Just say NO! to Winlink expansion!

kh6ty:
"So, even narrowband "voice" communications is right around the corner, and could make all the phone bands less crowded. "

Your Digi-talk program is quite a development so don't take what I have to say the wrong way. Narrow-band voice will be a help but there needs to be some big developments which are still on the horizon. None of the implementations that I have seen let yo basically sit on a frequency and call CQ. They all require some up from coordination so an analog connection is still required. This keeps the effective bandwidths used as high as the analog connection requires. I am not convinced from what I have seen that digital voice, using todays implementations, will be as qrm-tolerant as many seem to think. The digital voice implementation cannot use very aggressive ARQ operation and still remain usable in high-noise, high-qrm environments. There woud be too many long pauses to be acceptable to a lot of people. Less aggressive ARQ operation results in lots of holes and artifacts in the presence of high-noise and high-qrm. This may be more acceptable but is certainly not error-free.

I still think digital voice is a coming thing. I wonder,however, if a bandplan like WA0LYK's proposal is implemented if we won't see more narrowband FM actually hit the HF bands before digital voice. Since all modes with less than a 2.5khz necessary bandwidth would be allowed, narrow-band 2.5khz deviation FM would be allowed. Narrow-band FM provides for noise free voice just like digital voice will. While it doesn't have the narrow-bandwidth that digital voice promises to deliver in the future, it can certainly deliver noise-free operation today!

It will be interesting to see just what the amateur universe is expecting from digital voice, narrow bandwidth or less received noise.

I suppose the limiting factor will be what the equipment manufacturers provide for. I don't know if the FM module for my 751a can be adjusted for 2.5khz deviation or not. I suppose I should start looking!

tim ab0wr

 

Tim,

DigiTalk only works on the receive end. Transmission is still by keyboard entry. However, I can press a "CQ" macro and the person on the other end with DigiTalk will *HEAR*, "CQ CQ CQ DE KH6TY KH6TY KH6TY please K".

All this in 63 Hz bandwidth, with no background noise at all! If there is QSB or QRM, he gets an occasional "nonsense" word, but the program speaks each word as it comes in in text, "spells" out callsigns, and can be taught to "spell out" things like FT1000 or pronounce "pse" as "please". It is all quite simple, using a standard text-to-speech program integrated into DigiTalk.

The idea of narrowband speech is not only to get rid of background noise, but to let more people share the spectrum, so adjacent channel QRM is reduced because crowding is reduced. I use it often to "listen" to PSK31 while on the computer writing. Once, I "heard" a DX country I wanted and would have missed if DigiTalk had no been listening in the background.

However, with a 95% error rate, speech-to-text conversion so I can speak into the mic and have my words transmitted in PSK63 text is not possible real-time without editing out the errors. For example, my test phrase is, "My wife is a doll!". It prints out (and therfore transmits, as, "My wife is a dog!" - Not something to put straight on the air! Hi!

So, I am waiting for speech "recognition" to mature more, but you can see the potential - computer speech in only 150 Hz bandwidth - allowing 20 phone signals in the space of one phone signal today!

If it can somehow be done like MIDI sounds, with your own speech phenomes stored locally, than it can also sound just like yourself! This is what can be done with distributed computing power, and I think should be the direction encouraged by regulations, not just providing more and more bandwidth when overcrowding on phone is a real problem. I am very disappointed in the current digital voice - you can't break in, it drops out sharply with S/N degredation, and it is as wide as analog phone. QRM is the real problem with phone - not S/N!

That's a good point about narrowband FM!

There is no point anymore in hoping Winlink will cooperate. Domination and not sharing is their goal, it seems. What we have to do is convince the ARRL that Winlink, HF Packet, and Emcomm using Winlink can easily fit in the sub-bands with room to expand, by simply time-sharing in a reasonable amount. It seems this possibility has been overlooked, as Winlink asks for everything, but actually needs very, very, little. If fact, by default, wanting to use Pactor-III for everything, they are ALREADY time-sharing and completely in the sub-bands! We only have to convince the ARRL that what should be done is already being done! I have already written Haynie and my Director, but no acknowledgements yet. It is good to keep trying before the July Board meeting, where hopefully it will make a difference.

If you want to see how DigiTalk works, here is the link:
DigiTalk. You MUST download all the files listed so it is good to pay attention to the instructions on the page.

