Minutes of the ARRL Committee on Amateur Radio Digital Communications June 13, 1992 The ARRL Committee on Amateur Radio Digital Communications met at 8:30 CDT on June 13, 1992 at the DFW Marriott Hotel, Dallas, TX. Ed Juge, W5TOO, Chairman presided and Vic Poor, W5SMM acted as recording secretary. In addition the following members were present: Tom Comstock, N5TC, Craig McCartney, WA8DRZ, Paul Newland, AD71, and Dale Sinner, W6IWO. Bob Poirier, KODJ, was unable to attend. Comstock reviewed the role of digital communications in past emergencies including the Mexico City earthquake and hurricane Hugo. Poor reviewed the current state of the art of current and soon to be introduced digital modes and their impact on h.f. spectrum utilization. The Committee as a whole reviewed the responses from the Digital Survey conducted by QST and RTTY Journal. A lengthy discussion followed on all the issues raised in connection with the operation of unattended amateur h.f. digital stations. The recording secretary was directed to summarize these discussions and the unanimously approved recommendations to the ARRL Board in a separate report which is attached as a part of these minutes. Ed Juge, Chairman Vic Poor, Recording Secretary Report and Recommendation to the ARRL Board of Directors by the ARRL Committee on Amateur Radio Digital Communications June 13, 1992 The ARRL Digital Committee has been asked by the ARRL Board of study the issues related to use of automatic unattended control of amateur stations operating digital modes in the h.f. spectrum and to recommend what action the Board should take toward establishing permanent rules for such operation, if any. The Committee has carefully studied as many of the facts and opinions as were available within the Committee's resources. Data bearing on the question included: The results of the ARRL Digit Survey; Frequency usage and allocations in the U.S. and in other countries; The current state of the art for amateur h.f. digital modes; Potential abuse of unattended operation such as illegal third party traffic; The various competing interests for h.f. spectrum, particularly between existing digital modes; and Amateur operating practices and traditions. The ARRL Digital Survey The members of the Committee carefully studied the tallies of answers to the questions in the survey and read every written comment submitted by the respondents. The survey data showed that majority of respondents favored permanent authorization of unattended semi-automatic operation but limiting semi-automatic operation to sub-bands, and a substantial majority did not approve of unattended fully-automatic operation. A wide range of opinions and proposals were made in the comments attached to the survey, all of which were discussed and weighed by the Committee. The important issues raised are discussed below. Frequency Usage and Allocations in the U.S. and other Countries It is no secret that available space is very limited in the h.f. spectrum. Nowhere is that more evident than in the very popular 20 and 40 meter bands. The two oldest modes of operation, voice and c.w., have the lion's share of the spectrum in those bands since they were in heavy use before there were any digital modes. The digital modes have simply "squeezed in the crocks" between already established modes of operation. Since the digital modes have become established they have expanded gradually, a little at a time, primarily into space occupied by c.w. operation. Frequencies near the edges of digital mode operation continue to be shared by both digital and non-digital modes. Outside of the U.S., depending on the ITU region and the rules adopted by various administrations, digital operation for any given mode may not align with practice in this country and it does not seem possible to establish a sub-band plan that could be universally acceptable. It is simply inevitable that any band segment in the h.f. spectrum is going to be shared among differing modes of operation. This is not a new condition on the h.f. bands and has been accommodated for decades. Available Spectrum Space in the H. F. Bands Since all current h.f. band space is actively occupied by one or another mode of operation and since no current class of user is willing to give up space for another, the Committee is operating under the assumption that whatever rules are proposed there will not be a sudden significant change in the way the bands are currently used (at least this Committee is not prepared to make any such recommendation!). The Committee believes that gradual changes will continue to occur but that these changes will be due to natural migration as a larger percentage of amateurs shift to digital from other modes of operation and from one digital mode to another. The respondents to the survey strongly opposed the allocation of sub-bands by rule. The Committee also believes that any attempt to specify by rule sub-bands for a class of digital operation would soon grow obsolete as patterns of operation change, more digital modes are introduced, and more users shift to digital modes. Instead, the Committee believes that the amateur community will need to adjust itself to continued sharing of the spectrum by various modes and that such sharing should be facilitated through the publication by the ARRL of recommended sub-bands for the various modes and that such recommendations should be revised from time to time as operating patterns change. The Committee, as a subsequent action, will propose a revised band plan for consideration by the ARRL. In any case, the h.f. spectrum