HAM Radio 1/HamRadio.cdr/packet/amtor/amtor.txt

Text File | 1989-05-02 | 16KB | 317 lines

From:    N3Aia Packet Bbs
To:      All                                      Msg #98, 05-Apr-89 08:21pm
Subject: AMTOR Tutorial - Part 1 of 3

From: N9ANL @ N9GTM
To: All

AMTOR - Part one of a three part series.

Since they were introduced several years ago, many thousands of all-mode data
controllers have been sold. I'm referring to the AEA PK232 and PK64, and the
Kantronics KAM. These units offer a mode of communication called AMTOR. Most
people who purchase one of these units do so primarily for packet use, but are
curious about AMTOR. They would like to try the mode, but hesitate because
they don't understand it. Unfortunately, the manuals that come with the
all-mode units do a very poor job of explaining it. This series will be an
attempt to make AMTOR simple.

First I want to discuss how and for what purpose it was developed, and to
offer a plain-language explanation of how it works.

During the nineteen seventies, the Dutch government developed a system of
digital communication called SITOR. It was designed to provide reliable RTTY
communication under adverse conditions, while still maintaining an extremely
low error-rate. The SITOR protocol was meant primarily for maritime use. It
was so reliable a system that it was soon in use world wide.

In the early 1980's, Peter Martinez, G3PLX, made several minor changes to the
SITOR protocol and called it AMTOR. The FCC authorized its use by U S amateurs
beginning in January of 1983.

AMTOR is different than any digital mode you may have used, in that it is the
only synchronous type of RTTY authorized at the present time. "Synchronous"
means that the two stations in an AMTOR ARQ contact are synchronized (linked)
with each other.

There are three modes in AMTOR. The first is ARQ, an acronym for Automatic
Repeat Request, sometimes called Mode A. It makes the familiar chirp chirp
sound you have often heard. The second mode is FEC, an acronym for Forward
Error Correcting. It is sometimes called Mode B. The sound of an FEC
transmission is similar to that of Baudot RTTY. The third mode is called Mode
L, the Listen mode. It allows an operator to decode an ARQ signal even though
he is not one of the two stations that are linked.

First lets talk about the ARQ mode.

Let me begin by describing what's called the "ARQ Cycle". It is 450
milliseconds in length. It starts with the information sending station
transmitting a data-burst consisting of three characters. Since AMTOR uses a
seven-bit code, and the data-transfer rate is fixed at 100 baud, this
data-burst takes 210 milliseconds. Next there is an 85 millisecond period of
silence, during which the receiving station checks the 3 characters for
validity. If the data-burst passes the check, the receiving station prints it,
and sends back a single character control code. This takes 70 milliseconds.
The control code (an ACK) says, "OK, I got it, send the next burst". If the
data-burst does not pass the check, the receiving station prints nothing and
returns a control code (a NAK) that says "Hey, I didn't get it. Send it
again". The sending station will continue to repeat the data-burst until it
gets an ACK. The control code is followed by another 85 milliseconds of
silence, and the cycle begins again with either the next data-burst if the
information sending station received an ACK, or a repeat of the last
data-burst if it got a NAK.

The AMTOR ARQ mode has often been referred to as error free. That is not quite
true. Errors are possible, but the error rate is so extremely low that it is
VIRTUALLY error free.

The method of error detection used is quite simple. Five of the bits in each
character of the AMTOR code are information bearing bits. The other two are
used for error detection. The five information bearing bits in each character
are identical to the five information bearing bits for that character in the
Baudot code. Each of the two error detection bits can be either a mark or a
space. The code is arranged so that every AMTOR character contains four marks
and three spaces. A three character data-burst then, contains a total of
twelve marks and nine spaces. All the information receiving station must do
therefore, is count the total number of marks and spaces in the data-burst it
received to detect whether or not there has been an error.

While data is not being transferred, the information sending station transmits
idle signals to maintain the link.

Please note that throughout this description of the AMTOR ARQ mode I have
referred to the "information sending station" and the "information receiving
station". This is because when two stations are linked in Mode A, both are
transmitting at intervals, thus it would be incorrect to refer to one as the
"transmitting station" and the other as the "receiving station".

In an ARQ link, the timing is set by the station that initiated the contact.
It is called the "master" station, with the other station being the "slave".

<End Part 1 of 3>

--- FD 2.00
 * Origin: Import From [N3AIA @ N3AIA] Packet BBS - Schaumburg, IL.
(1:115/777.3)


From:    N3Aia Packet Bbs
To:      All                                      Msg #99, 05-Apr-89 08:23pm
Subject: AMTOR Tutorial - Part 2 of 3

From: N9ANL @ N9GTM

AMTOR - Part two of a three part series.

