Weak Signal Communications, by K1JT
Instruction Manual
INTRODUCTION
WSJT is the name of a computer program. It stands for "Weak Signal
communications, by K1JT." Plans
call for the program to include more than one communication mode.
The
first mode implemented, and the one that has given WSJT its wide popularity, is
called FSK441. It is designed for high
speed meteor scatter (HSMS) communication using "pings" reflected
from the underdense ionization trails of random meteors at approximately 100 km
height. Such pings are typically a few
dB above the noise and last for anywhere from ten to a few hundred milliseconds. WSJT makes QSOs possible in the amateur
2-meter and other VHF bands, over typical distances of 500 to 1400 miles (800
to 2200 km), using these brief pings. Such
QSOs can be successful with modest equipment and without waiting for "band
openings."
WSJT is designed for computers running the Windows
operating system; Windows 95, 98, ME, and 2000 have all been used successfully. The program sends messages using four-tone frequency
shift keying at 441 baud. The
name of the transmission mode is FSK441, although most people are just calling
it “the WSJT mode.” FSK441 makes very effective
use of the audio bandwidth of modern SSB transceivers. As implemented in WSJT, it generates a clean
transmitted spectrum. Each character in
a normal text message requires three tones that are sent sequentially. The transmission speed is 147 characters per
second, or 8820 letters per minute.
Unlike high speed CW, which has also been widely used for meteor scatter
communications, WSJT does not require the user to play back received “pings” and
decode the message by ear. Instead, the
decoded text appears in a scrolling window on the computer screen.
Minimum computer
requirements include a 75 MHz Pentium
or equivalent CPU, 24 MB of RAM and 40 MB of free disk space, a monitor with
800 x 600 or higher resolution, and a Windows compatible sound card. In addition you will need a simple computer-to-radio
interface like those required for PSK31, MFSK16, Hellschreiber, and similar
modes. The DTR or
RTS line of one of the computer’s serial communication ports (a “COM port”) is
used to key your transmitter's PTT line. Connections are also required between your radio's audio output
and the computer’s soundcard input, and vice-versa. Station accessories that accomplish these things are available commercially
from a number of sources. You will need
a method of synchronizing your computer clock with UTC to within about one
second.
3. OBTAINING AND INSTALLING THE PROGRAM
WSJT is available
for download free of charge at http://pulsar.princeton.edu/~joe/K1JT.
Download the file WSJT093.ZIP (or a similar file name with a higher version
number, if one exists). Unzip the distribution file into a convenient directory
(such as C:\TEMP) and then run SETUP.EXE in that directory to install WSJT to a
permanent location of your choice.
During installation you will likely get a couple of
messages that have the general form "An error has occurred while
registering file C:\WINDOWS\SYSTEM\<filename>", or perhaps
"MSVCRT.DLL access violation". You may safely click
"Ignore" to continue from such a message. If your system reports that you already have some file with a
newer date than one being installed, you may keep your existing file.
When SETUP reports the installation process to be
complete, you will probably want to create a shortcut on your desktop for
starting WSJT. Double-click on "My Computer", then on the relevant
hard drive symbol, and navigate from there on down to the directory on which
you have installed WSJT. Right-click on WSJT.EXE, and select "Create
Shortcut". Then drag the shortcut symbol from the installation folder onto
your desktop.
If you discover that an update to WSJT has been
released and you wish to upgrade, download the appropriate file from the same
web site. It will be named something like UPD094.ZIP, and will be a much
shorter file than the full installation package. Unzip the update file in a
suitable directory such as C:\TEMP and follow any directions in a README.TXT
file. Typically, the update will require simply copying new versions of the
files WSJT.EXE, WSJT1.DLL, and perhaps a few others over your existing
copies.
4.
INITIAL SETUP
Double-click the WSJT icon or select it from the
Windows START menu to start the program, and then go through the following
steps:
1. SELECT COM PORT FOR T/R SWITCHING
In the Setup menu, select the desired COM port and
the DTR or RTS line for controlling the PTT line of your radio. Note that you
can both "check" and "uncheck" items on this menu. If you don’t
know whether your station is using DTR or RTS, check both. On a first
installation (or following deletion of your WSJT.INI file) you may be prompted
to specify a COM port number. At this
prompt you may also specify a COM port number greater than 4, should that be
necessary for your setup.
