31.PDF
(
3794 KB
)
Pobierz
Contents
1 Sound Card Modes
1.1 Which Sound Card is Best?
1.2 Sound Card Interfaces
1.3 Transmit Audio Levels
1.4 Sound Card Software
2 Packet Radio
2.1 The Packet TNC
2.2 Talking to a TNC
2.3 TNCs and Radios
2.4 TNC Timing
2.5 Monitoring
2.6 “Connected” vs “Unconnected”
3 The Automatic Packet/Position Reporting
System (APRS)
3.1 Setting Up an APRS Station
3.2 Maps and APRS
3.3 APRS Position Encoders
3.4 APRS Networking Tips
4 PACTOR and WINMOR
5 High Speed Multimedia (HSMM)
6 Automatic Link Establishment (ALE)
7 D-STAR
7.1 What is D-STAR?
7.2 Digital Voice and Low-Speed Data (DV)
7.3 High-Speed Data (DD)
7.4 D-STAR Radios
8 APCO-25
8.1 APCO-25 and Amateur Radio
8.2 The APCO-25 Standard
8.3 APCO-25 “Phases”
9 HF Digital Voice
9.1 AOR and AMBE
9.2 FreeDV
10 EchoLink, IRLP and WIRES-II
10.1 EchoLink
10.2 IRLP
10.3 WIRES-II
11 Glossary of Digital Communication Terms
12 Bibliography and References
5.1 A Basic HSMM Radio Station
5.2 Running High Power
5.3 HSMM Antenna Options
Digital Communications
Amateur digital communication has always been a niche activity, but it has grown substan-
tially over the years. From the end of World War II until the early 1980s,
radioteletype
, better
known as
RTTY
, was
the
Amateur Radio digital mode. If you had visited an amateur RTTY
station prior to about 1977, you probably would have seen a mechanical teletype machine,
complete with rolls of yellow paper. The teletype may have been connected to the transceiver
through an interface known as a
TU
, or
terminal unit
. An oscilloscope would probably have
graced the layout as well, used for proper tuning of the received signal.
When affordable microprocessor technology appeared in the late 1970s, terminal units
evolved as well. Some included self-contained keyboards and video displays, making the
mechanical teletype obsolete. As personal computers evolved, they became perfect companions
for TUs. In this configuration, the PC functioned as a “dumb terminal,” displaying the received
data
from
the TU and sending data
to
the TU for transmission. TUs of this era offered ultra-
sharp receive filters that allowed hams to copy weak signals in the midst of interference.
In the late 1980s, conventional terminal units began to yield to sophisticated micropro-
cessor devices known as
multimode controllers
. As the name suggests, these compact units
handle several different digital modes in one package, typically RTTY, packet, AMTOR and
PACTOR. Like TUs, multimode controllers are stand-alone devices that communicate with
a personal computer acting as a dumb terminal. All of the heavy lifting is being done by the
controller and its self-contained software known as
firmware
.
In the early 1990s, sound cards appeared for personal computers. As sound cards became
more powerful, hams began to realize their potential. With the right software, a sound card
could take received audio directly from the radio and translate it into digital information. The
same sound card could also create various forms of digital audio modulation for transmis-
sion. The first “sound card mode” was PSK31, developed by Peter Martinez, G3PLX. In the
years that followed, sound cards became more powerful and versatile. Hams responded by
developing more new digital modes to take advantage of the advances.
Hardware controllers are still with us, but they are primarily used for modes like packet and
PACTOR that require more processing muscle and precision timing than a typical personal
computer can provide on its own. Other amateur digital modes such as D-STAR depend
on specially designed transceivers that combine the radio hardware with dedicated digital
processing firmware.
In this supplement, Steve Ford,
WB8IMY, discusses the tech-
niques involved in assembling and
configuring station components for
operating on the various HF and
VHF digital modes. Today’s digital
communication choices range from
keyboard based modes like classic
RTTY, packet radio and PSK31,
to digital voice, local high speed
multimedia networks and VHF/UHF
networks linked by the Internet.
Related information may be found
in the
Modulation
and
Digital
Modes
chapters. In previous
editions, this material formed
Chapter 31. Unless otherwise
noted, references to other chap-
ters refer to chapters in the print
version of the
ARRL Handbook
.
