Internet to Amateur Radio Gateway


John Watrous, Art Department Chair, 14Mar95

A gateway, using our existing Internet service to SRJC is possible, would cost the District nothing and could provide a number of interesting new pluses for the College. I see an Internet to Amateur Radio Gateway doing these things:

1. Furthering Interconectivity


Internet to Amateur Radio packet gateways already exist. In fact I am seeking this SRJC gateway because we have recently lost our local gateway at the US Coast Guard Training Facility in Two Rock. Many colleges and universities worldwide now provide such gateways and some are now in operation at Stanford and the University of Utah. Also, UCSD is providing top level name servers and routing hosts. I have a map of gateways throughout the world. Only amateur radio operators licensed by the FCC can use the gateway and their license is in jeopardy if they fail to follow the strict rules as specified by the FCC. The critical point is that amateurs care even more than SRJC about proper usage. Amateur radio licenses stress emergency preparedness and the conditions forbid any commercial uses. (see footnote)

2. Providing a Tool for Research


We have a unique resource here in Sonoma County. Due to the initial efforts and expertise of Glenn Elmore who works at HP, a small group of amateur radio operators are experimenting with very high speed radio frequency equipment which has the potential for new and interesting wireless applications such as real time interactive multi-media. The radio side of wireless network and applications development has not gotten the same attention as the computer side; it requires a greater degree of physical proximity to develop and that is part of what SRJC can help with. In addition, a Web page as 'common ground' for aiding this development worldwide is planned and would be a contribution.

Perhaps no where else in the world could a gateway provide as effective a research tool for non-commercial wireless networking at this time. The proposed Internet gateway is a part of a Northern California High Speed Digital Radio Frequency Network program currently being developed by several amateur radio operators. The local group includes an engineer and network specialist from Hewlett Packard, an independent technician (and pilot) from Seaview and a retired Pacific Bell technician in the East Bay. Keep in mind that all participants are volunteers, have independent Internet accounts for mail and Web browsing and experimentation normally takes place after work hours.

Although the gateway combined with the physical proximity of those developing the wireless network are what primarily enable this research, there is significant potential for worldwide involvement in these endeavors. In fact, some of the possible applications being considered would best be fulfilled this way. Not only would other gateways be involved, but individuals in many countries with a variety of expertise could participate. (see end paper)

3. Public Relations


Like the new SAREX station, a gateway will also attract a younger audience interested in science and technology who will see the SRJC logo and later make the connection when choosing a college. The PR value will extend far beyond those who directly use the gateway. The SRJC-gateway will be printed on special maps and the college will be seen as technologically innovative. Another possibility may be to move some existing Amateur Radio Networking Web pages to the gateway server to further link the gateway to SRJC. We would like to move Glenn Elmore's Web pages which are currently averaging a 4 Mbytes/day down load rate at it's current location, once a suitable server is available. If this rate were to continue here at SRJC it represents many new connections with technologically interested people to SRJC.

4. Cost


The Internet to Amateur Radio Gateway is possible at no cost to the District other than a very slight increase in network load (only 2 people presently have 230kbps access, others have 9600bps access--a lower speed and therefore less bandwidth used than most current dial ins), and small initial physical network connection. Preferably a direct, T1 connection would allow applications which require the lower latency and higher speed possible to operate. Interactive applications, time critical ones (wide area earthquake predictive applications, for example), images, etc.

I will provide the necessary equipment and maintain it; a computer, small radio and radio frequency modem and antenna (which is about half the size of a TV antenna). The equipment needs to be housed in my office so that I can interact with the software. The antenna needs to be located nearby, on Analy Hall and would be fairly inconspicuous.

Footnote
Excerpts from FCC Rules Regarding Amateur Service

97.1 Basis and purpose.
The rules and regulations in this Part are designed to provide an amateur radio service having a fundamental purpose as expressed in the following principles:

(a) Recognition and enhancement of the value of the amateur service to the public as a voluntary noncommercial communication service, particularly with respect to providing emergency communications.
(b) Continuation and extension of the amateur's proven ability to contribute to the advancement of the radio art.
(c) Encouragement and improvement of the amateur service through rules which provide for advancing skills in both the communications and technical phases of the art.
(d) Expansion of the existing reservoir within the amateur radio service of trained operators, technicians, and electronics experts.
(e) Continuation and extension of the amateur's unique ability to enhance international goodwill.

End Paper


A Little History of a Larger Development Relating to the Gateway>
a private note by Glenn Elmore

