FCC Releases "symbol rate" NPRM

Discussion in 'Amplitude Modulation' started by K5UJ, Jul 29, 2016.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: Left-3
ad: Subscribe
ad: L-MFJ
ad: abrind-2
  1. N1EN

    N1EN Premium Subscriber QRZ Page

    No, I don't think you've missed anything in the regs.

    I should point out that the technical side isn't my forte. For purposes of this subject, I'm more into the logic of what might be if a change is/isn't made.

    Traditional reasoning holds that faster signals require more bandwidth. The precise relationship between bandwidth and speed has changed over time, with the development of newer modes and perhaps better transmitters...but if you fuzz up your eyes and look at it at a high level, it's still generally true.

    Now, consider the use of data modes on HF. In amateur use, we primarily use data for keyboard-to-keyboard communication or to move relatively small text files.

    Keyboard-to-keyboard modes don't need to be high speed (and yes, I'm blurring the distinction between "wpm" and symbol rate here), and effective communication can be achieved, even in weak-signal or disturbed/fading conditions with relatively little bandwidth required. There's no reason to go wider.

    Sending text files is a different matter. In theory, we could achieve pretty speedy throughput at a symbol rate of 300 with an ultrawide signal. However, such a transmission would be exposed to various issues with fading (although modulation choice could limit the effects, and error correction can help), and the signal would have a little less "oomph" due to the frequency span covered than a narrow signal (see, e.g., how low power CW is more effective than low power SSB for DXing).

    Given those issues, constrained to a 300bps symbol rate, relatively narrow modes are likely more effective than ultrawide modes. Ultrawide modes would become more logical to use only when a faster symbol rate is available -- the ability to transmit symbols faster, either to increase throughput or to better accommodate the overhead of additional error-correction overcome the headaches of fading and spreading power across additional bandwidth.

    Outside the ham bands, ignoring OTH radar and the like, the cool kids seem to be playing with modes that are about as wide to slightly wider than SSB voice, with symbol rates of 2000bps or more....so I suspect that it will be quite some time before a 5+ kHz wide signal makes enough sense for someone to spend time doing the development work resulting in a viable mode (ignoring experimenters' who dabble "just because"). But maybe that time won't be too far off, depending on what happens with things like SDR.
  2. AC0OB

    AC0OB Platinum Subscriber Platinum Subscriber QRZ Page

    That is because I do not drink at the fountain of the ARRL.

    When I see the ARRL making what I think are ridiculous or demeaning statements, or proposing changes with which I do not agree, I have a responsibility to criticize those statements and or proposed changes.

    Again, I think the ARS community has to decide if it wants the ARS to become an extension of the internet with digital signals > 500 Hz bandwidths cluttering the bands and causing interference, or do we want a service that accommodates phone modes and narrow bandwidth digital modes.

    There are many narrow bandwidth digital modes and protocols, with open source software, that can accommodate digital traffic today.

    In my view, it was a mistake to allow any digital mode with a resulting badwidth of > 500 Hz.

    AC0OB - A Place where Thermionic Emitters Rule!
    Besides, when you're a Ham, you experiment with and improve boat anchors - that's what you do!. [​IMG]
    Last edited: Sep 6, 2016
  3. AF7TS

    AF7TS Ham Member QRZ Page

    [Responding to N1EN post #31]

    Okay, I see your point about the 'natural' datarate limit for HF, and agree with the bulk of what you say. A faster signal will require a wider bandwidth, and a wider bandwidth signal will mean (for the same total input power) less energy per bit and thus what is essentially a 'weaker' signal.

    (As an aside: there is no direct relation between bits per second and symbol rate; but in general the required signal/noise ratio goes down as you increase bandwidth for a given bit rate. At first people look at this and say whoot lets all use spread spectrum because the required signal/noise ratio just keeps going down. However they neglect that the received noise goes up as you increase bandwidth. With _very large error bars_ operating at a bandwidth of approximately 1Hz for each bit per second you want to move is pretty much as good as you can get in terms of energy per bit sent across a link.)

    So if someone wants to send faster than a keyboad chat they are going to quickly reach the point where their signal doesn't reach all that far because it will fall below the noise...and if they really want to send data short distances at megabit speeds (which one certainly can do using technology available to amateurs) then HF is pretty much _not_ the frequency range to use :)

    I disagree on one point: I don't see that the symbol rate rule makes much of a difference to the above.

