happens that Nature does the math for us in the continuous
domain where it would take a bunch of computes to do the running
FFTs and inverse FFTs, so we let Nature do it for us.

I would take the reg language and run with it: a signal not wider
than a communications-quality voice channel (done with SSB on HF),
it is *not* spread-spectrum of any flavor, thereby reserving that
term "spread-spectrum" for signals whose air-interface bandwidth is
*intentionally increased beyond that of a communications-quality voice 

that's basically a "limitation by bandwidth" view. However, I don't see 
any other interpretation that doesn't lead *directly* down the rabbit hole.


"Mathematics does not read regulations nor can it impute intent."

On 2/21/10 5:59 PM, Tom Azlin N4ZPT wrote:
> interested "debate" going on right now on the below reflector. Seems a
> new sound card mode came out where the author declared it to be spread
> spectrum. Thus a huge debate started along two paths. Was the mode SS in
> the first place and if so was it legal. Splinter arguments than all the
> other sound card mods are also thus not legal.
> I have not looked to see if there is a separate psuedorandom spreading
> code added to the carrier to pick up spreading gain. Of course there is
> another mode that is used in ARRL traffic nets called Chip64 which is
> also described by the author as SS. Which is being used.
> Given AMRAD was part of the original SS approvals, what do you all think?
> 73, Tom n4zpt
> -------- Original Message --------
> Subject: Re: [digitalradio] A closer look at ROS]]
> Date: Sun, 21 Feb 2010 17:30:50 -0500
> From: KH6TY<kh6ty at comcast.net>
> Reply-To: digitalradio at yahoogroups.com
> To: digitalradio at yahoogroups.com
> The FCC is only concerned with what happens to the resultant RF energy
> and what is done with it, not how that RF is generated. In the case of
> ROS, if the data is applied to an RF carrier and the frequency then
> hopped, that would classify it as spread spectrum.
> The rules are FCC rules and currently specifically specify spread
> spectrum to be used only at 222Mhz and above. If it were not for that
> specific reference and the statement by Jose that frequency hopping is
> used, then the rules might be subject to interpretation. As it presently
> is, Jose would have a tough time in a court of law to prove he does not
> use frequency hopping or spread spectrum, as he has already claimed.
> Our best chance to legally use ROS in the US is for the FCC to issue a
> ruling. As amateurs, and not even lawyers, we are not competent to
> second-guess the FCC's lawyers and as long as there are so many previous
> claims that ROS is spread spectrum, we are stuck with that definition.
> Our best hope is to get the FCC to amend the regulations, or make an
> exception, to allow spread spectrum as long as it is capable of being
> monitored by third parties and does not exceed the bandwidth of a phone
> signal, and ROS would meet all of those conditions.
> There are those who think that regulation by bandwidth would solve
> everything, but there are also those who would love that chance to take
> over the HF bands with automated messaging services so they do not have
> to worry about crowding anymore. You can be thankful for regulations
> that both protect, and also allow, with limitations, and that cannot be
> changed without a sufficient period of public comment from all users so
> that all sides can be heard from. The FCC adheres to such a process.
> 73 - Skip KH6TY
> w2xj wrote:
>> There are two very common misconceptions in that theory. The first is
>> that SS is unto itself not always a fully digital mode. and A, F, or J
>> in that case indicates the nature of the narrow band signal being
>> spread. So SS with a J designator would be an SSB signal digitally
>> spread by the PN code. When received, it is de-spread to an analog SSB
>> signal.
>> Sound card modes are not necessarily SSB. We use the SSB process as a
>> convenient easy to deploy up converter when operating in these modes,
>> the modulation occurs in the computer code and could be transmitted with
>> varying degrees of ease by other means. CW via sound card is still CW as
>> is the case with RTTY. ROS is no exception. It is quite possible to
>> drive a DDS chip on frequency and accomplish the exact same result only
>> at the expense of greater complexity.
>> KH6TY wrote:
>>> Rik,
>>> Did you see the recent post by K3DCW?
>>> The closest you get to a true definition in Part 97 is in section 97.3
>>> Definitions, Para C, line 8:
>>> /(8) SS/. Spread-spectrum emissions using bandwidth-expansion
>>> modulation emissions having designators with A, C, D, F, G, H, J
>>> or R as the first symbol; X as the second symbol; X as the third
>>> symbol.
>>> ROS uses SSB so the first designator is J (this meets the definition)
>>> and it uses bandwidth-expansion. (this meets that definition as well)
>>> Thus, taking this definition literally, it is indeed Spread Spectrum
>>> and is thus illegal below 222MHz....at least that the conservative
>>> interpretation that I'll stick with until we get a ruling otherwise.
>>> ====================
>>> Dave
>>> K3DCW
>>> Obviously, with such a new mode, there has been no ITU description of
>>> ROS. If it used bandwidth expansion (i.e. frequency hopping), it is
>>> obviously to be classified as spread spectrum. Whether or not modes
>>> like MT63 and Olivia are essentially the same is debatable. The
>>> problem seems to be the direct reference to bandwidth expansion (but
>>> within the width of a phone signal), which, until ruled otherwise (and
>>> I hope it will be) is spread spectrum according to the current FCC
>>> rules, and is currently legal only above 222 Mhz.
>>> 73 - Skip KH6TY
>>> Rik van Riel wrote:
>>>> On 02/21/2010 02:17 PM, w2xj wrote:
>>>>> I have spent the last hour looking through part 97. I find nothing
>>>> that
>>>>> would prohibit ROS in the HF bands subject to adhering to those
>>>> segments
>>>>> where the bandwidth is allowed. In fact the rules would appear to
>>>>> support such operation:
>>>> Lets look at it in another way. Part 97.3 is quite specific
>>>> about what modes are considered spread spectrum:
>>>> (8) SS. Spread-spectrum emissions using bandwidth-expansion
>>>> modulation emissions having designators with A, C, D, F,
>>>> G, H, J or R as the first symbol; X as the second symbol;
>>>> X as the third symbol.
>>>> ROS has no ITU designator marking it as spread spectrum.
>>>> Furthermore, from part 97.307 places this limitation on any
>>>> data mode transmitted in the HF bands:
>>>> (2) No non-phone emission shall exceed the bandwidth of a
>>>> communications quality phone emission of the same
>>>> modulation type. The total bandwidth of an independent
>>>> sideband emission (having B as the first symbol), or a
>>>> multiplexed image and phone emission, shall not exceed
>>>> that of a communications quality A3E emission.
>>>> ROS follows this rule.
>>>> In short, ROS has not been ruled to be a spread spectrum mode
>>>> by the FCC or the ITU, and fits within the bandwidth of a phone
>>>> communications signal on HF.
>>>> It also follows the common sense rule of not causing any harm
>>>> on the HF bands. It really is not much different from the
>>>> other data modulations out there. JT65, Throb and RTTY also
>>>> have empty space between carrier positions.
>>>> I would certainly try out ROS, if it weren't for the fact that
>>>> I don't have a Windows PC and ROS does not seem to run anywhere
>>>> else...
>>>> --
>>>> All rights reversed.
> _______________________________________________
> Tacos mailing list
> Tacos at amrad.org
> http://www.amrad.org/mailman/listinfo/tacos

"Of course it's hard!
If it was easy, we'd be buying it from somebody else!"

More information about the Tacos mailing list