2. TESTING T/R SWITHCING
Click one of the four
"Tune"
buttons labeled A, B, C, or D to be sure that T/R
switching works -- i.e., that the PTT line is keyed. Each tune button sends one
of the four audio tones to the soundcard speaker output, and from there to your
radio. You should be aware that all transmissions with this program are meant
to produce a full-amplitude, key-down carrier with 100% duty cycle. For
comparison, the on-off keying of CW (including HSCW) produces approximately 50%
duty cycle. (HSCW also produces a rather ugly spectrum, especially at the
higher speeds, with significant amounts of power in keying sidebands outside
the SSB bandwidth and consequent loss of signal-to-noise ratio.) If the high
duty cycle will overstress your final amplifier, you should reduce power. The
reduction in power output will be more than made up by the increased
signal-to-noise ratio provided by WSJT. Click TX Stop to
stop transmitting.
3. TESTING RX AUDIO LEVEL
The next step is to adjust
the receive audio level. With the receiver running on background noise, click Record to start a receiving period. On the left side of the status bar you should see "File: callsign_nnnnnnn_nnnnnnn"
in the left box. Next to this box you should see "File Position: n s."
This shows that the program is recording a new file. After recording a few
seconds of audio, click Stop. A message should appear in
the third panel of the status bar at the bottom of the screen.* This message
should say "RX Noise: n dB." This message describes the level
of received noise going into the soundcard. A jagged green line should appear
in the large plot area above, together with a two-dimensional grayscale spectrogram. The green line is a graph of
received noise power (vertical axis) over the time you were recording
(horizontal axis). The grayscale plot is a time vs. frequency spectrogram of
the received audio. If you see no green line and no grayscale plot, there’s
probably no noise (or not enough noise) going into the soundcard. Using the
receiver volume and/or the computer's volume control for "audio in"
repeat the short recordings until you get a noise level close to 0 dB. This
value is not highly critical: +/- 2 or 3 dB from the nominal 0 dB should be
fine. When properly adjusted, the green line should appear below the grayscale
spectrogram and approximately aligned with the short horizontal tick marks just
outside the edges of the plot area.
*If your radio
has an Automatic Gain Control (AGC), turn it off or to its “fast” position.
It may also help to back off on the RF
gain control. Also turn off your noise
blanker (NB) for now. You may want to experiment
with the NB later; with many radios, WSJT signals are not adversely affected by
the noise blanker, but in other cases pings are severely clipped.
WSJT displays a message if the audio level is extremely low (or if
there is no audio at all). If this occurs, check to make sure your cables are
connected and that you have selected the correct audio input channel on the
Windows mixer.
Unlike WinMSDSP, WSJT
displays a received signal at the end of a recording period, rather than
as-you-go. In addition, the graph of received signal strength vs. time – the
green line referred to above – represents power, not voltage. The power samples
have been smoothed over 0.1 s intervals for plotting. Pings appear on this plot
as upward-going spikes above the grassy baseline. Full scale (to the top of the
plot area) represents a signal 30 dB above the fluctuations in receiver noise
power.
Recordings of receiver
noise will also produce purple and red curves in the smaller plot area at the
upper right, together with four yellow marks above the curves. The purple curve
illustrates the average spectrum of received noise and therefore (in the
absence of an actual signal) represents your receiver's passband shape, including
the effects of IF and audio filters as well as the soundcard interface. The
vertical scale is in dB, and the length of the yellow tick marks is equal to 10
dB. The tick marks denote the frequencies of the four tones used in the FSK
character-encoding scheme, namely 882, 1323, 1764, and 2205 Hz. The red curve
is similar to the purple curve, but it represents the spectrum of the strongest
signal found by the decoding algorithm. There will be no red curve if no signal
was detected. Ideally, your receiver passband shape should be flat -- i.e., the
purple curve should be horizontal on the graph -- from about 600 to 2400 Hz. If
this is not the case with your receiver, WSJT software will compensate
automatically.