    There may be a small difference between a single 1200 baud signal and 4 'separate' 300 baud signals, but both will move roughly the same amount of data over the same bandwidth with the same sort of noise problems. One or the other _may_ be easier to implement, but given existing computational horsepower, I don't think they would be much different.

  4. N1EN

    N1EN Premium Subscriber QRZ Page

    Well, that's (a) part of the reason I cautioned about my lack of technical credentials, and (b) part of what I was trying to get at by acknowledging that I was blurring measures.

    (I was exhausted when I wrote that post. I might have made a jump or two in logic. :) )

    Another (perhaps better?) way to get at the point I was trying to make is this:

    Assume that we're talking about 300bps signals. All other things being equal, a transmission made up of 2×300bps signals has a better throughput than a 1×300bps signal. 3×300bps is better still, etc.

    The exposure to fading, reduction in average power per 300bps stream, etc. accelerate the diminishing incremental returns at a rate faster than implied by a (N+1)/(N) factor. There's a point beyond which it makes no sense to go any wider.

    To justify going very wide, you need something else, to keep the diminishing returns curve above the "why bother" level. And, one of those possible somethings could be a faster symbol rate.
  5. W6RZ

    W6RZ Premium Subscriber QRZ Page

    Technology wise, that day is already here. Here's a quick simulation of a 10 kHz OFDM modem described in Annex 4 of ITU-R M.1798.1 (the same document that describes Pactor III in Annex 3). I already had an OFDM generator block, so it only took a few hours to create this waveform. I tossed in some boosted pilots and a center null just for grins. The raw data rate would be 15,600 bps at 41.666 baud with QPSK modulation.



    But how would I monetize this modem? I can't use mainstream ham radio transceivers, so I would have to build a custom modem that included all the things in a transceiver (VFO's, display, knobs, 100 watt PA, etc). I'd have to be out of my mind to do that. It would be a super expensive box that nobody would buy.

    Also note that it takes up the entire 20-meter ACDS segment. HF e-mail users would be QRM'ing themselves with this modem.
  6. K5UJ

    K5UJ Ham Member QRZ Page

    One problem with any of the modern digital stations is that due to a FCC rule change years ago, they are required to identify only in the mode and protocol they are using to transmit the rest of their information. I believe the CW ID requirement was dropped sometime in the 1980s, when digital, pretty much meant analog SSTV and mechanical baudot RTTY. That was a bad idea then and now, I'll wager 90 to 95% of hams on HF are not equipped to identify a digital station that is deliberately or unwittingly causing QRM. I hate to beat this to death (I've mentioned this around 20 times elsewhere) but a CW ID at the tail end of a transmission that could be automatically included and encoded at around 30 wpm would make it possible for many hams to identify a station. With all these modes being implemented with software, it should be simple to encode a CW ID in the programs. An ID at the end would not disrupt anything--the transmission is about to end anyway. I'm not about to get a computer in my shack and all the software etc. just to find out who someone is, who is causing interference. As it is, I and many many other hams, are unable to determine if a digital signal is even a ham radio station signal. I might be hearing RFI from an appliance or an intruder. I can't think of any reasonable argument against reinstating the CW ID requirement.
  7. SM0AOM

    SM0AOM Ham Member QRZ Page

    This is one of the reasons that the serial-tone modem was pursued for HF data transmission in the mid-80's, with the introduction of STANAG 4285 and MIL-188-F serial waveforms. The use of parallel-tone modems such as the Kineplex (of 50's vintage) and Harris 39-tone was considered inefficient in terms of transmitter peak-to-average ratios and linearity requirements. It seemed difficult with the current state-of-the-art to exceed an 1 bps/Hz specific data rate, especially in dispersive ionospheric channels.
    This was also fuelled by the Watterson and Stephen Cook research efforts into error-control and equalisation schemes in the late 80's.

    In order to reliably achieve more than 1 bps/Hz the serial tone approach using adaptive equalisers and adaptive interference cancellation schemes were further
    pioneered by Harris developers and by Brakemeier et al at Deutsche Aerospace, with Mark Jorgensen at Rockwell-Collins further pushing the state-of-the-art.
    In the mid-90's several 4800 bps serial-tone modems using one SSB channel bandwidths surfaced.
    Some years later, Harris offered a 9600 bps HF modem with 3 kHz bandwidth for use in the STANAG 5066 architecture.
    Currently EADS/Thales, Rohde&Schwarz, Harris and Rockwell-Collins offer 19200 bps HF modems based on serial PSK/QAM waveforms.

