DF.TXT 6.4a          USING APRS FOR DIRECTION FINDING


OVERVIEW:  APRS NOT ONLY PLOTS BEAM HEADINGS (Both Manual and DOPPLER) BUT IT
ALSO HAS TWO METHODS FOR TRANSMITTER LOCATION USING ONLY OMNI DIRECTIONAL
SIGNAL STRENGTH CONTOURS!  The first omni technique displays overlaping
circular signal strength contours over the map based on signal reports from a
number of reporting stations.  The second omni technique plots lines of
bearing based on a single moving omni station (Aircraft or vehicle) plotting
three or more FADE points on his map.  All fade-points on a map where the
transmitter signal fades out should characterize a circle with the transmitter
at the center.  APRS now computes this circle and therefore the location of
the transmitter by computing the FADE circle based on these three or more
points.  This FADE circle technique was added in APRS version 5.8d and has
its own section below.  Since the use of omni directional signal strengths for
locating a hidden transmitter is all new, it is presented first, followed by
the FADE circle technique, followed by the more classical BEARING TRIANGULATION
method using both manual bearings and automatic serial interfaces to Doppler
DF equipment.


OMNI-DIRECTION FINDING:

APRS incorporates a whole new aspect to direction finding by permitting the
plotting of signal strength contours.  THIS PERMITS STATIONS WITH ONLY OMNI
ANTENNAS TO PARTICIPATE AND PROVIDE VALUABLE INFO!  This is possible since
APRS has a line-of-sight Power-Height-Gain (PHG) reporting and display format.
This format adds a station's power, antenna height above average terain, and
his antenna gain in his APRS position report.  APRS can then draw range
circles around each station showing his relative communication range.  For
backward compatibility, all stations not reporting the PHG format are plotted
assuming the default parameters of 10 Watts, 20 feet HAAT, and a 3 dB antenna.

     If each station includes these parameters in his position report, then
APRS plots a map of circles around all stations.  Where two circles inter-
sect or overlap, direct communications are possible.  This PHG plot is an
ideal tool for setting up ANY radio network WHETHER OR NOT APRS or PACKET
is being used!  Initially, my equations are straight from the textbook and
may need to be increased or decreased based on experience.  Please note that
these circles represent transmitting range based on your Power and Antenna
relative to a nominal 10 Watt station at ground level.  Your ability to hear
him, depends on his transmitter relative to 10 Watts.

DFING WITH OMNI SIGNAL STRENGTH REPORTS:  By modifying these PHG equations
for plotting received signal strengths, a weak signal is drawn as a larger
circle of probability than a very strong circle.  In the absence of any
precise signal strength indication on most VHF radios, I simply chose a
scale of 0 to 9 as a relative receive signal strength indication.  These
signal strength numbers replace the transmitter power in the PHG reporting
format and are preceeded with DFS to represent DF Signal strength.  APRS uses
these signal strengths to modify the normal PHG display circles as follows.
The numbers 1 to 9 will be plotted as circles from a dark gray up to a bright
red, with the radius of the circle decreasing the stronger the signal is
reported.  The number 0 represents a NULL or NEGATIVE report, meaning that
NOTHING within that stations horizon was detected.  These NEGATIVE reports
are plotted last as dark gray on top of everything else, since they identify
circles where the transmitting station is NOT.

     Since the PC can not ADD colors together, but only overlaps colors, the
user should visualize all the overlapping colors and not just the brightest
ones on top.  The probable location of the transmitting station will be in
the area of the most concentrated overlaps.  Do not be fooled by the brighter
circles.  Almost by definition, the location of the hidden transmitter will
never be at the center of a circle.  THE LOCATION OF THE HIDDEN TRANSMITTER
IS ALWAYS NEAR THE EDGE OF THESE CIRCLES.  If it was near the center, then
the signal would have been much stronger, and the circle would be a brighter
color and smaller!  Please load the DF-OMNI.BK backup file to see our first
omni-df attempt.  See the section below that describes what you will see in
the DF-OMNI.BK file.