is severely limited, especially for digital mode operation, and modes of operation that improve spectral efficiency must be strongly encouraged. The Committee will undertake a study proposing, in a subsequent action, voluntary technical standards which can be promoted among amateurs and vendors to significantly improve our current frequency usage. The State of the Art for Amateur HF Digital Operation While the current rules allow considerable latitude in what digital modes the amateur community uses, the actual practice is somewhat limited. Current practice includes "RTTY", a non-error-protected simplex mode, usually using the baudot code; "AMTOR", a partially error-protected half-duplex mode using the baudot code; "packet", an error-protected half-duplex mode using ascii code; and "PACTOR", an error-protected half-duplex mode using ascii code. In addition, a new DSP-based system has been demonstrated but is not yet generally available called "Clover" that is an error-protected full -duplex highly spectrum efficient mode. As currently used all of the above modes require approximately 500 to 1000 Hz. of bandwidth per channel except packet which requires 2000 Hz. per channel. Effective use of that bandwidth is terms of character throughput varies considerably as a function of the protocol used and the channel conditions. Partly because of the requirement for 2000 Hz. of space per channel and partly because of the nature of the AX.25 protocol, the performance figures for packet are the poorest per unit of bandwidth of any of the currently used modes. RTTY and AMTOR are better, and PACTOR is better still. Clover promises to exceed the throughput per unit of bandwidth of any of the above modes. Tolerance to poor channel conditions also varies among the modes with packet having the poorest performance, RTTY next, AMTOR and PACTOR being very much better. Digital techniques for h.f. operation are improving and newer technologies such as PACTOR and Clover promise significant near- term improvements in spectrum utilization, throughput, and performance under difficult h.f. radio conditions. The current rules do not appear to have contemplated these new modes in the h.f. portion of the spectrum and the Committee believes the rules require a modest change to encourage these and other new more effective digital modes and to promote operation in the narrowest possible bandwidth. Potential Abuse of Unattended Operation A few respondents to the Survey expressed opposition to any form of unattended operation because of possible illegal use of amateur bands for unauthorized third-party traffic, commercial purposes, or the support of illegal activities such as drug smuggling. The Committee is not aware of any pattern of such abuse nor does the Committee see any reason why illegal operation is not just as likely to occur directly between two attended stations as any other. The Committee did not consider this factor in making its recommendations. Competing Interests for HF Spectrum Space The most difficult issue the Committee has had to deal with is the demand for spectrum space from the many different classes of users. Many of these users are sharing (somewhat unwillingly) the same space and each would like the others to vacate to other locations. The most critical frequency bands (at the moment!) are 20 and 40 meters. On 20 meters the frequencies above 14,100 kHz. have been traditionally used for DX voice and below 14,100 KHz. for c.w. and data. With the advent of packet, and the STA authorizing unattended packet operation, packet operations began above 14,100 Hz. and has gradually occupied the region of 14,100 to 14,125 Hz. Due in large part to the fact that data is not allowed in this sub- band in some countries, packet operation has also extended downward into the band immediately below 14,100 attracting US operation in this sub-band as well. Non-US voice operators have taken exception to the use of the 14,100-14,125 space and RTTY operators have taken exception to the use of the space below 14,100. On the 40 meters packet operation began in the 7080-7100 Hz. region where traditionally RTTY and AMTOR operators had been active. This has forced the RTTY and AMTOR operations further down into the dismay of c.w. operators. This picture is further complicated by the fact that outside of region 2 data operation must be confined below 7050 kHz. The situation on other bands, especially below 21 mHz., though not as critical as on 20 and 40 meters, have similar conflicts. The informal 'sub-bands' used by the various modes are also somewhat fluid as propagation conditions change and usage shifts from one mode to another. The Committee does not believe that any subdivision of the bands by rule will best serve the amateur community in the long run. It also seems unlikely that any subdivision of the band by mode will work on a world wide basis because of the differences in the rules between regions and between individual administrations. Any subdivision of amateur bands by rule also imposes an unnecessary potential enforcement burden on the FCC. Amateur Operating Practices and Traditions Except in a very few special situations it has long been the tradition (and rule) that one amateur station must not willingly or knowingly interfere with a contract already in progress regardless of the mode of operation or the perceived importance of the communications in progress. It has also been a long standing tradition that no station or group of stations 'own' a frequency. (Frequency 'ownership' has admittedly become a practice on certain