WSJT uses four audio tones
to drive the transmitter. It is important that each of these tones is sent with
approximately the same amount of power. Press each of the four "Tune" buttons (A, B, C, and D)
in turn and check your transmitter power output (or the final amplifier
current, or some other relative indicator) while transmitting each tone.
Ideally the power should be the same with each tone; in practice, tolerances of
+/- 10% or even +/- 20% are acceptable, but 50% changes will deteriorate your
signal’s readability.
Many people using WinMSDSP have installed low-value
series coupling capacitors in their receiver-to-soundcard or
soundcard-to-transmitter interfaces. With the single-tone on-off keying of
WinMSDSP this causes no problem, and may even be an advantage. Although these
capacitors may provide some RF isolation, they also become high-pass filters in
the audio range. This can cause unequal power output among the four frequencies
used in WSJT.
At any given instant WSJT
transmits only one tone. As a result, WSJT does not require an amplifier to be
linear--you may drive a class-C amplifier if you wish. Unlike PSK-31 (and to a
lesser extent WinMSDSP), one does not need to reduce audio gain in the
transmitter to preserve linearity and minimize distortion. Turning up the audio
gain may be all you need do to equalize your transmitted power in the four
tones. Activating the speech processor may also help to level your transmitted
spectrum.
Decoded text for detected
signals appears in the decoded text box in center
of the screen:
Examples:
154000 15.0 160 8 26 -21 AA7A 2727 K0SM
27 AA7A 2
154100 17.2 40 2 16 109
R 3 1$
154500 6.7 100 5 26
-21 RRR
The first column shows the start time of the
recording time in HHMMSS format. The
second column shows the number of seconds into the receive period the program
detected the ping. Columns 3 and 4 give
the duration of the signal in milliseconds (ms) and its strength in dB above
the noise floor. Column 5 suggests an
appropriate signal report for this signal.
The sixth column displays a quantity called "DF," which
measures the apparent mistuning of the receiver, in Hz. If one of the specially encoded single-tone
messages has been detected, it is listed next: R26, R27, RRR (as in third
example), or 73. The decoded text from multi-tone
messages comes last.
In the example above, you can see that the first ping
contained two callsigns and a "27" signal report. This signal occurred 15 seconds into the
file that started at 15:40:00 UTC. The
signal lasted 160 ms and peaked at 8 dB above the noise floor. It was received 21 Hz lower than the
expected frequency. (WSJT will correctly
receive signals mistuned by 200 Hz or even more, but tuning to within +/- 100
Hz is desirable). The signal received 17.2
seconds into the interval starting at 15:41:00 was a noise burst that triggered
the decoder. It is very short and
contains no information, and the operator would be able to hear it as a noise
burst rather than an actual signal burst.
(WSJT signals “sound different” than static crashes.) The ping at 15:45:00 was decoded as a
single-tone "roger" message.
Notice that the RRR is in a separate column to the left of multi-tone
messages.
6. The "Pick Feature"
Clicking the mouse inside
the grayscale spectrogram or on the green line will force the program to
decode text at that particular time, regardless of signal strength. When the
mouse is inside spectrogram area, the arrow will change itself to a
cross. The location of the mouse is
displayed at the left, in seconds. The
Pick Feature is useful for decoding steady signals that do not trigger the
decoder of the program.
Normal
Operation
On-the-air use of WSJT benefite by following
procedures similar to those used in high-speed CW meteor scatter. Enter your callsign in the My
call text box and the other station's callsign in the To
radio text box. Next, enter the
length of the TX period (in seconds) in the Period box. (Most people are using 30 second periods,
the default value.) Press Generate Std Texts to create the most commonly used
messages. You can edit the messages, if
desired, and you can compose custom messages.
Toggle the Standard Texts / Custom Texts button,
and then edit any of the six text boxes.
WSJT will save your custom messages and it restores them the next time
you run the program.
When you are ready to start an automated sequence of
reception and transmission periods, switch Auto Period to
ON.