The second mode we will discuss is FEC.

This mode is not synchronous, and the stations involved are not linked, but
they do operate in phase with each other. In order for them to stay in phase,
each FEC transmission is started with several sets of "phasing pairs". These
are repeated at regular intervals during the course of the transmission, so
that the two stations can stay in phase. While no data is being transferred,
idle signals are transmitted to keep the two stations in phase. In Mode B,
each character is transmitted twice, 350 milliseconds apart. The receiving
station prints a character the first time it is received if the mark/space
count is correct. If it was received correctly the first time it was sent it
was printed, and will be ignored the second time it is received. If it was
incorrect the first time it was received, it is ignored, and will be printed
the second time it is sent if it is received correctly.

This method of error detection is much less effective than that used in ARQ,
and the error-rate is considerably higher than it is in that mode. Although
higher than in ARQ, the error-rate is still far lower than it is in other
forms of RTTY.

The third AMTOR mode is called Mode L, or the "Listen" mode.

It allows an operator to print ARQ data-bursts even though he is not linked to
the information sending station. There is no error detection at all in the
Listen mode, and if the station being monitored is asked for repeats (RQ's),
they will be printed.

Before discussing AMTOR operation, I want to explain two conventions that that
are used in all three types of RTTY.

The first is that lower sideband ONLY is used regardless of operating
frequency. Failure to observe this practice will result in reversed tones,
known as an "upside-down" signal.

The second convention regards how the frequency of operation is specified. In
single sideband or CW operation it is simple. If you are on say, 14,100.00 and
you are told to "move to 14,050.00", you just dial up the new frequency.

In RTTY when a frequency is specified, it does not refer to the frequency
shown on your display or dial. It refers to the frequency at which your MARK
TONE signal appears, regardless of what your display reads. This may seem
complicated, but there is a very good reason for it to be done this way.

Before I explain, let me mention two terms that may not be familiar to you.
One is Audio Frequency Shift Keying (AFSK), and the other is Frequency Shift
Keying (FSK).

When operating AFSK, your transmitter is keyed by the PTT line, and you feed
mark and space tones into the microphone jack. Since you will be on the lower
sideband, your mark tone signal will appear to be a carrier 2125 Hertz below
the frequency shown on your display.

When operating FSK, tones are not used. Your transmitter is keyed by the PTT
line, and produces a carrier just as though it was in the CW mode. The FSK
line from your data controller is merely opened and closed. As it opens and
closes, it shifts this carrier by 170 Hertz. In FSK operation, where your mark
tone signal appears depends on the radio you are using. It can appear either
170 HERTZ ABOVE, or DIRECTLY ON the frequency shown on your display.

Remember, the station you are in contact with does not know whether you are
using AFSK or FSK. If you asked him to move to another frequency, specified as
shown on your display, he would need know this. Also, if you are using FSK, he
would need to know whether your particular radio shows the position of the
mark or space tone signal on its display.

Using this convention to specify frequency eliminates all this. An operator
need only know where his own mark tone signal is relative to his display, and
specify his frequency accordingly. The other operator, using this convention,
can then tune him correctly regardless of whether he is using AFSK or FSK.

It is my experience, at least with my equipment, that FSK is the most
desirable method to use. That is because using the 500 Hertz filter eliminates
a great deal of QRM, and I cannot use it in the lower sideband mode. If your
radio allows you to use this filter while in lower sideband, then it really
doesn't make much differance whether you use AFSK or FSK.

Since the two stations in an ARQ contact are synchronized, they must recognize
each other, and the called station must respond ONLY to a call directed
specifically to it. For this reason a SELCAL system is used for
identification. Selcal is an acronym for Selective Call.

The convention is to derive your selcal from the first and the three trailing
letters of your callsign, thus my selcal would be NANL. In the case of a
one-by-two callsign such as W1AW, the first letter is repeated, followed by
the two trailing letters, so its selcal would be WWAW. If the callsign is a
two-by-one, the first letter is repeated twice, followed by the trailing
letter. The selcal of KK9H, for example, would be KKKH. A two-by-two callsign
such as KB9SV, results in a selcal of KBSV. For a two- by-three callsign like
KA9ABC, the selcal would be KABC.

While it is not necessary to follow this convention, since any four letters
can be a selcal, it is a good idea to do so. If you do, then any operator who
knows your callsign can deduce your selcal.

