[net.ham-radio.packet] Gateway V.1 No.20

jak@mtunh.UUCP (Jim Kutsch) (06/03/85)

Gateway: The ARRL Packet-Radio Newsletter
 Volume 1, Issue 20
 May 21, 1985

 Published by:
 ARRL
 225 Main Street
 Newington, CT 06111
 203-666-1541

 Editor:
 Jeffrey W. Ward, K8KA



 CALIFORNIA PACKETEERS PARTICIPATE IN "QUAKE '85"

 A simulated earthquake occurred on April 16 in southern
 California.   The simulated quake was 8.2 on the Richter Scale.
 A quake of this size would cause a great deal of destruction and
 would most likely result in the immediate loss of most
 communications over a large area for several hours.  As is the
 case for most large-scale exercises of this type, amateur radio
 played a part.  This year, for the first time, packet radio made
 major contributions.

 The goal was an ambitious one: to move traffic from the State
 Office of Emergency Services (OES) communications trailer in Los
 Alamitos to the OES office in Sacramento, 400 miles to the north,
 using VHF.  This required the use of seven digipeaters.  We
 learned several things during the exercise:  First, it is
 possible to move a large amount of traffic that distance through
 that many digipeaters, and second, it wasn't possible to do it in
 the way we had originally intended.

 We also had several unplanned events preceding and during the
 exercise which added to the simulated emergencies.  First,
 California is blessed with a geography that provides 4000 to 8000
 foot mountains and over-water paths.  The length of most of the
 paths used in the network (called WESTNET) is 90 miles, with one
 path of 120 miles and another close to 200 miles.  The longer
 paths require over-water ducts, which are in place for much of
 the year.  Two days before the exercise, a weather pattern went
 through that destroyed the ducts, which didn't return for several
 days.

 Also, three digipeaters failed the day before the exercise.  This
 was the largest network failure experienced before or since.  The
 systems were repaired in hours, and portable systems were
 readied.  During the exercise, one system was driven to a
 mountain top to supply backup for the lost duct, and another
 system was "car mobile" at the OES site, ready to be driven or
 helicoptered should the need arise.

 In addition to the amateur-owned equipment in place, the OES had
 previously purchased several TNCs; one was installed in their
 communications van in Los Alamitos and another in their office in
 Sacramento.  A third was at an intermediate digipeater site in
 northern California.  To make a long story short, soon after the
 exercise began, we established communications with Sacramento.
 Unfortunately, it was lost soon after, because of a flaky path at
 the far northern end.  We didn't find this out until much later,
 of course.  We interspersed an hour of attempted reconnects up
 north with passing traffic around the southern California area -
 San Diego, Santa Barbara, Glendale and others.  This activity was
 very successful.  After it became apparent that we couldn't stay
 connected to Sacramento, the goal became to get the traffic as
 close to there as possible.  We sent several CQ messages to San
 Francisco, which was outside of the simulated affected area, and
 attracted the attention of a ham with file store capabilities.
 We transferred two hours of accumulated traffic in short order to
 San Francisco, through four digipeater hops.  The San Francisco
 station then transferred the traffic to the Sacramento station
 though two hops, again with little difficulty.  Later in the
 afternoon as the paths improved, we were able to transfer traffic
 directly to Sacramento.

 We learned these lessons during the test: First, two teams are
 needed at the packet position in the communications trailer.  One
 team establishes and maintains the link, the other enters the
 traffic on a separate computer system.  When a link is available,
 the data can be moved quickly by moving a disk from the data-
 entry computer to the computer with the TNC attached.  During the
 "quake," even though two computers were available, message
 traffic piled up in the "IN" basket while we were trying to
 establish a path.  Once contact was made with the San Francisco
 station, it took half an hour to clear the backlog.  It would
 have taken less than five minutes had the data already been
 entered on disk.

 We also learned that it would have been better if we had planned
 on an intermediate file relay station part way up the link in the
 first place.  The cumulative effects of dropped packets increases
 with number of digipeaters until the probability of a packet
 making it all the way to the end and the ACK getting all the way
 back becomes small.  In our case, four hops was 90% reliable (one
 packet not ACKed in 10), but seven hops was less that 10%
 reliable.  Had we sent all traffic with to the intermediate
 station in San Francisco, we could have passed the traffic as it
 came in, instead of building up a large backlog.

 Finally, we learned that it was possible to pass traffic a long
 distance through a large number of digipeaters, provided that we
 did it the right way.  It was possible to maintain a clear
 channel; people with non-simulated-emergency traffic stood by or
 went elsewhere.  We also learned that as with other types of
 emergency exercise, success depends on planning and on the hard
 work of many amateurs.

 At the risk of leaving someone out, here is a list of the packet
 participants in "Quake 85" -- keyboarders, mountain-toppers, HQ
 and other office station setup, north and south:  KA6SOX, W1UUQ,
 WA6CFM, WD6FPY, WB6DAO, W6IXU, N6BGW, N6CXB, NK6K, AJ6T, WB6UCK,
 N6ZH, NG6P, K6QIF, KB6JM, WB6HHV, W6AMT.  There were 6 packet
 stations directly involved with traffic handling and 10
 digipeaters involved in the network.