OMNI-DF COMMAND SUMMARY:  The following list sumarizes all of the commands
used in performing direction finding both OMNI and with BEAMS.  Please note
that stations with BEAMS should NOT input OMNI signal strength readings,
since their gain will upset the consistency of the OMNI plots.  Beam stations
should always enter their BEAM HEADINGS.


INPUT-DF - Used to enter either a beam heading or a signal report.  If a beam
      heading of (0) is entered, then APRS assumes the entry is for an OMNI
      signal strength report.   Remember that 360 degrees means North.

INPUT-ADDobj - Used to add a voice reporting or non-APRS DF report to the map.
      By selecting the DF symbol, you will be prompted for the appropriate
      BeamHeading information.  If this is an OMNI report, then enter a beam
      heading of 0 to continue with the OMNI report.

INPUT-PwrHtGain - Not necessarily used for DF, but is used for showing
      your station Power, Height and Antenna gain parameters.

MAPS-PLOT-DF - This command is used to plot the OMNI-DF profiles.

MAPS-PLOT-HEARD - Plots only the DF rings around stations that have heard the
      fox.  This is for monochrome displays to separate the NOT-HEARD circles.

MAPS-PLOT-NOTheard - Plots only the DF rings that bound the stations that have
      not heard the fix.  This is to eliminate confusion on monochrome screens.

MAPS-PLOT-OVERLAY - Plots the DF rings overlayed on the existing road map.

MAPS-PLOT-PwrHtGain - Plots the Power-Height-Gain range rings around all stns.

MAPS-PLOT-Rings - Draws a circle of 1 and 0.5 of the selected map scale.


DESCRIPTION OF EVENTs STORED IN DF-OMNI.BK:  Although APRS can plot the
circles of signal strength, it will take some time to develop the skills to
interpret the result.  On this first attempt, there were NO other APRS
operators at home that I could raise that sunday afternoon, I scoured several
voice repeaters and got a few RF signal reports on the FOX. I then added
these stations to the map using the INPUT-ADD command.  FILES-LOAD the file
named DF-OMNI.BK and hit the MAPS-PLOTS-DF command.

   First, you will notice that APRS does a good job with the dark gray circles
of showing you where the FOX is NOT!  Actually, since the DF mobiles (not
aware of APRS and NOT in communications with me) took more than an hour just
to get close enough to hear the FOX, APRS users could have immediately begun
to drive to North Baltimore and cut at least an hour off of their search
times.  Second, notice the offset circle of KA3DZZ.  IF he had not added as
an afterthought that he had a ridge blocking his East view, the gray null
circle from him would have misslead us for a while.  Notice, that most of the
stations had ever participated in a fox hunt before, and had no talent in
estimating signal strength and some were even using HT's with rubber ducks!

    The most interesting thing is the report from W3PWF who said it was a very
strong signal and he was much further than either of the nearby mobiles that
reported weak signals.  ALthough he was in his driveway, he had almost 200
feet of height above average terain, but could not quantify it at the time.
This points out how tricky it will be to use the OMNI-DF plots.  Do NOT
rely on any one report.  You must visually take it all in.  His report is
correct, and although he has a large horizon, APRS draws his pink circle
smaller to show that the FOX could be closer to him.  Remember to look at the
edge of his circle, not the center.  If the FOX was closer to him, then his
signal strength would have been even stronger, and the circle even smaller!

    APRS draws stronger reports smaller for two very important reasons.  First
a stronger signal means the FOX is closer to to the reporting station.  Second,
since PC screens cannot MIX colors and only the last one drawn is visible,
APRS draws all OMNI-DF reports on the screen starting with the weakest
(largest) going up to the strongest and smallest.  After all of these colored
reports are plotted, then APRS goes back and plots all of the 0 or NULL
reports.  They are drawn on top, since they are a POSITIVE report that the FOX
is NOT within their range.  If we could have gotten a NULL report from a
station to the northeast of the pink circle, then it could have overlapped
the NE section of the PINK circle and told us that the signal was clearly
coming from the southwest of W3PF.

