Wide Bandwidth Digital Danger

Discussion in 'Ham Radio Discussions' started by W6EM, Dec 25, 2018.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: Subscribe
ad: Left-3
ad: L-MFJ
  1. N1FM

    N1FM Premium Subscriber QRZ Page

    Ok, I did some monitoring earlier. I found a free program, called Sorcerer, which is able to copy messages in Pactor 2 mode with one hundred percent accuracy. No problem. Sorcerer can also copy the headers of Pactor 3, which are apparently sent without using any proprietary compression. The beginning and end of the messages are sent in the clear, while the middle is gibberish.


    Note from Wikipedia:

    "As the data is compressed, and the decompression algorithm is not publicly known, the contents of PACTOR transmissions are unreadable to anyone without a PACTOR capable terminal node controller."


    The concern with the proprietary compression system, is that average hams, and the public (and potentially FCC and OOs), can't see the content of the message, and therefore the obscured text can't be monitored for legality. Pactor 4 apparently takes more bandwidth and has better (faster) throughput and uses a similar compression scheme.

    "PACTOR-4 reaches a higher data throughput than PACTOR-3 over the entire working range from -20 to +25 dB @ 4 KHz."


    Apparently, the Radio Mail Server (RMS) can see the full message, and the recipient/addressee can see the full message, but it's munged for everyone else, like this (actual OTA copy):

    ; KW1U de WB9FHP
    FC EM 108754WB9FHP 471 316 0
    FC EM 108755WB9FHP 447 295 0
    FC EM 108756WB9FHP 447 295 0
    FC EM 108758WB9FHP 466 310 0
    FC EM 108759WB9FHP 447 296 0
    F> 67
    nPEMBROKE MA¥ÔîfrØ~ÐoŽvnúÿûwJ¾º…Å
    ui^MigàþUýª>ü©6ÿïßæ•L8Â,Ät=ü(|ÇÁÌ^lé³À>﵇¾Æmz/.{ݾ¼3Vð¬(˜Âyb‘“…®Çªá$Š‰î`ì4ÂSx¢œIþþýý͆™ÕR„ƒ¯LïX•â$fþü|B÷öKÁ,†(?”TñÌ}ď‡¯È³'-4îÉcoßÀßyŒ'Å?eÛΊÓãÕVdØã÷ȯ!Y€TnPEMBROKE MA|,õþöî•…ûš”ëS¥-§€wùwíQwåI¸ÿ~ÿ4êÑF¥àŸïéCæ>nbógM™÷}¬}öP1mÑyuÞíõá[¸…aÅGÉ”Œ™|(]vJºDºIn_@nÁÌ}~n_«?ÁMAÿl<â:á<õø÷’â[(d3l?€
    JxÞç÷ȦB›ÝnPEMBROKE MAwíQwåI¸ÿ~ÿ4ªQF¥àŸïáÅæ>nbógM™÷}¬}öP5mÑyuÞíõàÛ¸…aDFÉ”Œœü(]vIeRUM˘؏¯Í›õgø)éÿíZ—v€wùwíQwåI¸ÿ~ÿ4ªQF¥àŸïáÕæ>nbógM™÷}¬}öP1mÑysÞíõá[¸…aÅGÉ”Œ™|(]vJºDºIn_@nÁÌ}~n_«?ÁMAÿl<âܸ °<1K¿þÝtw‚צÀà¥<p|ûù´5M€Þ; KW1U de WB9FHP
    FC EM 108760WB9FHP 491 329 0
    FC EM 108761WB9FHP 474 319 0
    F> CB
    nPEMBROKE MA/K©ÜÌå°ý |,õ|
    üªûT}ùr®?ß¿Í:”q„i‡è{úHœÇáÌ^lé³À>﵏¾Ê-º/.»Ý¾¼CXp¬8¸âyb‘“…ÏH°¼Au‚t»3‰?ßÐ$¿i&wþoßÎ¥éÃ’X2‹û‚<K½Í1׶wç¹±Ï$Puéë,<Dœßß‘†O>ðK!Êoåtð!çúñwq#ÈYìYözúО™òÿóÕgžkãêàGîwkÂY`[r¼üoÿ¿@iyNIvò»´v¡«-îûA6v8‰nPEMBROKE MAmf÷yÄáð°<|î/K©ÜÌå°ý |,õý{íÝ{
    ü¬ûT}ùr®?ß¿Í:´q„i‡è{úHœÇáÌ^lé³À>﵏¾Ê-º/.»Ý¾¼CWð¬8¸âyb‘“…ÏHª’‚*'¹‚…ÏHª’‚*'¹‚ì°£G‰h?L=ãäpPu9ý%F„TïN»7‰?ßÐ#È¿ §þ‰ßÎ'çú
    ; KW1U de WB9FHP SK