The Internet and it's protocols grew out of (D)ARPA research and activities more than 20 years ago. There was the need for interconnecting diverse computer systems of military, educational institutions and industry so that they could all collaborate.
The Internet Protocol (IP), which is actually a 'stack' of protocols, resulted from this common need and joint effort. Somehow, in spite of the defense/military presence, a rather open and evolving research and development environment resulted. Needs and requirements were discussed and then Requests For Comment (RFCs) were submitted by the physically diverse members as a sort of suggestion as to how/what a new protocol should be. These RFCs were made generally available, haggled over, implemented and changed until they served. This process has continued until the present and is the primary means for improvement.
As with most developments, a relatively small number of experts emerged in the various areas of endeavor. Some wrangled out lower level issues. The Ethernet protocol started at DEC, I think, and was later codified into IEEE 802.3 after a few changes (Ethernet and 802.3 aren't identical but similar). Similarly, higher layer protocols, like FTP, Telnet, SMTP which were closer to the users need were also proposed, improved and implemented, tested, improved ...
All this made a great "playground" for academic types while also providing useful results. The Internet began to emerge out of an assemblage of interconnected subnets. ARPAnet, MILnet and others I think were early but others were added and the Internet rapidly grew to be what it is today. Somewhere along the way, the Internet Engineering Task Force was created. It was more or less the 'group of experts' involved in the above. Somewhere along in the early 1980's not long after TAPR introduced the TNC and amateur packet began, a number of folks, some of whom were involved in the IETF and Internet development and were also hams, thought it would be interesting and worthwhile to run IP over amateur radio. This very much fit with amateur radio's charter and history as a development ground for new technologies. Phil Karn, one of those involved in Internet protocol development (he has his name on several RFCs I think as well as perhaps more than one implemented and active protocol, VanJacobson/Karn being a combination I remember) is also KA9Q and wrote some code to run on a readily available CPU board and also interface to TAPR TNCs. (I think I have this more or less right, don't hold me to it).
There are probably a dozen names on TAPR's WWW page of "pioneers" in packet. I can't recite them, but they include Phil Karn KA9Q , Bdale Garbee N3EUA (at HP in Colorado), Mike Chepponis K3MC (Freemont, CA, old roommate of Bdale's who got him to take the ham test), Harold Price NK6K and others.
In addition to these there are many amateurs worldwide involved professionally with networking and the Internet who are also active in developing and researching wireless networking with amateur radio.
Brian Kantor at UCSD (major sysop, as far as I can tell, also card carrying Hell's Angel!) started a distribution list in which amateurs could further develop IP as applied to radio. There had been some previous work in wireless networking but the bulk of Internet development used existing connections (phone lines and the like) rather than radio. The University of Hawaii is noted for some of it's early wireless work. The ALOHA environment came from them. (ALOHA has to do with a node shouting "hello" rather than with UofH per se, I think) This distribution list, tcp-group@ucsd.edu, now has many thousands of readers worldwide, largely but not entirely amateur. Phil's original "KA9Q" code has been ported to a very wide variety of hosts and is used throughout the world. It's free to amateurs but must be licensed by others. HP here in Santa Rosa is one of the many who have so licensed it. The myriad of adaptations of Phil's original code have shown up as PA0GRI, JNOS, TNOS, MFNOS etc., etc. and are currently being jointly developed worldwide. It really is a widespread development project and many of the participants are "major networking types", many of whom are located at Universities around the world.
Along with the collaboration among amateurs using the Internet itself, development was further fostered when the American Radio Relay League got interested in networking and early in 1980's started the (then) ARRL Computer Networking Conferences. These are annual conferences which accept papers on a wide variety of wireless networking topics and publish proceedings. These proceedings have filled something of the same role in amateur radio networking as RFCs have in network development.
Interestingly, all though it is primarily an amateur radio group, most of the contributions (starting with Phil's code itself) have been on the computer and software side. Not much fundamentally new has been added in the radio/hardware side. This is in part due to the difficulty of collaborating across states/countries/continents on hardware. Software is easier to clone and probably easier to test and measure for the vast majority.
I came along about 1986 or so and started to get interested in first packet and then in IP. I was rather unique in the development group in having primarily RF/microwave experience rather than a CS background. I stumbled into packet/IP after having successfully led 'projects' to build a TVRO system from scratch, a group project to build an EME station and another group endeavor to break the 10GHZ terrestrial DX record. All of these had aspects of microwave design and signal propagation which prove relevant to high speed data flow over a wireless network.
It didn't take long to discover that amateur networking was severely hampered by the lack of suitable RF development. I looked at the fundamentals and produced my "teaser"; the 10 GHz 2 MBps microwave data link made from door-opener transceivers. Kevin Rowett, N6RCE (at Tandem Computers at the time), handled the CS and networking side, I did the RF. This was first published in Ham Radio Magazine in December of 1989 (following the 3 part "Designing a Station for the Microwave Bands" the year before). It was also published in the 8th ARRL CNC proceedings as was a 5 way article (Bdale, Mike, Phil, Kevin and myself) dealing with the general potential for high speed amateur and wireless networks. The ARRL picked it up and has included it in the last several Handbooks as well as the new "Advanced Packet" publication (I just got my complementary copy and you're welcome to browse it, but it's mostly reprints and old news).
The microwave link was meant to expose the fact that orders of magnitude of improvement were possible for no more investment if things were done "right". I followed it up the next year with my paper "Physical Layer Considerations in Building an Amateur Radio Network". This justified the approach based on fundamentals. The same year (CNC #9) as my physical layer paper, Kevin N6RCE, Ed Satterthwaite N6PLO and I presented the "Hubmaster" paper which was a new protocol based on the preceding work and appropriate for the amateur environment. Kevin and I presented a "status" paper in CNC 10 the following year. The existing 904 radios were being developed and deployed along with "MIO", a digital controller Kevin designed and built, at this time.
Kevin got busy and I was left on my own to try to make things work properly. Developing/supporting MIO was more than I was up for alone(having radios to do as well) so I first bought 10 MHz TNCs and put them on with Bob, Dale's help at 38.4 kbps, and then PI2 cards in 386DX40 hosts at 230.4 kbps. I think you came in along about that time and know the rest.

Glenn Elmore
Internet: glenne@sr.hp.com