YOU MUST REMEMBER TO LOOK AT THE EDGES OF ALL CIRCLES, NOT THE CENTERS!  THE
FOX SHOULD BE NEAR THE LOCATION WHERE THE MOST CIRCLES INTERSECT OR OVERLAP.

     This was just my first test, and unplanned.  Notice that with all of the
stations that we rounded up, only 4 of 13 even heard the FOX at all.  For
serious work, each station reporting should have a very good idea of his
Height above average terrain and general geographic horizon.  If each of
those stations was also watching the APRS plots unfold, they could have
modified their reports to be more meaningful!


RECOMMENDED OMNI-DF PROCEDURE:  As soon as the APRS net is alerted of a FOX
or a hidden transmitter, each APRS station should first listen on the reported
frequency and enter his signal strength.  Next  each of the APRS operators
should go onto the local voice repeaters and ask for OMNI-SIGNAL strengths
from mobiles and any other fixed stations.  The APRS operators use the INPUT-
ADD command to add these stations to the map.  By having one APRS operator
listening on EACH local voice repeater, and solliciting reports, the maximum
number of reports can be gathered with a minimum amount of chatter.  Having
random APRS statios randomly soliciting reports on a random number of voice
repeaters causes a lot of duplication and repeats.  Be sure to get the
stations reported signal strength, location, Antenna height-ABOVE-AVERAGE-
TERRAIN (not sea level or above ground) and any offset in his horizon.  My
interpretation of the signal strength scale is as folllows:

   0   No signal detected what-so-ever
   1   Detectible signal (Maybe)
   2   Detectible signal (certain but not copyable)
   3   Weak signal marginally readable
   4   Noisy but copyable
   5   Some noise but easy to copy
   6   Good signal with detectible noise
   7   Near Full-quieting signal
   8   Dead Full-quieting signal no noise detectible
   9   Extremely strong signal "pins the meter"

Don't forget that stations DO NOT NEED TO BE APRS stations to participate!
Any voice report can be entered on the map by any other APRS station using
the INPUT-ADD command and selecting the DF symbol type.  By entering a beam
heading of 0, the user is prompted for ths signal strength report.  For more
information on the Power-Height-Gain formats, see the DIGIs.txt and
PROTOCOL.txt files.

PLOTTING DETAILS FOR OMNI-DF CIRCELS;  I used the radio horizon forumla for
the radius of the circles, modified by the siignal strength value.  There is
presently no mathematical basis for the factor by which I reduce the size of
the circle, porportionaly to the greater signal strength.  I guessed at a
relationship that would roughly match the subjective weightings of the reported
values anyway.  Here is the present equation for the four DFSshgd or PHGphgd
characters.  If you have a better idea, let me know!

         P = 10 / s        For Power plots, P = p;  For DFS, P is INVERSLY
                              proportional to signal strength s.
         H = 10 * 2 ^ h    Convert character to power in Watts
         G = 10 ^ (g / 10) Convert from dB
         D = 45 * VAL(d)   Convert to degrees.  If D is not zero, then the
                              circle is offset in the indicated direction
                              by 1/3rd radius
         R = SQR(2 * H * SQR((P / 10) * (G / 2)))   radio range equation
                                                    modified by the additional
                                                    SQR(P/10 *G/2) to make it
                                                    unity at 10 watts and 3 dB
      
         R = R * .85  Present fudge factor

EQUAL FADE CIRCLE TECHNIQUE FOR MOBILE OMNI DFING:

     This method has been used for years by Airborne search and rescue teams
to locate downed aircraft based on the location of points where the signal is
just detectable.  The advantage of this technique is that NO BEARING info
and NO SIGNAL STRENGTH info is required.  The key factor, is that ALL points
where the signal fades to zero are located on the edge of a large circle with
the hidden transmitter at the center.  By simply flying (driving) through the
area of the hidden transmitter and plotting at least three points where the
signal fades out, you can identify the circle and therefore the location of
the transmitter.  For aircraft searches, this technique can be repeated at
lower and lower altitudes to repeatedly reduce the size of the circle and
therefore increase the accuracy.  For ground based searches, an attenuator
or tighter squelch can be used to reduce the size of the circle for successive
runs.