    Click here for info on KW1U: http://wma.arrl.org/stm/

    The association advocating the use of the protocol believes they are in the best position to monitor the system for legality and they say they continually store and monitor the messages for legality, and they're better equipped to do so than the great unwashed and the OO corps. That's really not hard to fathom. Below, is a portion of a letter they wrote to the FCC to allay any concerns about pending petitions:

    Dear Chairman Pai,
    (CC anyone/everyone in authority at FCC)

    "I am writing on behalf of the Board of Directors of the Amateur Radio Safety Foundation, Inc. We govern the non-profit organization that
    supports the Winlink Global Radio Email® system. We are deeply concerned that the Commission may succumb to an internet and social
    media campaign led by Theodore Rappaport, resulting in a multitude of comments that echo false and misleading technical points, driven
    by highly emotional arguments about “national security, crime and terrorism.”

    "In our attachment we show that within a system like Winlink, our numerous, interested and knowledgeable participants—with the help of
    smart automation—are doing a better job of monitoring and inspecting hard-to-intercept transmissions than if less knowledgeable
    operators participated on-air. Over 2500 active, experienced and licensed Winlink sysops inspect their station's daily traffic, and they
    outnumber the ARRL's volunteer monitors plus they exclusively focus upon the digital band segments we use. Further, administrators and
    sysops employ powerful tools to search and sort thousands of messages for inspection. All communications are logged in detail and
    messages are archived, and available to the FCC and anyone else on request. Similar logging and archiving are generally employed by
    other digital systems we're aware of."

    "It should be noted that within Winlink, station IDs, message and transaction information are always sent in clear text and can be
    intercepted easily by laymen on-air using the same equipment and software the sender and receiver use."

    "Theodore Rappaport and the opponents he informs offer an emotional, layman's conjecture in their assertions that hard-to-monitor,
    advanced digital protocols used in the amateur radio service will encourage crime, terrorism, and are a threat to national security. They
    clearly do not know or appreciate what monitoring and inspection routinely occurs, and are thus not qualified to judge. Rappaport himself
    has never held a Winlink account or ever used one to learn, or rationally evaluate it. "

    (The letter is much longer, but that's the gist of it).

    Loring A. Kutchins, W3QA
    Amateur Radio Safety Foundation, Inc.

  2. W6RZ

    W6RZ Premium Subscriber QRZ Page

    Incorrect. Pactor 4 requires the same bandwidth as Pactor 3.

    From the NPRM:

    "For example, ARRL points out that PACTOR 3, which has a data rate of up to 3600 bits per second and a symbol rate of 100 bauds, is permitted in the HF bands; but PACTOR 4, which is capable of a data rate of 5800 bits per second without occupying any more spectrum, is prohibited at HF by the current rules because it has a symbol rate of 1800 bauds."

    The difference is that Pactor 3 uses a multiple carrier OFDM modem while Pactor 4 uses a single carrier modem. OFDM renders the baud rate limitation completely ineffective as a way to limit bandwidth. With enough carriers, you can have an 8 MHz wide signal and still be under 300 baud.