 

You hit a big nail on the head, Tim. The people developing digital for cellular made promises of CD quality, greater range, no static, less bandwidth needed. What we saw was reduced range because the C/I had to be good enough to keep the BER in the range where the listener would not notice. No static, yes....but it was replaced with artifacts that users used to call "Darth Vader". LOL. CD quality was not attainable in the bandwidth. In order to offer the same coverage with digital, carriers had to build more sites which added to C/I problems.

Amateur Radio digital drum beaters have less spectrum than cellular did on a per time slot basis and proposes to used HF, a band more prone to C/I issues. Yet the promises of clear audio, increased range, etc abound. Any promise that 3kHz (or less) digital voice will be anything but tinty, raspy, artifact ridden, artifical sounding crap should be viewed with suspect. Should we move forward? YES! But should we cause any jeopardy for legacy modes in the name of pie-in-the-sky claims? NO!!!!!

One more thing, Tim. On your 23 char/sec calculation above, how many BPS is that by your calculation? I put it at about 1.8k BPS without overhead, and probably 1.2K BPS with error correction.

 

ag4yo:
"One more thing, Tim. On your 23 char/sec calculation above, how many BPS is that by your calculation? I put it at about 1.8k BPS without overhead, and probably 1.2K BPS with error correction."

If you assume no compression and 8 bits per character it would be a straight (23) x (8) = 184 bits/sec

Not very impressive for an effective data rate. If the messages were sent with compression, the effective data rate would be even less.

At 1200bps and 8 bits per char you would expect to get about 150 char/sec.

I think what Skip is pointing out is that P3 is apparently not a very effective mode for short text messages, there is just too much session setup negotiation and speed negotiation involved. I don't feel so bad for having never purchased an SCS modem and just sticking with an old pk-232mbx and Pactor 1.

It makes one wonder why the WL2K people are pushing Pactor 3. Is it because they just don't understand what is going on or because they plan on using WL2k to pass large graphic and data files from the internet to the users? That would, of course, make it more than just an internet email system, it would put it in a whole different arena.

tim ab0wr

 

And, it should be noted that "2.5 kHz deviation" means PLUS AND MINUS 2.5 kHz. That is a 5 kHz width, without guardband.

Obviously, it cannot be used in a 2.5 kHz or 3 kHz bandwidth.

 

I would point out that there is NO DIFFERENCE between inputting a 1000 Hz tone and inputting the human voice. The Balanced Modulator is not set up differently, and does not have a mechanism to know the difference.

And, I would point out that the multitude of frequencies in the human voice are sine waves. They combine to produce an envelope made up of sine waves.

I would also point out that the transmitted output is a linear function of the input amplitude for both types of input.

Finally, I would point out that amplitude modulation involves varying the amplitude of the output signal as a linear function of the input amplitude - whether or not a carrier is transmitted.

Therefore: 1) A 1000 Hz tone input is a modulation. 2) It produces the same modulation that would be produced by any tone (or combination of tones) that would be input. 3) The modulation type is Amplitude Modulation, suppressed carrier.

From these facts, it can be easily - and clearly - seen that your premise is factually wrong, and that the math used to "explain" was simply lifted from the Handbook (chapter 11, to be precise).

If a person sings a pure note (fundamental, no harmonics), the result is a sine wave at a single frequency. This is the same single tone sine wave input that you allege magically creates CW.

I would urge you to explain to all of us by what mechanism the modulator "knows" to treat two different sinewave inputs differently, so that it outputs DSB / SSB for one and CW for the other.

The correct answer to your math: The carrier and audio input frequency(ies) "disappear" (absorbed in networks), and are "replaced' by the carrier frequency + audio input frequency(ies) and carrier frequency - audio input frequency(ies).

 

The answer is "enhanced" Winlink 2000, which means large attachments and images:

Of course, the ARRL immediately gave Winlink everything they asked for, without asking for comments from anyone else!

What's next? Winlink 2000 voice mailboxes? That the "future of amateur radio is dependent upon digital voice, image and data?

Let's see... "Image" swapping, like SSTV or HDSSTV, is a minor activity, requiring only one frequency per band, digital data transfer is mostly Winlink (<1%), and digital voice is the same bandwidth as analog voice, so that does not accomodate any additional growth in the radio amateur community.

 

I'm sorry Larry, but what does the output of the product detector look like when a single tone is applied to the input of a SSB transmitter. It sees the same output as would be generated by a CW signal. The only way to tell the difference is to look at the entire spectrum of the transmitted signal on a spectrum analyzer. You should know that. The electronic circuitry, certainly, can not tell the difference and I submitted your ear can not tell the difference either!