     The only assumption in this process, is that the radiation pattern
from the transmitter is relatively omnidirectional.  An advantage of this
this technique is that the aircraft does NOT have to fly over the transmitter
to find a signal peak (which is very ambiguous, considering that there is
often a NULL directly overhead of an OMNI transmitter).   See the following
plot to see how the data is plotted.  Between each pair of fade points, a
line is computed and then a line of bearing is drawn midway between the points
and perpendicular.  The intersection of these lines-of-bearing give the
location of the transmitter.  The sketch below is symetrical due to the
limitations of the angle of the slash characters used in drawing it, but the
technique does work no matter where the flight paths intersect the circle!

       Entry               .   .   .  Fade Circle
     Flight path      .                 .
             \     .                    *  .
               \ .       *           *       .  / Exit flight path
               A.\          *     *          D/
                .  \           T            / .
                .    \      *     *       /   .
                 .     \ *           *  /    .
                   .  *  \            / *  .
                   *  .    \        /    . *
                *          . \ . C/.          * Perpendicular
                             B \/               lines of bearing
                              /  \
                             |    |
                              \__/ oops, nothing heard,
                                   turn the other way!

     APRS has now implemented this algorithm.  No matter what pattern you
drive (or fly), simply drive until you first aquire the signal and hit the
F5 key.  Then continue driving in the same general direction until you just
lose the signal.  At this point hit F5 again.  APRS will then compute a line
of bearing perpendicular to the line connecting those two points and bisecting
the distance.  This perpendicular line of bearing is represented by the
asterixed lines above.  Turn and choose a new line to drive until you
re-aquire the signal and do the same process again.  Hit F5 on aquisition
and hit F5 again when the signal fades.  When APRS plots this second line of
bearing,  you will have two intersecting lines of bearing that roughly
indicate the location of the hidden transmitter.  Drive directly to that
point and insert enough attenuation in your antenna to make the signal weak
enough to do the whole process again but with a much smaller FADE circle.
This added attenuation is similar to aircraft reducing altitude to reduce the
fade circle for each additional run.

    Note that each time you press the F5 key to mark a fade point on the map,
APRS asks you if this is a NEW CONFIGURATION or not.  This is important,
because APRS should use only the points made by the same station and in the
same configuration for each plot.  To keep track of these, APRS labels each
new fade point with your callsign suffix in parentheses and then a letter
for the given configuration and then a sequential number.  Whenever the
MAPS-PLOTS-FADE commmand is given, APRS only computes bisectors and bearing
lines from each group of points from the same station, and from the same
configuration group (letter).  So, for any given configuration (antenna and
attenuation combination) just hit return at the configuration prompt.  When
either the antenna or attenuation are changed, then answer Yes for the first
point in the new configuration.

NOTE!   It is very important to understand that this is just a technique.
The operator MUST have experience in DFing and must thoroughly appreciate
the vagaries of propogation and antenna height-gain.  Just pressing F5 does
NOT find the FOX!  Give me a violin and it will NOT make music!  Garbage in
implies garbage out! ETC.  What I am saying, is to make sure that each time
you are ready to mark a new fade point, consider the average terrain and be
sure you are in a comparable propogation position.  Obviously, if you have
some kind of S-meter, you do NOT have to drive all the way to a fade
condition, but just to a measureable and repeatable signal strength level.
As long as you press F5 at multiple points of equal signal strength, the
fade technique will work.