     From NK6K.


 IBM PC PACKET ADAPTER

 Jack Botner, VE3LNY, whose article "A Packet Radio Adapter for
 the IBM PC" appeared in the January 1985 issue of QEX, has
 received many inquiries concerning software for his 8273 HDLC
 adapter.  He reports that an implementation of the VADCG protocol
 is available from the Hamilton Area Packet Network (HAPN).  The
 protocol software is written in assembler, and the support
 programs are written in C.  To receive the software, send $15, a
 diskette and a disk mailer to:

     Jack Botner, VE3LNY
     35 Wynford Hights Circle #1708
     Don Mills, Ontario
     CANADA M3C 1L1

 Software supporting the AX.25 protocol is not yet available for
 Jack's adapter.  HAPN would like to hear from anyone interested
 in developing AX.25 software for the interface.

     From VE3LNY.


 COLLECTION OF PACKET PAPERS

 If you have been reading any of the books recommended in Gateway,
 you have probably noticed that a few "classic" papers on packet
 switching are referred to over and over.  Tutorial Principles of
 Communication and Networking Protocols, edited by Simon S. Lam is
 a comprehensive collection that includes many of these essential
 papers on digital communications.  The papers cover link
 protocols, multiple access methods, local-area networks, resource
 allocation in networks, network and internet services, and
 protocol verification.  The collection is published by The
 Institute of Electrical and Electronics Engineers (IEEE), and it
 is ISBN 0-8186-0582-0.

     Via KA9Q.


 PACSAT FUNDING

 As reported in Gateway 15, the Radio Amateur Satellite
 Corporation (AMSAT) and The Volunteers in Technical Assistance
 (VITA), have completed engineering specifications for PACSAT, a
 proposed store-and-forward, packet-radio satellite.  In the two
 months since the design and strategy meeting, no sources of the
 necessary funding have been identified.  Recently, VITA was
 granted several thousand dollars to retain a professional fund
 raiser for the PACSAT project.  While this should help, it is
 still important that amateurs assist by identifying potential
 sources of major contributions.  Also, as in past amateur-
 satellite projects, individual and club donations to the PACSAT
 fund will be necessary.  If you can help, contact:

     VITA
     1815 N. Lynn
     Arlington, VA 22209
     Attn: Gary Garriott, PACSAT Project Manager

     Ed.


 JAS-1 PACKET INTERFACE
 The  April issue of the Japanese magazine CQ Ham Radio has an
 article by  JA1ANG, based on material from  Miki  Nakayama,
 JR1SWB, describing the user interface to the packet mailbox that
 will fly on the JAS-1 satellite.  It shows a sample message  from
 W3IWI to JR1SWB as it would be received at JR1SWB and then a
 reply.  The dialog shown in  the magazine is reconstructed below:

 (Legend in left-hand column:
 $: JAS-1 ==> User
 %: User  ==> JAS-1
 &: TNC  <==> User)

 & cmd:CONNECT JAS-1
 & *** CONNECTED to JAS-1

 $ Hello JR1SWB de JAS-1
 $ Your last access was on 85/01/11 at 1922 UTC

 $ NO. DATE   FROM   TO      SUBJECT         LINES
 $ 15  01/28  W3IWI  JR1SWB  AX.25 SOFTWARE  3

 $ Command ?

 % READ 15

 $ Posted: Mon 85/01/28 From:   W3IWI
 $ To:     JR1SWB
 $ Subj:   AX.25 SOFTWARE
 $ I have the disks ready to mail to you and the
 $ packet is all sealed. $    Tom

 $ COMMAND ?

 % WRITE W3IWI

 $ Subject:

 % Re: AX.25 SOFT.
 $ Text:

 % Tom, I'm looking forward to receiving the
 % packet.
 %    73's  Miki
 % .

 $ Message from JR1SWB to W3IWI saved as #18
 $ COMMAND?

 % BYE

 & *** DISCONNECTED

 This protocol is a combination of features of  the W0RLI MailBox
 (JR1SWB  has  a Xerox 820 and is running the MailBox) and  of
 TeleMail (the message in the example was sent via TeleMail).
 JAS-1 is scheduled for launch in 1986.  Uplinks will be on 2
 meters (145.85, .87, .89, .91 MHz) with a single downlink on 70
 cm (435.91 MHz) using 1200 bauds, phase-shift keyed (PSK).

     From W3IWI.


 SIGHTLESS HAM ON PACKET RADIO

 April 25, 1985 marked an important event for the Rochester
 packet-radio fraternity.  1985 has seen rapid growth of packet in
 the Rochester area, but the highlight has to be the appearance on
 packet of Walt Keleher, KA2ASL.  What's so unusual?  Walt is
 blind.

 Operating on voice or CW without sight presents some challenge,
 but think about operating on packet radio without your sight.
 Sending is not a big problem, since Walt can touch type, but
 receiving is another story. Walt has an Apple IIe computer, an
 Echo voice synthesizer and a specially developed terminal
 program. This terminal program unites the computer/terminal and
 the text-to-speech functions of the synthesizer or "talker."
 Whatever appears on the screen of the computer is "read" by the
 talker, in nearly perfect speech. If pronunciation errors are
 made, Walt can back up to get a repeat or to have the unit spell
 out the unknown word or phrase.