    Eliminating the baud rate limitation is technically correct. It's a shame that it's been conflated with Winlink and HF e-mail.
    N8NOE and KX4O like this.
  3. K0IDT

    K0IDT Ham Member QRZ Page

    Not directly. The RMS sysops have to log on to the master Winlink database to see message content, nothing locally, in real time, and they have the equipment.
  4. K0IDT

    K0IDT Ham Member QRZ Page

    ARRL driven by Winlink to legalize Pactor 4 is responsible for the conflation. Yes, the baud rate limit is silly and should be removed, but if you look at the very early comments on RM-11708 it's plain to see Winlink had a hand in writing
    the RM. The early comments are "we want P4", nothing about experimenting or what other protocols are being "held back". Check the time stamps on the early comments, sort of looks like Winlink had a heads up on the timing.
    NL7W likes this.
  5. N1FM

    N1FM Premium Subscriber QRZ Page

    JUNE 7, 2004
    Concerned hams question amateur radio choice of closed code and controller
    Author: Jay Lyman

    "Yet another issue arises from the fact that -- as is often the case with proprietary solutions and people's positions -- two of the ARRL committee members are also on the Winlink Development Team."

    "The ARRL ARESCOM is a work-in-process and I have been asked not to provide details until they have been made officially public by the ARRL," said the ARRL's Steve Waterman, who is also a Winlink 2000 developer. "I can tell you that the objective is meant to provide seamless, end-to-end email from the desktops of the offices in the served agencies with no invasive software behind the agency firewall. It is meant to do this using the existing email agents on the desktops. Further the objective is to accomplish this with no Internet available," Waterman added. "Incidentally, Pactor plays a minor role, if any at all, in this scenario."


    The absolute bandwidth of Pactor-3 is simply described as "2.4 kHz." The manufacturer abandons that simple description when describing Pactor-4.

    "PACTOR-4 reaches a higher data throughput than PACTOR-3 over the entire working range from -20 to +25 dB @ 4 KHz."


    "In the modulator, the 1800 Bd pulses are shaped with a pseudo-RC filter and finally shifted in frequency by +1500 Hz. The center frequency of the (real) audio signal of the DR-7800 modem is 1500 Hz. The bandwidth of the signal is approximately 2400 Hz (@ -25 dB)."

    Assuming the signal is 25 dB down at the edges of 2400 Hz, the bandwidth scheme of P4 is still a bit perplexing, because the manufacturer describes the center of the audio frequency of the transmitted pulses as 1,500 Hz. If the absolute bandwidth is "approximately" 2,400 Hz, as stated in their spec sheet, then the first pulse is at 300 Hz with the center at 1,500 Hz, and the last pulse is at 2,700 Hz = 2,400 Hz total, suggesting they're measuring audio bandwidth in a 3,000 Hz filter.

    The pulse begins at 300; add 1200 to get to the middle = 1500, and; add 1200 more to get to the last pulse, at 2700 Hz. 1,200 + 1,200 = 2,400 Hz occupied bandwidth within a 3,000 Hz filter window.

    If transmitted with a typical ham radio, in USB, (in theory) you'd need a 2.7 filter width (minimum) at the upper end, to accommodate the entire transmitted and received signal. Many ham rigs have filters that begin at about 100 Hz and end at about 3,000 Hz. In practice, that means a frequency readout of 3.590 USB will actually be transmitting from 3.590.3 to 3.592.7 mHz.

    If the filter width of the rig's receiver is 3 kHz (typical) then the receiver will hear approximately 3.590 to 3.593 in USB. While, technically, the occupied transmit bandwidth is 2,400 Hz, in practice, a strong signal in the first or last 300 Hz of that 3 kHz receive passband, could harm reception. However, the manufacturer has compensated for that potential, with a high degree of compression. Apparently that's why the pulses are only 25 dB down at the edges.

    "The chirp mode is robust against narrow band and part-channel interference, as well as against selective fading. Together with a much improved method of memory-ARQ it allows a link to continue up to a SNR of -20 dB @ 4 kHz. Special importance was given during the development of PACTOR-4 that this could be used without problem on industry standard transceivers with a 2.4 kHz IF bandwidth (”amateur radio transceiver”). Due to the adaptive equalizer, the form of the IF filter curve (as compared against OFDM, PACTOR-3) is non critical. PACTOR-4 requires only slightly more SNR in order to equalize even ”heavily bent” IF filter curves. P4dragon modems use a very complex newly developed synchronization algorithm for the link initialization (both ”normal” and ”robust”). This allows that on receipt of a ”connect” request from a calling station, an immediate and loss free automatic frequency compensation of up to ±280 Hz! can be applied to the receiver."