1 Sound Card Modes
Sound card technology dominates the amateur HF digital communications world. Although
the term sound
card
is commonly used, the techniques discussed in this section apply to
motherboard-embedded sound chip sets, which are extremely common in computer systems
today, and to external sound processing devices. We’ll use the term “sound card” to refer to
any of these hardware implementations. See
Fig 1
.
The sound card modes in use today include PSK, RTTY, MFSK16, Olivia, Hellschreiber
and MT63. There are also sound-card applications for digital voice, discussed later in this
chapter and for slow-scan TV, discussed in the
Image Communications
supplement.
The 31-baud version of PSK, known as PSK31, is by far the most popular mode for casual HF
digital communication. The popularity of PSK31 notwithstanding, radioteletype (RTTY) still
remains the chief HF mode for contesting and DXing. All other HF digital modes play minor
Digital Communications 1
(A)
Fig 1 — Sound cards some in various shapes
and sizes, but virtually all of them will handle
Amateur Radio sound card modes. At A,
a basic sound card that plugs into a PC
expansion slot. Others, such as B, plug into
a USB port. Some high-end sound devices
use an internal
card with the
processing
circuitry
and an
external
breakout
box with
multiple
input and
output
connectors.
sion from analog to digital isn’t perfect, for
several reasons. Here are some parameters
to consider.
Sample Size
When a sound card takes a sample of
the input voltage, it expresses it as a binary
number with a certain number of bits. This
is the
sample resolution,
or
sample size
. The
sample resolution determines the number of
steps between the smallest and the largest
signal the card can measure. The greater the
number of steps and the smaller they are, the
more precise the samples will be. Larger steps
introduce more
quantization noise
, so a sound
card’s signal-to-noise ratio is limited by the
number of bits of resolution in each sample.
For example, a card taking 8-bit samples mea-
sures only 256 voltage steps and cannot yield
a signal-to-noise ratio better than about 49 dB.
With 16-bit samples, there are 65,536 steps,
and the ideal S/N rises to 98 dB.
(B)
roles, but each has characteristics that provide
benefits depending on the use case. Olivia,
MFSK16 and MT63, for example, provide
more robust copy under poor conditions.
On VHF and above, the
WSJT
software
suite is the sound card mode of choice for
meteor scatter and moonbounce work.
WSJT
has also found experimental application on
the HF bands. See the sidebar, “The WSJT
Revolution.”
Regardless of the mode in question, the
sound card functions as the critical link. It
is put to work as a digital-to-analog (D/A)
and analog-to-digital (A/D) converter. In its
A/D role, the sound card takes receiver audio
and converts it to digital information. Dur-
ing transmission, the sound card is used as
a D/A converter, taking digital information
from the software application and creating a
corresponding analog signal that is fed to the
transceiver. (For more information on A/D
and D/A converters, see the
Analog Basics
chapter.)
hardware necessary to be competitive in digi-
tal DX hunting or contesting, a good sound
card can give you a substantial edge. Other
applications that require a lot of process-
ing power, such as software-defined radio
(SDR), require a high-performance sound
card. Often vendors will offer a list of sound
devices known to perform well with their
equipment.
Sound cards convert analog audio signals
to a set of digital samples, but this conver-
Sample Rate
The clock that drives the A/D converter
runs at a steady rate, known as the
sample
rate
. As you might expect, a higher sample
rate is required to accurately capture higher-
frequency sounds. A waveform can be ac-
The
WSJT
Revolution
Hams routinely use meteors and the moon as radio relectors for
meteor scat-
ter
and
moonbounce
communication. You can read more about these activities in
the chapter on
Propagation of RF Signals
and in the
Space Communications
supplement on the
Handbook
CD. For many years, meteor scatter and moonbounce
required large antennas and high power RF ampliiers. CW was the mode of choice
since a concentrated, narrow-bandwidth signal had the best chance to survive the
journey and still be intelligible at the receiving station. That changed in 2001 when
Joe Taylor, K1JT, unveiled a suite of sound card applications known simply as
WSJT
.
By using the sound card and computer as powerful digital signal processors,
WSJT
greatly reduced the station hardware requirements, making it possible for
amateurs with modest stations to make meteor scatter and moonbounce contacts.
Hams have also experimented with using some
WSJT
modes on the HF bands to
make contacts using extremely low power.