FURTHER DETAILS:  When you press the F5 key for manual reports, APRS creates
a Fade marker at the location of the cursor.  If you are GPS equipped, simply
press the Go key first to move the cursor to your present location.  This
For each press of F5, a new fade spot is created.  Once APRS has two or more
of these locations, it can plot the lines of bearing.  Use the MAPS-PLOT-FADE
command to display the plot of all of the lines of bearing.  Although this
FADE circle technique is one of the neat optional features provided to
registered DF users of APRS, I have also made it available in the basic
package as well for up to three fade points so that everyone can try it out.
Registered DF users, of course, can plot any number of points.


PLEASE NOTE!  *******************************************************
              The difference between this technique and the OMNI-DF function
in APRS, is that the FADE circle technique takes advantage of mobile direction
finding stations to locate the edge of the FADE circle.  FIXED stations
can NOT provide ANY useful information for the FADE circle technique.  The
chances that they are exactly on the FADE circle edge is a chance in a
million.  Yes, they can induce attenuation to cause the signal to just fade,
but their exact sensitivity, antenna gain, and antenna height above average
terrain CANNOT be reproduced anywhere else, by anyone else, to find a second
or even third comparable point.  So that is the difference between the two
techniques.

The FADE circle is for mobile OMNI fox hunters, and the OMNI-DF capability
which plots signal strength contours is for fixed OMNI stations.


*****************************************************************************


     APRS DIRECTION FINDING WITH BEAM HEADINGS AND DOPPLER DF UNITS


     APRS is an excellent tool for instantly plotting and diseminating DF
bearing information.  APRS has several methods of obtaining lines of bearing
for plotting:

     MANUAL APRS    - Any APRS station simply selects the INPUT-DF command and
                      types in his beam heading

     MANUAL OTHER   - Any APRS station can take voice reports from other
                      stations, and place them as DF reporting OBJECTS on his
                      APRS map

     AUTODF D.S.Inc - Connecting your second COM port to the serial data out-
                      put of a Doppler Systems Inc system will automatically
                      plot and transmit the bearing of the FOX.
            
     AUTODF N7LUE   - An innexpensive APRS compatible interface to permit
                      connecting ANY doppler DF unit to the APRS serial port.

DF DEMONSTRATIONS:  To see the results of manual DF bearings in a Baltimore
foxhunt, FILE-LOAD the FOXDF.BK file.  You will see the multiple lines of
bearing all converging to within 1/2 mile of the final location of the Fox.
Notice that none of our stations were any closer than 15 miles away and more
than half of our DF stations were more than 25 miles away!  Notice too, that
none of these stations were particularly calibrated and only two stations
were actual APRS stations.  The others just reported their position and
bearing by voice and we put them on the map.  MAKE SURE you know how to
convert from magnetic to true bearings. We did it the wrong way and were
10 miles off the first time!

     To see what the AUTOmatic Doppler DF interface looks like, zoom into
Phoenix, Arizona and FILE-REPLAY the AUTODF.HST file.  You will see N7LUE's
DF unit make multiple hits on three local repeaters in the area.  If you are
doing a DF exercise, you can enable APRS to save all DF reports in a track
history file by setting the CONTROLS-POSFIL to off.  With the Position Filter
off, APRS will save every DF posit to the track history file.

CAUTION:  APRS does not do spherical geometry, it assumes a flat earth.  This
will not be noticable unless you attempt to use DF bearings beyond a few
hundred miles.  Even tracking balloons over 200 miles, this error will probably
be less than the typical innaccuracies of the average HAM beam antenna.  For
this reason, APRS will not draw a DF bearing line beyond 256 miles.


MANUAL APRS STATION DF REPORT:  Each APRS station can include a beam heading
in his position report by entering the INPUT-DF command.  Unless the station
indicates Permanent, this bearing will normally time out after 2 hours to
eliminate any confusion caused by old/stale reports.  A solid yellow line
indicates an excellent line of bearing, and a more dotted line indicates
less and less quality.  As a further aid, the MAPS-PLOTS-RINGS command can be
used to superimpose a set of range rings on the screen around any one station
for estimating distances for subjective analysis of signal strnegths.  If you
are running the WX station option, then the DF report will override your WX
station report with the Beam Heading report.