When you have finished a recording, pings (short
bursts of signal) are detected and decoded automatically. Minimum thresholds for the duration (defined
by the Width parameter) and Strength of
detected pings can be set by choosing Parameters |
Decoding Parameters or by using the +/- buttons on the window. You may reduce the Width and Strength
thresholds if you want the program to be more sensitive to weak signals. This will also increase the number of noise
bursts causing false decodes. Suggested
threshold values are width = 20 ms and strength = 2 or 3 dB. To reduce false decoding from static crashes,
increase these settings to 40 ms and 4 dB, or perhaps 60 ms and 5 or 6 dB. Ping widths are measured in 20 ms
increments. Note that at 147 characters
per second, a 20 ms ping contains only three characters. Nevertheless, it is not unusual to see two
to four characters correctly decoded from a 20 ms ping only 3 or 4 dB above the
noise.
The ST parameter is the strength
limit for the specially encoded single-tone messages. The great advantage of these messages is the improved
signal-to-noise ratio they provide, as much as 3 to 6 dB. A Single Tone limit around –1 dB (or, say, 4
dB lower than the primary Strength limit) seems to be about right for most conditions.
WSJT attempts to compensate for some mistuning
between transmitting and receiving stations.
However, mistuning by more than about 200 Hz may cause decoding to
fail. For every detected ping, the
program lists the mistuning in Hz under the DF column in the decoded text
box. You should consistently see
similar numbers in the DF column during a QSO that is producing usable signals. If the DF number is consistently outside the
range +/- 150 Hz, it will help to retune your receiver to compensate. For example, if you see repeated DF values
at -280 Hz (or observe that the red spectral curve for a strong ping seems to
be displaced significantly to the left), tune your receiver lower in frequency
by the appropriate amount. Do this with
your RIT control, or by using split T/R mode if your rig has such a
feature. You want your transmitted
frequency to remain constant, since your QSO partner may me trying to tune to
your signal, also. You do not want to
confuse him or her!
Although retuning the receiver is the preferred
method, you can also use a software parameter called Expected DF. Expected DF causes the frequency range for
signal decoding to be moved upward or downward by a specified number of
Hz. You can also narrow the range of frequencies
searched by reducing the value of Tolerance from its default
value of 400 Hz. If you have clearly
identified the frequency at which you are receiving your QSO partner, and your
two rigs are stable, setting Tolerance to a low value such as 100 Hz will
significantly filter out noise bursts.
Small buttons near the Decode Again button allow
adjustment of the decoding parameters.
The Reset Defaults button quickly returns all parameters
to reasonable starting values.
Whatever you use for the ping limits, you will still
see errors in the decoded messages, especially near the beginnings and ends of
pings where the signal fades up from and back into the noise. Of course, the same uncertainties occur with
HSCW or any other mode of communication when signals are marginal. Unlike some digital modes that require high
accuracy and therefore employ error-correcting schemes, the design criteria for
WSJT willingly relax accuracy requirements in order to gain speed.
MENU
OPTIONS:
This allows you to open a
wave file from the hard drive. The file
must be a standard WAV file recorded 8-bit monaural format with 11025 Hz
sampling. The file will be decoded
automatically as if it had been recorded from the radio.
WSJT saves files in the
RxWav subdirectory. Selecting this
option will delete all .WAV files in the RxWav subdirectory. WSJT will ask for a confirmation before
proceeding.
| Save text in file
DECODED.CUM
Selecting this option will
save all decoded text in a file named DECODED.CUM. This file may become quite large after using the program many
times. If DECODED.CUM is not present in
the WSJT directory, the program will create it.
This
will delete the file DECODED.CUM.
These options allow you to select
which COM port will activate T/R switching with the DTR or RTS line. If you need to use a COM port number higher
than 4, exit WSJT, delete the file WSJT.INI in the installation directory,
restart the program, and enter the desired number when prompted for a COM port.