WSJT
does not support conversational
contacts with lengthy exchanges of information. Instead, the software allows for
basic information exchange suficient to meet the requirements that a contact has
taken place.
WSJT
is available for
both
Windows
and
Linux
.
It is a software suite that
supports ive different
operating modes:
FSK441
for meteor scatter;
JT65
for
moonbounce, but also being
used occasionally on HF;
JT6M
, optimized for meteor
scatter on 6 m;
EME Echo
for measuring the echoes
of your own signal from the
moon; and
CW
for moon-
bounce communication using
15 WPM Morse code.
The software is available
for free downloading from
physics.princeton.edu/
pulsar/K1JT
.
1.1 Which Sound Card
is Best?
This is one of the most popular questions
among HF digital operators. After all, the
sound device is second only to the radio as the
most critical link in the performance chain.
Sound cards have traditional analog audio
amplifiers, mixers and filters, all of which can
introduce noise, distortion and crosstalk. A
poor sound device will bury weak signals in
noise of its own making and will potentially
distort your transmit audio as well.
If you have a modest station and intend to
enjoy casual chats and a bit of DXing, save
your money. An inexpensive sound card, or
the sound chipset that is probably on your
computer’s motherboard, is adequate for the
task. There is little point in investing in a
luxury sound card if you lack the radio or
antennas to hear weak signals to begin with,
or if they cannot hear you.
On the other hand, if you own the station
Fig A1 — WSJT software operating in the JT-65
mode.
2
Digital Communications
curately captured by sampling at twice the
highest frequency of interest. Energy at higher
frequencies produces
aliases
, so sound cards
put a low-pass filter ahead of the A/D con-
verter, running at a cutoff frequency equal to
one-half the sample rate. These filters cannot
be perfect, so there’s bound to be either some
high frequency roll-off, or some distortion due
to aliases sneaking through.
of the nominal 11025 Hz. This error not only
affects the apparent frequency of an incoming
tone, but can keep a program that requires a
high degree of frequency stability, such as
MFSK16, from maintaining consistent fram-
ing on incoming data. The result will be poor
or inconsistent copy.
VoIP applications such as EchoLink and
IRLP send a continuous stream of data from
one sound card to another over the Internet.
If the sender and receiver aren’t running at
the same rate, it can cause audio drop-outs,
as the buffer at the receiving end becomes
empty or overflows.
WSJT requires a high degree of accuracy
but works around this problem by measuring
the actual sample rate for sound card input and
output (on some computers they are not the
same) and then doing appropriate manipula-
tions of the data. Most software doesn’t have
active sample error correction, however. If
your software and hardware support it, you
might realize better performance at a higher
sample rate such as 12 kHz.
Sample rate accuracy is usually less of an
issue for modes such as RTTY, PSK31, Olivia
and DominoEX that synchronize the receiver
with the sender frequently. In these modes, the
sound card’s clock is still used as the timing
reference, but exact sound card timing is far
less critical.
just to name a few.
Depending on your computer, you may be
able to choose your receive audio connection
from either
MIC INPUT
or
LINE INPUT
. Anything
else you find is not an analog audio input. If
you do have a choice, use the
LINE INPUT
for
the receive audio from your radio. Although
the
MIC INPUT
jack can be used, it will have
much more gain than you need and you may
find adjustment quite critical. Some sound
cards have an “advanced” option to select a 20
dB attenuator that will reduce gain and make
the
MIC INPUT
jack easier to use.
If your only goal is to decode received sig-
nals, you need nothing more than an audio
cable between the transceiver and the sound
card. You should not need ground isolation
for receive audio as it is at a high level and
is normally not susceptible to ground loops.
You may also be able to choose from several
outputs that appear on the radio. Your radio
may have
SPEAKER
,
HEADPHONE
,
LINE OUT
,
RECORD
,
PHONE PATCH
and
DATA OUTPUT
jacks
available. These may be fixed output or vari-
able output (using the radio’s volume control).
Be careful with radio’s
DATA OUTPUT
jack –– it
may not work on all modes.