 On the evening of April 25, Dave Denz, N2DWL, and Mark Winrock,
 both members of the Rochester packet group, went to Walt's home
 and set up his TNC and radio.  After a short period of
 instruction and coaching, the big moment arrived.  The equipment
 worked perfectly.  Walt worked WB2NBU, K2YNW and W2DUC in rapid
 succession. While Walt was taking his time learning about packet
 operation, the rest of the gang was taking turns sending to him.
 Walt reports that we are perfect copy, but all seem to have a
 Swedish accent (a common trait of text-to-voice processors).

 It was exciting evening, introducing Walt to packet radio and
 helping willpower overcome a great barrier.

     From W2DUC.


 CONTACTS IN THE SOUTH

 At a recent packet-radio forum at the Baton Rouge, Louisiana
 hamfest, several amateurs from Louisiana and Mississippi
 expressed the need for better communications among the scattered
 centers of packet radio in the southern U.S.  If you are
 interested in helping connect the western reaches of "SOUTHNET,"
 write:

     Ken Shutt, K5GUU
     12433 Archery Dr.
     Baton Rouge, LA 70815

 or

     Alan Clark, WD5IKD
     2325 Milam Street
     Pearl, MS 39208.

 In areas where packet stations are few and far between, it is
 important that those who are on the air or interested in getting
 started can get in touch with others who've been bitten by the
 packet bug.

     Via WD5IKD, K5GUU.


 INTERMOUNTAIN REPORT

 Dave Pedersen, N7BHC, of the Utah Packet Radio Association
 (UPRA), recently gave a packet-radio introduction and
 demonstration to more than thirty people in Boise, Idaho.  There
 are several packet stations under construction in Boise.  Dave,
 located in Salt Lake City, Utah,  reports that he has been able
 to make a 300-mile direct connection with Jeff Bishop, W7ID, in
 Boise.  Unfortunately, the path is not quite solid on 2-meter FM.
 Dave and Jeff regularly hold 2-meter SSB skeds, and they hope to
 try some SSB packet (real FSK instead of AFSK on FM) soon.  A
 digipeater link between Salt Lake City and Boise should be
 completed this summer.

    From N7BHC.


 HF PACKET

 If your VHF packet network is getting a bit crowded or you just
 want to try out another facet of packet radio, connect your TNC
 to your HF rig and give HF packet a try.  HF packet stations
 transmit at 300 bauds, using a 200-Hz shift.  While some stations
 are using Bell 103 modems modified to transmit and receive on one
 set of tones (simplex), any 200-Hz shift modem can be used.  If
 your transmit tones are the same as your receive tones, just tune
 in another station and you're on your way.  Packet-radio employs
 NRZI encoding, so which sideband you use is not important.  Most
 of the HF packet activity is on 30 and 20 meters, with 10.147 MHz
 and 14.102 MHz the popular frequencies.  The following stations
 are running the W0RLI GateWay software on HF packet:  KI0Q, Ames,
 Iowa; N4CI, Conyers, Georgia; KF4JF, Hahira, Georgia; K7PYK,
 Scottsdale, Arizona; WA4SZK, Florence, South Carolina; W0RLI,
 Westport, Massachusetts; KD6SQ, Cucamonga, California; KA1PN-1,
 Nashua, New Hampshire.  Hank Oredson, W0RLI, reports that a new
 GateWay station comes on the air each week.  So, slow down your
 TNC and narrow your modem to try HF packet.

     Via W0RLI.



 TAPR "OLD BUSINESS DAYS"

 From Monday, May 20, through Friday, May 31, 9:00 AM until 5:30
 PM PST, TAPR will double its office staff and disconnect the
 telephone recorder.  During this time they will have a human
 answering the phone, and they will be returning the backlog of
 calls that has built up.

 What exactly is  "Old Business Days"?  It's a marathon session
 during which TAPR hopes to clear out the last loose ends of the
 TNC-1 product line.  Call if you have any outstanding business
 with TAPR regarding any of the following items:

     TNC-1 Kits
     TNC-1 cabinet kits.
     Beta Upgrade kits.
     Outstanding spare parts orders.

 Also take this opportunity to order spare or replacement parts
 for TNC-1.

 If you are calling about a backlog order, please have the date of
 the transaction available and information on the method of
 payment.  If you call after hours, you will get the answering
 machine;  Please leave a number where you can be reached during
 the day.

 Please do not call about any of the following items:

     Orders for TNC-2
     New TNC-1 kit or cabinet orders
     9600-bit/s modems
     TNC software or hardware questions

 Both TNC-2 and the 9600-bit/s modem boards are in beta test now.
 TAPR will begin taking orders when they are available.

 The TAPR phone number is (602) 746-1166.

     From NK6K.


 REPRODUCTION OF GATEWAY MATERIAL

 Material may be exerpted from Gateway without prior permission,
 provided that the original contributor is credited and Gateway is
 identified as the source.