These are two signal lines in
the COM port, and you may select either one (or even both) to activate your T/R
switch. You can find a very simple TR
switch interface can be found at http://www.qsl.net/k0sm/interface.gif.
| North American standard
messages, European standard messages
These options establish the
style of messages created when you push the Generate Std
Texts button. Because there are
different meteor scatter operating procedures in different IARU regions, WSJT
can automatically create standard messages for both North American and European
operators. Operators in other parts of
the world should decide what format they wish to use.
If your computer clock is
not set to UTC, use this option to offset the WSJT clock to UTC. If your computer clock is behind UTC, enter
a positive number of hours. If your clock is ahead of UTC, enter a negative
number.
This will display the
Windows Volume Control. By default,
windows will display the playback control. This will allow you to adjust the TX audio level from the
computer. If you want to adjust the recording
level, you will need to go to Options | Properties and select Recording.
This will create a short
delay between the time the time the TX and RX mode in the program. The purpose of this feature is to avoid
false signals that are created by amplifier and preamplifier switching.
This option enables an
automatic station identification. You
must record a file named ID.WAV and place it in the same directory as
WSJT.EXE. The file may give your
station identification in either voice or CW, and it should be recorded in
8-bit monaural WAV format at 11025 samples per second. The ID interval (set on the Parameters menu)
determines how often WSJT sends the identification.
Currently, HSMS is the only
mode supported in WSJT. In the future,
PUA-43 and EME-2 modes will be added.
|Decoding parameters and ID interval
This box allows you to
adjust the decoding parameters of WSJT.
The decode parameters are Width, Strength,
Single Tone Strength, DF, and Tolerance.
The "identification
interval" will determine how often you send the ID.WAV file tosend your
station identification.
For
more information on the decoding parameters, see decoding
algorithm.
|Display received tone levels
This will display the
relative strength of tones in a decoded signal. These numbers appear in a separate column after the decoded text
in the decoded text box.
| Help
Display
the online Help File.
| About WSJT
This item displays version
and copyright information. On most
Windows installations the "System Info" button displays information
about your computer and operating system.
This button turns the timed
T/R switching on and off. You should
set the transmit and receive duration in the Period text
box (in seconds). During a schedule, you and your partner alternate so that
only one station is sending at one moment.
If the TX First box is checked, your station will send starting on the
minute. Only one station should have
the TX first box checked! In North
America the convention is for the westernmost station to transmit first.
This slider will adjust the
brightness of the grayscale spectrogram.
This slider will adjust the
contrast of the grayscale spectrogram.
This button decodes the
last recorded file again. After
recording and decoding a file, you can adjust the decoding
parameters. Pressing Decode Again will decode the file with the new decoding
parameters.
The + and - buttons adjust
the DF parameter that WSJT uses in the decoding process. You can see the current settings in the
status bar at the bottom right of the program window. It is also available under Parameters |
Decoding parameters and ID interval.
If you receive a signal
that is more than about 100 Hz away from your frequency, you can adjust this
parameter to improve copy. You can
adjust this parameter with the +/- buttons and press Decode
Again. You can use DF to substitute
for the RIT if you radio does not have one.
It will not be as effective as an actual retuning of the receiver,
however.
This erases all data in the
decoded text box, spectral plots, and graphs at the
top of the screen. It will not
delete any files.
This button will generate
the most often used messages in meteor-scatter QSOs. These messages will use the information you enter into the My call, To radio, and Report text boxes.
The format of the messages is slightly different depending upon your
IARU region. WSJT provides both North American and European formats. Go to the Setup menu option and select the
format for your region.
Use this button to monitor
a calling frequency or two other stations in a QSO. It records continuously (except for a small gap between periods) and
displays data after each recording period.
This control plays a
recorded file through the speakers (and through the radio if you are not
careful!). It functions much like the "Play" button on a cassette
recorder.
This button starts
recording receiver audio. The program
will record for the length of time entered in the Period
box or until you press the Stop button. When the program has stopped recording, the
file will be plotted and analyzed for signals.
This works much like the "Record" button on a cassette player.
This button will reset the decoding parameters (Width, Strength,
ST, DF and Tolerance) to
their defaults. It does not affect any
other settings.