Ken

 

Yes, Ken, if you observe the output of a DSB generator on a scope, before the carrier is suppressed, you can easily see the modulating waveform impressed on the carrier. Then, as you null out the carrier, you see the two sidebands only. Then, if you pass that output through a SSB filter, you get a pure carrier on the scope, just the same as a CW signal.

 

wa5ben:
"I would point out that there is NO DIFFERENCE between inputting a 1000 Hz tone and inputting the human voice.  The Balanced Modulator is not set up differently, and does not have a mechanism to know the difference.

And, I would point out that the multitude of frequencies in the human voice are sine waves.  They combine to produce an envelope made up of sine waves."

Well, you got these two right!

wa5ben:
"I would also point out that the transmitted output is a linear function of the input amplitude for both types of input."

So what? The variables that are involved in changing a frquency  are radians/sec and time, not amplitude. While the amplitude transfer function may be linear, i.e. the multiplication of f(t) by a constant will result in an output of Cf(t), that has no bearing on the multiplication of the frequencies which is what results in the frequency translation, i.e.  f(t) * g(t). Only a non-linear transfer function in frequency or time can cause a multiplication to occur.

For a diode-ring modulator, you are using a time-variant gain function to create the multiplication. I.e the gain of the modulator varies between unity and zero.

          G  = K * f(t)    or
          G  = K * (cos at)

(either the carrier signal or the modulating signal can be used to vary the gain between unity and zero)

The output function is:  K * C * f(t) * (cos at)
where C is your "linear amplification factor".

Please note that since K is, in effect, a square wave with many odd harmonics, the output will have many harmonics also. This is one factor that can give rise to IMD.

For AM, the frequencies added and then convolved with a transfer function such as p(t). p(t) could be the transfer function of a diode (e.g. square-law)

e.g.  f(t) + C(cos At)   convolved with p(t)


wa5ben"
"Finally, I would point out that amplitude modulation involves varying the amplitude of the output signal as a linear function of the input amplitude - whether or not a carrier is transmitted."

Of course it is linear. Do you see a second order amplitude term in the equation?
         Vout = (AB/2) [ cos (at - bt) - cos (at + bt)]

That doesn't mean that a device with a linear amplitude transfer function can do multiplication of two frequencies.

"Therefore:  1)  A 1000 Hz tone input is a modulation.  2)  It produces the same modulation that would be produced by any tone (or combination of tones) that would be input.  3)  The modulation type is Amplitude Modulation, suppressed carrier. "

I'm not sure you understand the difference here. Do you understand WHY the carrier is suppressed? Do you understand what the term "suppressed" means? What happens when a non-balanced modulator is used? Are you trying to say that a SSB-SC signal cannot be generated using a non-balanced modulator?

wa5ben:
"From these facts, it can be easily - and clearly - seen that your premise is factually wrong, and that the math used to "explain" was simply lifted from the Handbook (chapter 11, to be precise)."

It was YOUR premise that was wrong, Larry. Collins Radio realized this as far back as 1960. They were generating CW in their equipment using a sidetone input to the audio chain. I guess the sarcasm that WA0LYK was using just went right over your head.

In other words, YOU at the receiving end, cannot tell if the originating end is using a sidetone input to the audio chain or is using a direct RF oscillator.

Actually the math was "lifted" from the textbooks

Electronic Integrated Circuits and Systems
author Franklin C Fitchen, Electrical Engineering Department
South Dakota State University,
published by Van Nostrand Reinhold Company, Copyright 1970.
Chapters 8-4 and 8-6
and
Signals, Systems, and Communications
author B. P. Lathi
Assoc Professor of Elec. Eng.
Bradley University
published by John Wiley & Sons, Inc
Copyright 1965
Chapters 4.8, 11.2, 11.3, 11.4, 11.5

I tried to keep the math pretty straight-forward. It's nice that it matches what is in the handbook. Here is a more technical explanation if you want it.

The mixing can be done by using the Fourier transform or the convolution theorem.