v.h.f. frequencies, but this practice has never been established on the h.f. bands and the Committee strongly rejects the concept of doing so now.) On h.f. the use of sub-bands with various classes of operation gravitating to specific locations is largely self regulating simply by virtue of the fact that a station occupying a frequency is not driven off the frequency by deliberate interference by a station operating another mode. (There are always isolated exceptions to this but it is not condoned in the rules or by the vast majority of amateur operators.) As greater numbers of amateurs use a particular mode that part of the band becomes recognized informally as a mode-specific sub-band. There is always a significant overlap in the sub-bands between modes - packet sharing with RTTY, RTTY sharing with AMTOR, AMTOR sharing with c.w., and so on. The greatest conflicts come where the overlapping modes have significantly different bandwidth, i.e., AM vrs. ssb, packet vrs. RTTY. Types of Automatic Operation Two types of automatic digital operation are under consideration for use on the amateur h.f. bands. One is fully-automatic operation where messages are passed between amateur stations without any operator intervention and no operator may need be present at either station. The other is semi-automatic operation where messages are passed between amateur stations with an operator initiating the contact from one of the two stations. Both fully- and semi-automatic operation is permissible today under the rules provided there is a control operator present at both stations. (Stations authorized under the STA may operate unattended.) Digital operation with one station functioning in a semi-automatic mode has long been a practice dating back to the '60s. Fully-Automatic Unattended Operation The proposal to authorize fully-automatic unattended operation represents distinct departure from past practices. A clear majority of the respondents to the survey opposed any fully- automatic operation on the amateur h.f. bands. To authorize fully-automatic operation without restriction, as some of the respondents to the survey advocate, would seriously undermine the fiber of mutual cooperation that h.f. operation requires. The Committee rejects such operation as undesirable on its face. It was also proposed to authorize fully-automatic operation with restrictions, either to the frequencies allowed, to a few privileged stations, or both. The committee saw no purpose in limiting the frequency bands alone since the number of stations that would attempt unattended operation would make the mode and allocated frequency useless to everyone. Limiting the number of participating stations was also rejected by the committee because there was no conceivable way to equitably allocate the privilege to specific stations nor was the committee willing to set aside any portion of the band to stations with special privileges. Fully-automatic operation, by it's very nature is mode-specific and must 'own' the frequency it operates on an cannot be effectively shared by other modes of operation. To authorize fully-automatic operation on the necessary mode- specific sub-bands raises serious problems. There are no likely sub-bands that can be used on a world-wide basis or that will not cause interference to other users under some circumstances. The only mode of operation that is currently a prospect for fully- automatic authorization is packet, based on the AX.25 protocol, using 2 kHz. channel spacing. This mode delivers the poorest performance with respect to spectrum utilization or survivability under adverse propagation conditions of any the digital modes currently in use. j The Committee does not believe that, if a protected mode-specific sub-band is to be authorized, that it should be a mode that is as inefficient in its resource utilization as current packet practice represents. Such an authorization will discourage the development and use of a more suitable mode. Further, the Committee does not believe that these is any service being provided by fully-automatic operation that is not also available by other means without the associated problems of fully- automatic operation. Nor does the Committee know of any reason why packet operation cannot also be operated in semi-automatic mode, thereby eliminating the need for a rule-mandated sub-band. Semi-Automatic Unattended Operation There are many reasons, however, why some form of automatic digital operation is desirable. It permits amateurs to exchange communications when there is a time difference between the operating times available to the two amateurs, and it permits the quick exchange of messages rather than taking air time with long calls and keyboard-to-keyboard operation. (This not a suggestion by the Committee that keyboard-to-keyboard is undesirable but simply that there are many cases where moving messages at machine speeds is more spectrum efficient and makes more frequency time available to direct keyboard operation.) It is very evident that some form of automatic operation is highly desirable when handling NTS and personal messages between amateurs through intermediate stations. This capability forms the very heart of the amateur community's preparedness for emergency service. Respondents to the survey favored semi-automatic unattended operation over those opposed by a two-to-one ratio. The Committee does recognize that there is some potential for interference using a semi-automatic unattended mode even as there is such potential in purely manual modes. However, so long as there is a control operator