<End Part 2 of 3>

--- FD 2.00
 * Origin: Import From [N3AIA @ N3AIA] Packet BBS - Schaumburg, IL.
(1:115/777.3)


From:    N3Aia Packet Bbs
To:      All                                      Msg #100, 05-Apr-89 08:24pm
Subject: AMTOR Tutorial - Part 3 of 3

From: N9ANL @ N9GTM

AMTOR - Part three of a three part series.

There are three ways to initiate an ARQ contact in AMTOR. You can call CQ, you
can respond to another stations CQ call, or you can call a specific station.

In the first case, always call CQ in the FEC mode, NOT ARQ. When you do this,
be sure to include your selcal. Following your CQ, return to AMTOR Standby and
wait for a reply. If you receive a response to your CQ, it could be in FEC,
although it is much more likely that it will be in ARQ. If the response is in
ARQ, your station will begin the ARQ cycle, and you will see that idle signals
are being received. At this point, the link is established and your station is
ready to receive text. Now the other operator need only open his transmit
buffer, type the text, and it will be sent to you.

If text stops appearing on your screen, just stand by. Your station is
probably requesting repeats because it received a data-burst incorrectly. You
will soon receive the block correctly, and traffic will begin to flow again.

When the other station is ready to receive information from you, he will turn
the link over to you, and it will be your turn to send text to him. He will do
this by sending you a turnover sequence. That is the two characters
"Plus-Question mark" (+?).

You will know when the turnover has been made by the change in rhythm of your
stations transmissions. Before the turnover, your station was sending single
character data-bursts 70 milliseconds in length. When the turnover occurs, it
will begin transmitting three character data- bursts 210 milliseconds in
length.

When the turnover has been made, simply open your transmit buffer and begin
typing text for transmission.

When you are transmitting text in ARQ, you will see it pop onto your screen in
three character blocks, as the blocks are acknowleged by the receiving
station. If this text stops appearing and nothing seems to be happening, it
just means that the receiving station is requesting repeats because it
received the last data-burst your station sent incorrectly. Stand by. The
incorrect block will soon be acknowledged, and traffic will flow again.

When you are ready to receive text again, reverse the link by sending the
turnover sequence (+?).

The second way to initiate a contact is to answer another stations CQ.

When a station calls CQ in FEC, its selcal should be included. If it is not,
you can deduce what it is. To answer the CQ, just start an ARQ transmission.
When you do this, your program will ask for the selcal of the station you wish
to call. When you enter it, the ARQ cycle will start, with your station
transmitting the other stations selcal. You will soon hear the other station
responding with control codes, and you will see an indication that your
station is sending idle signals. Now the link is established. Just open your
transmit buffer and type in text for transmission. When you are ready for the
other stations reply, send the turnover (+?) and reverse the link.

The third way to initiate a contact is to call a specific station in ARQ.

Usually, this will happen when you have been watching an ARQ contact in the
Listen mode, and you want to call one of the stations when they finish. In
this case, since you have been "eavesdropping", you already know the selcal of
the station you want to call. You also know it is on frequency, since the
contact has just been terminated. Call your station by starting an ARQ
transmission just as though you were replying to a CQ call.

There are two control codes used in AMTOR that you should be familiar with.

The first is "Control-C". It is used to force a changeover. If the other
station in an ARQ link is information sending station, and you want to reverse
the link without waiting for it to send the changeover sequence, you can do so
by sending it a Control- C.

The second code is "Control-D". It is used to break an ARQ link. An ARQ
contact should ALWAYS be ended with a Control D.

There are a couple of other things to be aware of.

The T/R switching time of your rig should be on the order of 25 milliseconds
for best ARQ operation. All newer rigs and most older ones easily meet this
standard. Some rigs though, designed before AMTOR came into use, may require
simple modifications. Notably, these are the Kenwood TS930S, and the earlier
ICOM rigs. In any case, if you are in doubt, a call to the manufacturer will
get you any information you need.

Since the ARQ cycle requires your rig to return to full receive sensitivity
quickly, always set your AGC for fast release, or turn it off entirely.

Most AMTOR activity will be found on 20 meters. Look for it around 14,075.00.
You will also find activity around 3,650.00, 7,050.00, and 21,075.00.

I hope I've made AMTOR easier for you to understand than the manual you got
with your data controller did.

If you enjoy a good rag chew, it's lots of fun. Give it a try.

Best 73,
Red, N9ANL
Chicago, Illinois

--- FD 2.00
 * Origin: Import From [N3AIA @ N3AIA] Packet BBS - Schaumburg, IL.
(1:115/777.3)