NON PACKET DF REPORTS:  Even for stations not running packet or APRS, their
lines of bearings can be quickly entered by any APRS station using the INPUT-
ADD command which adds them to everyone's map in real time.  In this case,
simply select the DF symbol, enter a beam heading, and enter a quality number
between 1 and 8, where 8 is best.

DUMB PACKET TERMINAL DF REPORTS:  Non APRS packet stations can also
automatically report their lines of bearing into the system by simply
entering a beacon text in the APRS format with their line of bearing.  The
details for the following APRS DF report format are included in the
PROTOCOL.txt file:

 BText  !DDMM.xxN/DDDMM.xxW\CSE/SPD/BRG/N0Q/DF report...

 Where: DDMM.xxN is Latitude, DDDMM.xxW is Longitude
        \ (Backslash indicates a Triangle symbol for DFing)
        CSE is course (000 for fixed station)
        SPD is speed (000 for fixedstation)
        BRG is the DF bearing in degrees True
        N0Q is a Quality indicator


AUTOMATIC DOPPLER DF UNIT INTERFACE:

     Randy, KA7UUS and Bob N7LUE developed a serial interface to the popular
ROANOKE Doppler DF unit (or any other DF unit that drives an LED display).
They have added a divide by N counter and a UART to produce a single ASCII
character report 8 times a second or so.  Each character is a letter from
@,A,B,.. ,O indicating the azimuth of the 16 LEDS.  For some DF units that
rotate counterclockwise, the board will optionally use lower case letters
for the opposite rotation.  A VOX circuit disables data output when there
is no DF signal, and an optional PTT circuit can be used to disable the DF
unit when ever a co-located TNC transmits the resulting DF data.  This last
circuit was necessary to prevent the DF unit from generating false bearings
whenever the packet TNC transmitted!

     APRS accumulates, averages and calculates the deviation of these samples.
It then plots a bearing line in the average direction and shows the variance
of the data by the "dottedness" of the line.  A solid line is a solid
non-varying signal, whereas a very dotted line, had a lot of variance in the
reports.  Since APRS averages the data and computes the deviation and
average to 1 degree, the fact that the DF unit is only reporting in 16ths
of the compass is averaged out.  Anyone who has watched a doppler DF unit in
action, understands that the signal bounces everywhere due to reflections
and the distribution of the data is broad enough that the quantization of the
raw data to 4 bits is insignificant.  The add-on N7LUE universal APRS serial
interface is available from N7LUE at the following address:
     Robert Swain, N7LUE        410-766-2494 eves
     820 38th St West
     Bradenton, FL 34205

Marty Mitchell, N6ZAV at 340? Otero St, Costa Mesa, CA 92626 is selling an
improved version of the ROANOAK DF unit.  His phone number is 714 760-6060.

REMOTE DF SITE:   ALthough any APRS site with the DF interface can be an
automatic DF station, the APRS PC computer can be eliminated for remote
site operations.  All that is needed is a DF receiver, the DF unit and
serial interface, and a TNC and packet radio.  By setting the TNC in the
UNPROTO CONVERSE mode, it will simply packetize the data out of the DF unit
periodically for display on all APRS stations on the network!  It is simple
to configure the TNC to do this as follows:

  A.  Take the 8 characters per second data from the DF unit and connect
them to the serial data input of the TNC.  Take the PTT output of the TNC
and connect it to the optional PTT-SUPPRESS input of the N7LUE interface
to prevent the DF unit from generating erroneous data when the TNC transmits
(and overloads the DF unit).

  B.  Set the TNC packet length PACLEN to 75.  On a continuous signal, then,
the TNC will transmit once every 10 seconds after it has accumulated a full
packet of 75 characters.  Each transmission will contain the last 75 samples
from the DF unit!

  C.  So that APRS knows the location of the remote DF unit and that it is
a DF station, the BText of the DF TNC must contain the LAT/LONG and the APRS
DF symbol character (\):     BT !3856.55N/07629.11W\DF station...