Put a check in this box to
save every .WAV file WSJT records. The
files are saved in the RxWav subdirectory.
WARNING: Each 30-second file requires about 0.33 MB of disk space. You can delete all .WAV files in the RxWav
subdirectory by choosing File | Delete files in RxWav. To save individual files, use Save Last.
Each received file is
deleted when the next recording begins, unless you have clicked Save Last to save a single file, or checked Save All to save all received files.
Clicking this button will
cause the last decoded file to be in the RxWav subdirectory. You can reopen the file at a later time by
using File | Open.
Each received file is
deleted when the next recording begins, unless you have clicked Save Last to save a single file, or checked Save All to save all received files.
Pressing these buttons will
activate the transmitter. The text will
be sent for the duration listed in the Period text
box. The text box becomes colored to
remind the operator which text is being sent to the transmitter. Yellow backgrounds denote multi-tone
messages, while light blue backgrounds signify single-tone messages.
This button switches
between two sets of messages. The
standard texts are those most commonly used in meteor-scatter QSOs. WSJT provides both North
American and European formats. Use
the custom texts to store other messages, such as grid square or contest
information. Your custom messages will
be saved when you exit WSJT and restored when you next run the program.
This stops a record, monitor, or play
operation. It functions much like the
"Stop" button on a cassette recorder.
This sets the minimum
signal strength of a multi-tone signal that WSJT will decode. If this is set at 2 dB, WSJT will only try
to decode signals that are 2 dB or more above the noise floor and that are
longer than the time defined in the Width parameter. Setting Strength to a lower value allows to
you decode weaker signals, but it also displays more noise.
The + and - buttons adjust
the Strength parameter that WSJT uses in the decoding process. You can see the current settings in the
status bar at the bottom right of the program window. It is also available under Parameters |
Decoding parameters and ID interval.
This setting sets the
minimum strength of a single-tone signal that WSJT will decode. It works in much the same way as the
Strength parameter does for multi-tone signals.
The + and - buttons adjust
the Strength parameter that WSJT uses in the decoding process. You can see the current settings in the
status bar at the bottom of the program window. It is also available under Parameters |
Decoding parameters and ID interval.
This setting is a type of
"narrow filter." WSJT will
not present decodings for signals mistuned by more than this number of Hz. At the beginning of a schedule, you should
leave this setting at 400 Hz until you know on what frequency the other station
is. Decreasing this value can eliminate
much noise on the screen.
The + and - buttons adjust
the Strength parameter that WSJT uses in the decoding process. You can see the current settings in the
status bar at the bottom of the program window. It is also available under Parameters |
Decoding parameters and ID interval.
These buttons send a steady
tone on each of the WSJT frequencies.
Use these buttons to adjust your transmit audio level.
Check this box to transmit
on the first period of the minute.
Leave it empty if your schedule partner is transmitting on the first
period of the minute.
Checking this box will disable
the transmitter without interrupting the Auto Period. The program no longer keys the PTT and it
will send no TX audio to the radio. The
program continues to receive normally on the RX periods.
This will stop your
transmitter for this TX Period only. It
will NOT stop the Auto Period. The transmitter will automatically key on your next Period unless
you turn off the Auto Period. To stop
sending completely press TX Stop and turn off the Auto Period.
This enables the
single-tone R26, R27, RRR, and 73 messages.
Unlike the other messages, these messages can be sent without using
FSK. The program is able to send and
detect steady, single tones at each of the four WSJT frequencies. Most operators have found that the
single-tone messages are very effective for completing QSOs when pings are weak
and scarce. See Definition
of Code.
The + and - buttons adjust
the Width parameter that WSJT uses in the decoding
process. You can see the current
settings in the status bar at the bottom of the program window. It is also available under Parameters | Decoding parameters and ID interval. See also: Decoding
Algorithm.
This is the large screen in
the center of the program window. It
displays the decoded text,
and other useful information.
My call
Enter your callsign in this
box
Enter the callsign of the
other station in this box. This box
will also determine the file name for all recorded files. Therefore, you cannot enter any symbol that
cannot appear in a windows files, such as "/" or "?". If you must use these in a message, edit them
into the message manually.