For the sound card transmit connection,
you will have a choice of the computer’s
HEAD-
PHONE OUTPUT
,
LINE OUTPUT
,
SPEAKER OUT-
PUT
or a combination of these. The
SPEAKER
OUTPUT
is usually the best choice as it will
drive almost anything you hook to it. The
SPEAKER OUTPUT
has a low source impedance
making it less susceptible to load current and
RF. Any one of these outputs will usually
work fine, provided you do not load down a
line or headphone output by using very low
impedance speakers or headphones to moni-
tor computer-transmitted audio. The transmit
audio connection must have full ground iso-
lation through your interface, especially if it
drives the
MIC INPUT
of the radio. This usu-
ally means adding an isolation transformer,
as discussed in the next section.
Linearity
If the sample steps aren’t all exactly the
same size, or the clock drifts up and down a
little bit in frequency, distortion is introduced.
The ideal sound card would have a perfectly
linear A/D converter and a perfectly stable
clock, but of course these are impossible to
achieve. Good-quality sound cards do, how-
ever, have crystal-controlled oscillators as
their clock source.
Sample Rate Accuracy
Even if the clock runs at a stable frequency,
we won’t get the desired result if it’s running
at the
wrong
frequency. Sample rate accuracy
can be important for analog modes that aren’t
continuously synchronized. For these modes,
one sound card is generating a signal and the
other is receiving it, and the two cards are
expected to be running at exactly the same
sample rate. Distortion, or even loss of data,
can occur if the rates are slightly different.
Remember that sound cards work by tak-
ing the analog audio signal at the input and
converting it to digital information by sam-
pling the audio signal at a very high rate,
typically between 8000 and 11025 Hz for
most ham applications. Sound cards can have
sampling rate errors that will seriously affect
weak-signal copy. For example, the laptop
on which this chapter is being written has an
actual sampling rate of 11098.79 Hz instead
SOUND CARD STATION SETUP
Fig 2
shows a typical sound card station
setup. A simplified sound card is shown here,
but even the simplest of sound cards can be
complicated. They can have as few as two
external connections but there may be as
many as twelve or more. You may find ports
labeled
LINE IN
,
MIC IN
,
LINE OUT, SPEAKER
OUT
,
PCM OUT
,
PCM IN
,
JOYSTICK
,
FIREWIRE
,
S/PDIF
,
REAR CHANNELS
or
SURROUND
jacks,
Transceiver
Computer Serial Port
or
USB Port
1.2 Sound Card Interfaces
In addition to providing audio connections
between the sound card and your transceiver,
you also need to provide a way for your com-
puter to switch the radio between receive and
transmit. This is where the
sound card inter-
face
comes into play.
Commercial sound card interfaces such
as the one shown in
Fig 3
match the audio
levels, isolate the audio lines and provide
transmit/receive switching, usually with your
computer’s serial (COM) or USB port. You
can also make your own interface by simply
isolating the audio lines and cobbling together
a single-transistor switch to connect to your
COM port.
Fig 4
shows some commonly used
interface circuits.
In Fig 2 you’ll notice that the transmit audio
Transmit/
Receive
Keying
Sound Card
Interface
Transmit Audio
Transmit Audio
Receive Audio
Sound Card
Inputs and Outputs
HBK0147
Fig 2 — A typical interface connection between a computer and a transceiver. Note
that the transmit audio connects to the radio through the interface, and TR keying is
provided by the computer serial port. Newer sound card interfaces are often designed
to work with computer USB ports.
Digital Communications 3
is known as a
mark
. A “0” bit is represented
by a 2295-Hz tone called a
space
. The differ-
ence between the mark and space is 170 Hz,
called the
shift
. When applied to a single-
sideband transceiver, this AFSK audio signal
effectively generates an RF output signal that
shifts back and forth between the mark and
space frequencies.
A transceiver that supports FSK, however,
can accept mark/space digital data directly
from the computer and will use that informa-
tion to automatically generate the frequency-
shifting RF output. No audio signal is applied
to the transceiver when operating FSK.
Is there an advantage to using AFSK or
FSK? In years past, transceivers that did not
support FSK operation often did not allow
the use of narrow IF filtering. Those filters
were reserved for CW, not the SSB voice
mode used with AFSK. If you wanted to use
RTTY with such a rig, you had to use AFSK
and contend with the wider (2.4 kHz or so)
SSB IF bandwidth, or else add an external
audio filter. FSK-capable transceivers, on the
other hand, allowed the RTTY operator to
select narrow CW filters, reducing receive
interference in crowded bands.