  Fourier transform:

          f(t)  <-> F(w)

          f(t)(e**jWt)  <->  F(w - W)

A sinusoidal signal can be expressed as the sum of exponentials:

    cos At  =  (1/2)[ (e**jWt) + (e**-jWt) ]

So  f(t)(cos At)  = (1/2)[ f(t)(e**jWt) + f(t)(e**-jWt)]

the Fourier transform is:

 f(t)(cos At)  <->  (1/2)[ F(w + W) - F(w - W) ]


Using the convolution theorem we use the transform of
cos at  :
   cos at <->   pi * [ impulse(w - W) + impulse(w + W) ] = phi(w)

According to the convolution theorem:
 
  f(t)(cos at) <-> (1/2pi) [ F(w) * phi(w)]

using the identity:
         f(t) * impulse(t-T)  = f(t-T)

we get   f(t) (cos at) <-> (1/2)[ F(w + W)  +  F(w-W)]

which is the same as the Fourier transform.

wa5ben"
"If a person sings a pure note (fundamental, no harmonics), the result is a sine wave at a single frequency.  This is the same single tone sine wave input that you allege magically creates CW."

No kidding! That's what a 1000hz tone is? A person singing and a sidetone generator can cause the same frequency signal to occur? Will wonders never cease!

wa5ben:
"I would urge you to explain to all of us  by what mechanism the modulator "knows" to treat two different sinewave inputs differently, so that it outputs DSB / SSB for one and CW for the other."

Larry, that's the whole point we have been trying to get across.

A 1000hz tone into a balanced modulator will product a DSB output. It doesn't matter if it is generated by a human voice or a tone generator! It becomes SSB by using a filter to remove one sideband - whether it is generated by a human voice or a tone generator.

Once that is done there is exactly one frequency output - and you can't tell it from a CW signal generated directly by an RF oscillator!

wa5ben"
"The correct answer to your math:  The carrier and audio input frequency(ies) "disappear" (absorbed in networks), and are "replaced' by the carrier frequency + audio input frequency(ies) and carrier frequency - audio input frequency(ies). "

"disappear"? What in blazes does this mean?

"(absorbed in networks)"  This is closer. Do you understand what that means?

ae1x, wa0lyk and I have tried to explain this to you. It's obvious we are just wasting our time.

The math just doesn't explain the real world, I guess.
(At least the world according to Larry)

tim ab0wr

 

wa5ben:
"And, it should be noted that "2.5 kHz deviation" means PLUS AND MINUS 2.5 kHz.   That is a 5 kHz width, without guardband."

And your point is?

If you had actually read WA0LYK's proposal you would have seen the statement:
"A total bandwidth limit can be specified as "no emissions whose necessary bandwidth is greater than 9K00 (9 kHz). "

Since the next step up in FM deviation is usually considered to by 5khz deviation, don't you suppose that would be past what his recommended deviation would be?

wa5ben:
"Obviously, it cannot be used in a 2.5 kHz or 3 kHz bandwidth. "

And your point is?

The discussion was what could happen with the proposal by WA0LYK, not the proposal by the ARRL.

tim ab0wr

 

Skip,

The footprint for Pactor 3 is very much more efficient than for Pactor 2. This is why the user community on Winlink 2000 uses it over 90 percent of the time. You would, also if your safety and well-being were at stake.
Take a look at the comparitive footprint on
http://winlink.org/Presentations/RFfootprints.PDF.

Secondly, as I have mentioned on another thread, the ARRL HSMM Working group has recommended a 20 KHz bandwidth limit for all HF bands. Here is a piece of their Janurary, 2005, band plan recommendation:

"The ARRL High Speed Multimedia Working Group is extremely concerned regarding the nature many of the current proposals for band plans. We think that most such proposals, as progressive as they may appear to be at this time, will ultimately, in the future, severely restrict the growth and development of Amateur Radio into the digital age of radio communications of the 21st Century.

Any change in FCC regulations will freeze band plans in stone for the Next 20 years. To allow for future development, the HSMM working group recommends that FCC regulations should be simplified with only a single
maximum allowed emission bandwidth for each amateur band:

160m: 10 kHz
80m-10m: 20 kHz
6m-2m: 200 kHz
125cm+: within band

The ARRL can then issue band plans that create segments with lower emission bandwidths and these can change over time as different operating modes become popular. On the 160 through 10 meter bands, one band plan must cover the U.S. to prevent interference. Above 30 MHz, the band plans may be regional."