Enter the TX period in this
box, in seconds. If you are
transmitting for 30 sec and receiving for 30 sec, you should but "30"
in this box. If you wish to run 1 min
sequences, enter 60.
Enter the signal report for
the other station in this box. Once you
have done this, push Generate Std Texts to create the
messages automatically.
Charts
and Graphs
This is a time vs.
frequency display of the most recently decoded file. The yellow tick marks are the four frequencies used in WSJT
multi-tone FSK transmission. Signals
will appear in white. You may adjust
the brightness and contrast
with the sliders below the window.
Click Decode Again to
update the display after adjusting the sliders.
You should use this display
to identify possible signals that were not decoded by the program. You can use the Pick Feature
to decode these signals.
This is the information at
the bottom of the program window. It
displays important information such as file names, file position, RX audio
level, and decode parameters.
Single-Tone
and Multi-Tone messages
Multi-tone messages are the
default in WSJT. These use a total of
four tones at 882, 1323, 1764, and 2205 Hz to encode the data. Single-tone messages use a single, steady
tone to convey special messages. See Definition of Code.
Decoding
Algorithm
The algorithm that
determines ping length presently does so as follows. It starts with intensity data based on signal over the full
receiver passband. The power is then
smoothed and sampled at 20 ms intervals.
When the signal exceeds the baseline level by more than the threshold specified
by the Strength parameter, a ping is said to have
started. When the power has dropped to
at least 1 dB below threshold, the ping is said to have ended. The difference between start and end times
is the ping width to the nearest 20 ms. Pings with deep fading may be
interpreted as several closely spaced pings.
If you have the Width limit set to 60 ms, then the
program will not attempt plain-text decoding for 20 and 40 ms pings, no matter
how strong they be. WSJT performs
decoding of single tones for pings of any length.
Multi-Tone messages
FSK441 uses Frequency Shift Keying at a baud rate of 441 Hz. Four distinct tones are used, namely 882, 1323, 1764, and 2205 Hz. Each encoded character uses three tone intervals and therefore requires 3/441 seconds (approximately 2.3 ms) for its transmission. FSK441 can accommodate an available "alphabet" of 48 characters. The present encoding scheme uses 43 of these characters, the same ones used in the PUA-43 alphabet.
Single-Tone Messages
In addition to the
Multi-Tone messages, the messages R26, R27, RRR, and 73 can be send in
single-tone mode. This encoding sends a
steady carrier at one for the four possible frequencies. When WSJT receives a single-tone message it
is displayed in a separate column between the DF parameter and where multi-tone
messages normally appear (see Decoded Text). The program always attempts to decode single-tone
messages on receive. However, the
program will send single-tone messages only if the Use
Single-Tone Msgs box is checked.
PLANNED ENHANCEMENTS
WSJT is a work in progress. Many enhancements are planned, including correction of some minor
bugs, improved performance of both multi-tone and single-tone decoding
algorithms, implementation of PUA-43 and EME-2 modes, and a number of
operational convenience features.
BUG
REPORTS
WSJT is a real-time program that runs on an operating
system not designed for tight control of critical real-time events. It attempts to support the vast range of
different hardware configurations that can run Windows. Of the many hundreds (perhaps a thousand or
more) of hams who have installed WSJT on their computers, probably only a
handful have nearly identical setups.
It is hardly surprising, therefore, that some difficulties have come up
that could not be foreseen in my own program development and tests.
Thanks to a dedicated group of early users of WSJT
and an extremely efficient means of exchanging hints and suggestions via
internet reflectors and email, newcomers to WSJT have had their questions
answered and their problems solved, usually very quickly.
If you have trouble installing or using WSJT, please
ask for help. The best source of help
is probably one of the internet reflectors dealing with amateur meteor scatter communications,
such as hsms@qth.net in North America or meteor-scatter@qth.net in Europe.
If you find bugs the operation of WSJT, please report
them to me at k1jt@arrl.net.
I hope you will enjoy using this piece of
just-for-fun software.
Joe
Taylor, K1JT