  D.  APRS will then plot a new bearing line for each DF packet received.

  E.  For short FOX transmissions, the TNC should have PACTIME set to AFTER
10 (1 sec) and CPACTIME to ON.  The PACTIME setting was chosen relatively
short so that a packet is transmitted at the end of each FOX transmission,
but before another station keys up.

  F.  To prevent all DF sites from keying up at once at the end of the FOX
transmission, each automatic DF site must have a differnet value of DWait.
Each additional site should have an additional 100 ms.

   With the design noted above, each DF site will transmit a maximum of one
packet every 10 seconds, or one packet for every short transmission of the
fox.  With the parameters chosen above for 5 stations, the network would be
pretty well saturated just handling the data from all sites.  This is fine
for intensive operations in search of a FOX or jammer, but a more routine
level of operation could be realized by reducing the data rate from the the
DF unit from 8 to 4 characters per second or less.  This would result in
only one packet report every 20 seconds or more which might be more suitable.
At these high data rates, and since a good DF site should have good altitude,
digipeater paths for routing the data should be avoided if possible.


AUTOMATIC REMOTE SITE DF NETWORK CONTROL:

Since the automatic DF interface between a TNC and a DF unit will generate a
lot of packets, there has to be some means for remotely turning it on and
off.  I consider that beyond the realm of APRS, since for a remote DF site,
there must already be some kind of control link in place in order to command
the DF receiver what frequency to listen to.  If such a link already exists,
then the capability is probably also there for enabling or diasabling the
DF/TNC interface.

     In the absence of such a control link, however, a very simple remote
control and receiver command link can be derived from the TNC itself!  First,
take the voltage from the CONECTED LED and use it to enable the DF unit
output to the TNC input (some TNC's bring this signal out on one of the RS-
232 pins).  This way, the automatic reporting will begin as soon as the DF
Net Control station connects to the TNC.  This means of control also has the
advantage of using the serial data channel from the DF Net Control SYSOP up
to the site for setting the frequency of the receiver!  Since APRS software
only checks the TO address for valid APRS data, and does not care whether the
packet is connected or not, it will still be able to monitor all data from
the remote site.  To facilitate this process, APRS now also accepts packets
addressed to DFNET which should be used as the callsign of the NET CONTROL
station.  This is legal, as long as the NET CONTROL station also places his
true call in his BText and sends his beacon once every 10 minutes.

DF NET CONTROL OPERATION:    The scenario for this kind of operation, would
be for the network SYSOP to use a dumb terminal in the multi-stream connect
mode to connect in turn to each of the remote sites.  Once each of these
connections is established, each DF station begins sending DF data as long as
the connection is in place.  To disable a site, the SYSOP simply disconnects
from that station.  The only disadvantage of this means of control is the
additional QRM on frequency from all the ACKs required from the SYSOP TNC for
every DF packet transmitted.   Having an alternate means of control, avoids
this CONNECTED environment but adds complexity.


MOBILE APRS DIRECTION FINDING

    APRS is the ideal tool for integrating together all of the DF equipment in
modern DFing, the Doppler DF, the GPS, and the TNC packet link.  If you have
a dual serial port LAPTOP computer, the first serial port is connected to the
TNC and if available, a GPS, using the Hardware SIngle Port mode.  The second
COMM port is dedicated to the DF unit.  With this arrangement, the GPS provides
continuous data on the location of the vehicle and the TNC provides the
communication links to the APRS DFing network.  The DF unit provides the
DF data whenever the FOX transmits.  With the GPS data, APRS will do an
automatic conversion from the relative bearings from the DF unit to the
heading of the vehicle.  With this arrangement, the mobile DF unit will be
seen in the APRS network, moving along and providing constant bearings to the
hidden transmitter.  In practicality, however, there are problems in this
plug-and-play scenario.