You can read about it on http://www.arrl.org/announce/reports-2005/january/23a-HSMM.doc. and other such recommendations on http://www.arrl.org/announce/reports-2005/january/


Unattended space, How wide to allow bandwidth, where should this or that protocol be, Where should the protocols that do not yet exist be, Where to operate DX contests without regard to others, and many, many more considerations must be worked out. However, asking the FCC to regulate such matters is not in keeping with the signals they are presently giving, or what the ARRL is proposing.

This does not mean that the ARRL will necessarily follow the advice of the now almost three-year old ad-hoc HF digital committee, (filtered through the Voluntary Resources committee) that you have fixated on, or that they will follow the most recent advice of the HSMM Working Group Taskforce. What it means is they first must contend with the ability to get out from under the static nature of the existing regulatory structure. Smart move!

So, the first issue is who is going to deal with these matters, and how. What the majority of amateurs do today, has little to do with what the majority of Amateurs will do tomorrow. You can prove this by putting out your "DigiTalk" 125 KHz narrowband FEC digital voice application out to the public, and make a major impact on illustrating how very narrow-band digital voice may be practically used instead of widerband digital developments.



Steve, k4cjx

 

While you're here, Steve, are you going to reply to the question I posted earlier? The executive summary is "why we would we trust a voluntary band plan when the frequencies used by your PBMO violate the current ARRL band plan?" My earlier post contains URLs citing both the current ARRL band plan and the Winlink page listing your PMBO's operating frequencies.

73,

Dave, AA6YQ

 

"Any change in FCC regulations will freeze band plans in stone for the Next 20 years. To allow for future development, the HSMM working group recommends that FCC regulations should be simplified with only a single
maximum allowed emission bandwidth for each amateur band:"

Let's look at some of the other stuff the HSSM report says:

"On HF, the 20 kHz bandwidth limitation aligns amateur standards with shortwave broadcasting standards. 20 kHz is the maximum bandwidth of DRM (Digital Radio Mondiale), the worldwide standard for digital audio
broadcasting between 0.1 and 30 MHz. This is also the minimum bandwidth that would allow HSMM applications as defined by the ARRL Technology Task Force (56 KBPS minimum data rate)."

"One goal of the new HF allocations is to encourage development of new higher-speed digital modes and this can only be accomplished by giving developers the flexibility of using wider bandwidths. This would allow
sharing of the sub-band by numerous stations with bursty traffic and be more efficient than multiple lower speed connections. High-speed links will be invaluable during emergency situations by allowing very efficient simultaneous transfer of voice mail, email and facsimile traffic from the affected area to the outside world."

"We understand that amateurs using the existing HF CW and phone bands want protection, and if the ARRL must reflect that we recommend 200 Hz statutory limits at the lower for CW and low-speed data, and 6 kHz statutory limits at the upper end of each band to allow existing SSB, ISB and DSB AM operation plus any other modes that fall within the bandwidth limitations. At a minimum, 20 kHz wide emissions should be allowed in the following segments:

   3.58 - 3.725 MHz
   7.035 - 7.125 MHz
   14.065 - 14.15 MHz
   21.08 - 21.2 MHz
   29 - 29.7 MHz"


Where do I start?
   1. The statement "freeze bandplans in stone for the next 20 years" is wildly irresponsible rhetoric with no other purpose that to exact a viscereal response from the reader that :   "Woah!! We can't have that! We need to do what the group wants!!"

    The FCC has been very responsive to changing bandplans   ----  WHEN A VERY GOOD REASON IS PUT FORTH!!!!

   2. "aligns amateur standards with shortwave broadcasting standards"

    Since when has amateur standards been driven by shortwave broadcasting standards? We don't have to follow broadcasting standards for digital voice in order to implement a narrow version of an ODFM protocol on the ham bands.

    There is absolutely NO reason given for why we should align with broadcasting standards.

   3. "This is also the minimum bandwidth that would allow HSMM applications as defined by the ARRL Technology Task Force (56 KBPS minimum data rate)."

Here is the REAL reason, cleverly hidden behind a plea for adherence to worldwide audio broadcasting standards!

The real goal is to open up the ham bands to even wider bandwidth data signals. All for the purpose of passing "voice mail, email and facsimile traffic" "

Please note carefully the addition of VOICE MAIL to this list. Not just email and facsimile, but VOICE MAIL.

Slowly but surely, just as Skip has pointed out, we begin to see an erosion of the purposes for amateur communications.