    All present serial-tone modem realisations are more or less descendants from this heritage.
    When your system is peak-power limited (which is very prevalent in airborne HF systems) it makes good sense to use the available power in the most efficient way.
    In order to do this, the peak-to-average ratio must be controlled, and this is difficult in the parallel-tone realisations, as this ratio easily can be 10-12 dB or more.

    Working since the mid-80's with civilian air/ground and military general purpose HF data transmission systems, I have followed this evolution closely.

    The early/mid 80's inexpensive modem state-of-the-art that was represented by the 150/300 bps ARINC-Parker FEK modem, the Global-Wulfsberg MSK 300 bps, and the CRC-Canadian Marconi-Glenayre 600 bps MFSK "narrow-band-spread-spectrum" modem was considered not to have sufficient performance to form the backbone of a proposed HF datalink for air traffic management purposes. Also, a subset of the 24-tone or Kineplex modem tailored to fit into the 400-2500 Hz nominal passband of an ARINC 719 HF radio was discussed.

    For some 9 years, I was an adjunct to the Airlines Electronic Engineering Committee and the ICAO AMCP HF datalink advisory groups,
    where these questions were discussed in-depth. Performance issues with parallel-tone modems lead to the proposal of a modified MIL-188 waveform
    with a nominal DQPSK highest modulation rate of 1800 baud, and with more and more robust fallback rates. 3600 bps using 8-QAM was as an improvement path.

    When first fielded in 1990 the then Sundstrand Data Control (soon to be Allied Signal and finally Honeywell Aerospace) system worked very well, and became the core of the ARINC 653 and 753 Characteristics which eventually found their way into ICAO guidelines.

    Military users also found that parallel-tone or MSK modems had insufficient performance.
    After extensive evaluations in the mid/late 90's the tailor-made 30-tone "Arctic Waveform" that was developed by Marconi-Selenia for the HF2000 system became abandoned in favour of serial MIL-188 and STANAG 5066 waveforms.
    Incidentally, this waveform had a very strong family resemblance to the HF Digital Voice system that was developed by G4GUO and briefly marketed by AOR in Japan.

    The FCC is stuck in the early 80's state-of-the-art if keeping serial tone modems illegal.

    It is highly unlikely that wide-band HF data waveforms used by radio amateurs are surfacing within the foreseeable future, as the current generation of SSB transmitters can be expected to have a service life of 20-30 years.

    By the same reasoning they could as well outlaw 850 Hz shift RTTY on HF as a form of "spread-spectrum".

    N6YW likes this.
  8. N6YW

    N6YW Ham Member Volunteer Moderator QRZ Page

    This is fascinating.
    Although it has very little to do with AM aside from what was initially posted at the beginning of the thread,
    I am going to allow this discussion to continue. The talented people who have spent quite a lot of effort
    in this thread are to be commended in helping us understand the intricacies involved with this subject.
    I would like to see various solution schemes discussed further. If the ARRL is lurking here, perhaps they
    take notice and gain some useful insight. Let us hope that common sense prevails and in the end,
    leave AM alone.
    Billy N6YW
  9. K4KYV

    K4KYV Premium Subscriber Volunteer Moderator QRZ Page

    The main reason the AM community should be concerned over this NPRM is that it could open the door to specific bandwidth limits for all modes. One comment I read recommended that the FCC impose the 2.8 kHz limit for both digital and phone signals, adding a 6-kHz "exception" for AM, essentially a redux of the ARRL's failed "regulation by bandwidth" petition. Not only would this be harmful for AM; if strictly enforced, it would constrict all forms of amateur radio experimentation, development and building, since a ham would have to worry about his transmitter meeting the legal numerical bandwidth standard before even putting it on the air for the first time. This would further promote the plastic radio/appliance operator aspect of ham radio at the expense of the technical.

    The CW community may have a concern, out of fear that digital hash may eventually take over major portions of the CW bands. Another point to ponder is the purpose of mode sub-bands in the first place, if SSB-wide digital signals are allowed to operate in what is traditionally known as the "CW bands".

    The FCC has chosen not to go the route of specified bandwidth limits with this NPRM, but as I read it, they have nevertheless opened for discussion whether or not that would be a good idea.
    N6YW likes this.
  10. N6YW

    N6YW Ham Member Volunteer Moderator QRZ Page

    Thanks for the further clarification Don.
    It appears there lurks a dangerous precedent. I wonder if the FCC would like to do away with amateur radio altogether.
    It certainly would make sense seeing how our government really doesn't profit from us and we as a general rule can be
    quite a pain. Just sayin. I certainly hope not.

Share This Page