Many of today’s transceivers offer ad-
justable-bandwidth digital signal processing
(DSP) filters in the IF stages that can be used
with any operating mode. This has effectively
eliminated the FSK advantage, at least for
receiving.
The appeal of FSK remains, however,
when it comes to transmitting. A
properly
modulated
AFSK signal is indistinguishable
from FSK, but it is relatively easy to overdrive
an SSB transmitter when applying an audio
signal from a sound card (more on this in the
next section). With FSK this is never an issue.
You simply feed data from the computer to
the radio; the radio does the rest.
This is why a number of RTTY operators
still use FSK, and it’s why transceiver manu-
facturers still offer FSK modes (sometimes
labeled
DATA
or
RTTY
) in their products. Sev-
eral sound card interfaces support FSK by
providing a dedicated TTL circuit between
the computer COM port, where the FSK data
appears, and the transceiver FSK input. When
used in this fashion, the sound card does not
generate a transmit audio signal at its output.
Instead, the RTTY software keys the various
lines at the COM port to send the FSK data.
If the sound card interface you’ve chosen
doesn’t support FSK, you can build your own
TTL interface using the circuit shown in Fig
4. This simple circuit uses a transistor that is
keyed on and off by data pulses appearing
on the COM port TxD pin (pin 3 on a 9-pin
COM port).
It is important to note that FSK transceiver
inputs can only be used for modes that are
based on binary FSK, typically with 170 or
200-Hz shifts. These modes include RTTY,
Fig 3 — The MicroHam Digi Keyer is an example of a multifunction sound card
interface. Some commercial devices include CW keyers, transceiver control interfaces
and support for multiple transceivers.
connects to the radio through the interface.
By doing so, the interface can provide isola-
tion. Some interfaces also include a transmit
audio adjustment, although this can also be
accomplished at the computer, as described
in the previous section.
It’s important to mention that you may also
be able to use the VOX function on your trans-
ceiver to automatically switch from receive
to transmit when it senses the transmit audio
from your sound card. This approach com-
pletely removes the need for a TR switching
circuit, COM port and so on.
The weakness of this technique is that it will
cause your radio to transmit when it senses
any
audio from your computer — including
miscellaneous beeps, sounds or music. It’s
best to turn off these sounds from your com-
puter so you don’t transmit them accidentally.
Another approach is to use a second sound
card so that one can be used for regular com-
puter audio applications and one dedicated to
interfacing with the radio.
AFSK AND FSK
Most sound card modes rely on some form
of frequency and/or phase-shift keying to
create digitally modulated RF signals. This
modulation takes place at audio frequencies
with the sound card audio output applied di-
rectly to an SSB voice transceiver, either at
the microphone jack or at a rear-panel acces-
sory jack, and is called
audio frequency shift
keying
(
AFSK
).
RTTY, PACTOR I and AMTOR signals
can be sent using AFSK, and often are. It
is also possible to transmit these modes by
applying discrete binary data
directly
to the
transceiver. This technique is known simply
as direct
frequency-shift keying (FSK)
.
For example, each character in the Bau-
dot RTTY code is composed of five bits.
When modulated with AFSK, a “1” bit is
usually represented by a 2125-Hz tone and
Fig 4 — Some commonly used interface
circuits. At A, isolating the sound card
and transceiver audio lines. T1 and T2
are 1:1 audio isolation transformers
such as the RadioShack 273-1374. At
B, a simple circuit to use the computer
COM port to key your transceiver PTT,
and at C, a similar circuit for FSK keying.
Q1 is a general purpose NPN transistor
(MPS2222A, 2N3904 or equiv). At D, an
optocoupler can be used to provide more
isolation between radio and computer. On
a DB9 serial port connector: RTS, pin 7;
DTR, pin 4; TxD, pin 3; GND, pin 5.
4
Digital Communications
Plik z chomika:
umts
Inne pliki z tego folderu:
22.PDF
(7750 KB)
24.PDF
(8887 KB)
21.PDF
(7438 KB)
17.PDF
(5987 KB)
25.PDF
(9493 KB)
Inne foldery tego chomika:
arrl 2006
arrl 2010
arrl 2011
arrl 2012
arrl 2013
Zgłoś jeśli
naruszono regulamin