First it began as personal messages ala NTS traffic. Then it went to personal emails. Then it went to safety of life and property text messages such as PIRACY notices, weather reports, etc. Then it went to internal emails for emergency agencies - whether they are related to a declared emergency or are routine daily business. Now it has progressed to internet graphics and VOICE MAIL.

Remember, a lot of us do internet browsing on SLOWER speed links than 56kbps. And this group wants to dedicate 30%  - that's ONE THIRD - of the 80m ham band to links of this speed. Links which will allow direct internet browsing.

How long do you think it will take for these links (you can only get about 6 of them in the 145khz they are requesting) to become saturated with internet traffic and the **** voluntary **** bandplan to begin to be violated?

   4. Remember, the plan has this statement:
"The ARRL can then issue band plans that create segments with lower emission bandwidths and these can change over time as different operating modes become popular."

Now, we can assume ONE of two things here. Either these ARRL issued band plans will be voluntary or they will carry the force of law.

If they are voluntary, then I repeat the question I asked in Item 3. How long do you think it will take for these links (you can only get about 6 of them in the 145khz they are requesting) to become saturated with internet traffic and the **** voluntary **** bandplan to begin to be violated?

If they carry the force of law, then ask yourself - DO YOU WANT THE ARRL TO BECOME A QUASI-GOVERNMENT AGENCY controlling the use of the ham bands? To me that would be a dangerous, dangerous precedent. There is no way I could condone such a thing.

The only other alternative would be for the FCC to set such an allocation up by regulation and by the committees own statement that would be in place for the next 20 years! It's obvious that they either don't believe their own rhetoric or they understand that voluntary bandplans will be unable to stop their unregulated, unrestricted growth - which is their goal.

k4cjx:
"So, the first issue is who is going to deal with these matters, and how."

You are absolutely right. And it is my firm belief that it should NOT be the people that are making the proposals today.

Folks, even if you don't believe that what WL2K is carrying today is commercial traffic in competition with legitimate telecommunications business - you should all be able to make the correlation between VOICE MAIL traffic and telecommuncations. That's *commercial* business - nothing else. And to those that don't understand that this is just the top of the slippery slope - I hope you aren't really interested in anything else in ham radio because if we continue to follow the policies being presented by this cabal we will see ham radio become commercial service bands in the next decade - and we will have lost our dear hobby.

k4cjx:
"What the majority of amateurs do today, has little to do with what the majority of Amateurs will do tomorrow."

Steve, your hubris is amazing. Your belief that the majority of hams today are going to become totally invested in what you see as the most important part of ham radio is just unbelieveable. The internet has been around for over 20 years. Yet the use of the ham radio spectrum for access to the internet amounts to a miniscule part of the hobby even today. JNOS has been around providing access to the internet via ham radio for what? 15years? Yet it still is used by only a small percentage of ham radio.

Yes, you can provide free internet access to all kinds of people and get them to generate all kinds of usage on the ham bands. But it won't be HAM RADIO. 99.9% of those people won't care what the link is from their "black box" and won't do any more than they need to in order to get a "operating license". Then those of us, to whom the art of radio is an avocation as well as a hobby, will just get drowned out in the noise generated by these free "internet access links". As WA0LYK has pointed out, "beware the internet"!

tim ab0wr

 

k4cjx:
"The footprint for Pactor 3 is very much more efficient than for Pactor 2. This is why the user community on Winlink 2000 uses it over 90 percent of the time. You would, also if your safety and well-being were at stake.
Take a look at the comparitive footprint on
http://winlink.org/Presentations/RFfootprints.PDF."

Steve, what Skip has posted blows this out of the water. The rf footprint calculations shown on this web site are misleading at best, and irresponsible at worst. You are misleading people into thinking that the faster protocol is best for short messages when the link setup times may actually INCREASE the amount of time needed to pass traffic over a slower, but more quickly negotiated, link.

It is obvious from *YOUR OWN TRAFFIC DATA* that Pactor 3 has a huge overhead which tremendously impacts your throughput levels on short text messages. Yet your "rf footprint" calculation doesn't take this into account in any way.

The fact that you have not designed either your PMBO software or requested that client software consider the message makeup in deciding what protocol to use in sending the message shows that you have exactly ZERO interest in managing the spectrum use of WL2K to minimum levels, let alone mimize the latency that your subscribers see. That makes your spectrum use irresponsible and that IS going to be in any comments to the FCC that I make concerning this proposal.

You know, you need to go look up Lincon's quote about fooling people.

tim ab0wr