present at one end of the link, monitoring the progress of an exchange, such interference can be held to a minimum. The benefits of semi-automatic operation outweigh the risk of inadvertent interference. The Committee believes that in view of the long successful history of semi-automatic operation that authorizing unattended semi- automatic operation is in the best interests of the amateur community. RECOMMENDATIONS I. Unattended fully-automatic operation of amateur digital stations should not be authorized below 30 mHz. II. The FCC rules should be amended to allow unattended semi- automatic operation of digital stations on any frequency on which digital modes are authorized. Unattended semi-automatic stations may not initiate a contact, either with another station or via an undirected broadcast. An operator initiating a contact with an unattended station must first ascertain that no interference will be caused to existing communications, and must monitor the progress of communications. If it becomes evident that the communications with an unattended semi-automatic station is interfering with other amateur communications then the link with the semi- automatic station must be discontinued. An unattended semi- automatic station must be equipped with a time-out timer to insure that no signal is transmitted longer than five minutes in the event of the malfunction of control equipment or the loss of contact with the initiating station. Suggested wording for such an amendment is included in the appendix. III. The FCC rules should be amended to allow the use of modem- dependent codes for the purpose of efficient data compression and error control on h.f. radio channels. The bandwidth of such signals should be restricted to 500 Hz, below 28 mHz, and 2000 Hz. between 28.0 and 28.3 mHz The appendix to this report suggests specific wording for the recommended rule change. A station using a modem-dependent code must still comply with 96.119 Station Identification. IV. The League should publish a comprehensive tutorial-style operator's guide for h.f. digital operations clearly defining acceptable operating practices. Such a manual would delineate currently used informal sub-bands for the various modes and styles of operation, and the good operating practices that are required for effective mutual cooperation and coexistence. This Committee will make specific recommendations for the content of this guide. V. The League should publish technical standards or guidelines for the characteristics of signals generated by digital mode stations for the purpose of achieving the best possible use of the h.f. spectrum. QST should be used as a forum to educate that amateur community on the benefits and means of achieving acceptable signal quality and should review the technical characteristics of digital mode products with respect to published standards. This Committee will make specific recommendations for these technical standards. APPENDIX A The following is suggested wording for an addition to Part 97 authorizing unattended semi-automatic digital mode operation. 97.3 Definitions ( ) Unattended Digital Station - A station in the amateur service using an RTTY or data emission that is operated without a control operator present. 97.216 Unattended Digital Station (a) Any amateur station licensed to a holder of a General, Advanced or Amateur Extra Class operation license may be an unattended digital station. (b) An unattended digital station may operate on any frequency below 30 mHz. that is authorized for RTTY or data emission for the class of operator license held. (c) An unattended digital station may only use those RTTY or data emissions authorized by 97.305 and 97.307. (d) No unattended digital station may initiate a contact with another station or may broadcast any undirected signal. (e) The transmitter of an unattended digital station must be equipped with a time-out timer that will insure that no signal is transmitted for longer than five minutes in the event of the malfunction of control equipment or loss of contact with the initiating station. (f) Any amateur operator initiating contact with an unattended digital station must first ascertain that no interference will be caused to existing communications, must be present for the duration of the contact, and must discontinue the contact if it becomes evident that communications with the unattended digital station is interfering with other amateur communications. APPENDIX B To encourage improvements in digital mode communications and especially to improved spectrum utilization on amateur h.f. bands Part 97, 97.307 (f) (3) and 97.307 (f) (4), should read as follows: (3) A RTTY or data emission using a specified code listed in 97.309 (a) of this Part may be transmitted. The symbol rate must not exceed 300 baud, and for frequency-shift keying, the frequency shift between mark and space must not exceed 300 Hz. A RTTY or data emission using an unspecified digital code under the limitations listed in 97.309 (b) of the Part also may be transmitted. If an unspecified digital code is transmitted the authorized bandwidth is 500 Hz. (4) A RTTY or data emission using a specified code listed in 97.309 (a) of this Part may be transmitted. The symbol rate must not exceed 1200 baud, and for frequency-shift keying, the frequency shift between mark and space must not exceed 1 kHz. A RTTY or date emission using an unspecified digital code under the limitations listed in 97.309 (b) of the Part also may be transmitted. If an unspecified digital code is transmitted the authorized bandwidth is 2 kHz.