    Add another +/-280 Hz in automatic frequency compensation for the Dragon 4. If your rig uses a 3 kHz filter, the P4 modem will automatically use that +/- 280 Hz signal seeking technology to find the other station's transmitted signal. This seems to be a kind of automatic frequency control. So while the transmit bandwidth may be approximately 2.4 kHz, the P4 modem will seek another station's transmission, either 280 Hz above the expected frequency, or below the expected frequency. In this way, you can see how two stations could occupy a full 3 kHz typical filter width, if they are off the expected freq. by plus or minus 280 Hz.

    https://www.p4dragon.com/download/PACTOR-4 Protocol.pdf

    Ted Rappaport (as quoted below) states:

    "Current amateur radio activity by some of these data advocates that run Winlink email servers result in daily violations of Part 97 rules that are not being enforced, and this problem has never been addressed by ARRL or Mr. Waterman. This has caused some to speculate that there is a hidden agenda based on monetary interests, or some other secret motivation for the ARRL, Steve Waterman, and Winlink-associated groups to continue their pursuits for wideband digital access in the very limited amateur radio HF spectrum, when much greater bandwidths and more suitable spectrum for such experimentation and usage exists in the VHF and UHF amateur allocations.

    The Commission should consider the possibility of a hidden agenda behind the positons of ARRL and Steve Waterman. It is worth noting that the famous computer entrepreneur Vic Poor, W5SMM, was a driving force for digital data in HF bands, was a founding member of Winlink, and was a leader in ARRL's original RM-11306 ad-hoc drafting committee. The ARRL presented Mr. Poor with its President's Award, the highest recognition given by ARRL to a member, just a few months prior to Mr. Poor's death in 2012, causing some to conjecture that the continual and relentless digital data pursuits in the HF spectrum by ARRL and Mr. Waterman might be motivated by a contingent bequeath from Mr. Poor's estate, or perhaps there is some monetary connection between ARRL, Winlink and/or SCS, the sole vendor of the proprietary Pactor signaling scheme.

    While this is speculation, it is interesting that ARRL's own press release, made about a year before ARRL filed RM-11708, specifically mentioned Mr. Waterman just as they did in their FCC filing for RM-11306 seven years earlier:

    http://www.arrl.org/news/vic-poor-w5smm-receives-arrl-president-s-award. The Commission may wish to consider asking ARRL and Mr. Waterman to publicly state any conflicts of interest that might be at work for their continued advocacy for wideband email data in the HF amateur radio service, given that VHF and UHF frequencies are widely available for experimentation and development of wideband communications, have national networks of repeaters, and well suited for emergencies."

  6. NL7W

    NL7W Premium Subscriber QRZ Page

    Others' half-truths are the "game." ;)
  7. K7JEM

    K7JEM Ham Member QRZ Page

    It seems to be that anything uttered by Ted is taken as gospel truth, regardless of degree of speculation, while anything claimed by the other side is either bold face lies or terrorist enabling tactics.
  8. NL7W

    NL7W Premium Subscriber QRZ Page

  9. NL7W

    NL7W Premium Subscriber QRZ Page

    Make the compression / decompression algorithm publicly known!

    After 1 to 2 years of publicly available hardware and software releases allowing WINLINK message content monitoring, let's review the ARRL / WINLINK proposal. Let's review after HF emails' message content have been reviewed by the general amateur population.

    Or, unless WINLINK cares to release all of their users' email content (minus headers) over the last 1 or 2 years, I cannot accept WINLINK's claim that Part 97 hasn't been flagrantly broken. IMHO, WINLINK will claim their emails were lost when pressed... if you know what I mean.

    Till then, to hell with them and their request. It plainly stinks to high heaven!
    Last edited: Dec 29, 2018
    K0IDT and N0NB like this.
  10. N0NB

    N0NB Subscriber QRZ Page

    I agree.

    One thing that interests me is the commentary that winlink is allowing boaters to avoid a subscription to Sail Mail. If so, wouldn't the Sail Mail people be involved in this and opposing a further erosion to their revenue base?

    Another thing, is that the winlink people would probably do well to rename their product. These days it recalls the bad old days of win modems and win printers and other such MS Windows only hardware. It sounds so cutting edge late '90s tech!

Share This Page