  1)  First, The heading information from the GPS is ONLY ACCURATE, AS LONG AS
  THE VEHICLE IS MOVING!  When the vehicle stops, the GPS has no way of
  computing heading.  Therefore, the heading information is meaningless.  To
  minimize this error, APRS will only use the LAST Heading for which the
  vehicle velocity was over 10 MPH.  The problem is, that as the vehicle
  steers from that heading, the DF data that he is plotting and transmitting
  will be incorrect.

  2)  Whenever the TNC transmits APRS DF or position data, it totally garbles
  the DF unit!  Even placing the packet network on a different band, still
  garbles the DF unit.  To solve this problem, the DF unit should be wired
  to the TNC PTT lead so that the DF unit is DISABLED whenever the TNC is
  transmitting.  Diode ORing of the PTT leads of every transmitter in the
  vehicle would prevent any self-generated DF errors caused by other radios
  in the vehicle.

  3)  Sometimes the GPS is obscured or otherwise is not putting out good
  fixes.  More often, the DF unit is putting out GARBAGE!  With everything
  running automaticallly, the driver usualy knows when the data is good and
  when to ignore it, but the garbage data is still being processed by APRS
  and being transmitted to everyone else.  I would guess that more erroneous
  DF bearings would be transmitted than good ones!

  4)  Most laptop computers only have one usable COMM port!

   As I began to explore various ways for all three devices, the DF, TNC and
GPS to share the same COMM port, Joe Moell K0OV suggested that it should all
be manual anyway!  By having two push buttons on the dash board, one labeled
GPS and the other labeled DF, the operator could easily determine when he
wanted GPS data, and when he wanted DF data.  All the rest of the time, he
would be connected to the TNC and be in communication with the rest of the
APRS team.  This way, the only DF reports injected into the DF network would
be under the control of the operator, and he could make the complex judgment
whether he was seeing good DF data, and whether his vehicle was making a good
consistent heading!  Also, with this arrangement, no complex interface is
involved, other than two DPDT momentary push button switches.  The following
schematic shows how the serial data from all three devices is switched and
how the second pair of contacts is used to tell APRS, which device is connected
at any time.

                     POSITION               DF
                        S1                  S2             LAPTOP

TNC DATA >-----------*
                     V
                    ---------*------------*
                                          V
                     ^                   ----------*----------> RXD
                     |
GPS DATA >-----------*                    ^
                                          |
DF DATA  >--------------------------------*

                                   diode
TNC DTR  <-------------------*------->|------------*
   (RTS)                     |                     |
                    ---------*           ----------*----------> DSR

                     ^                    ^
                     |                    |
                   -----                -----
                   /////                /////

     The second pole of the DPDT push button switches the DTR pin of the TNC
to ground to force it to buffer all data while either of the buttons is
pressed.  Some TNC's use the RTS line for this purpose (PACCOMM) so check your
TNC manual.  Also note that the DSR or RTS line should have an internal Pull-
UP resistor inside the TNC.  SImilarly, this ground signal is steered by the
diode to the DSR input of the LAPTOP so that APRS can detect when the DF unit
is connected.  APRS can distinguish between the TNC and DF data using the
normal HSP logic, but it needs this DSR signal to let it know to process all
incommming letters as DF data.  Notice that this circuit also eliminates the
need for the normal APRS Hardware Single Port mode two-transistor switch.

     Actually, other switches might be added to swap the GPS and TNC lines
so that operation of the GPS and DF unit alone, without the TNC, could be
accomplished.  In this case, S2 is left in the DOWN position, so that DF
data is constantly displayed, and S2 is released momentairly to get a GPS
fix.  Or, you might also want to enable true GPS HSP mode so that APRS can
track you hands free as you drive to and from the area of interest.

CONFIGURATIOIN:  Since APRS does not need to toggle the normal HSP interface
to switch on the GPS, APRS should be in the SINGLE-PORT-MODE, vice the HSP
mode.  In the SPM mode, APRS will always look for GPS data interleaved with
TNC data, and there is no internal timing going on in the program.  Any time
the operator presses S1 (and holds it for 2 seconds), APRS should see the
GPS data and grab a fix.  Similarly, if the APRS program also has DF-SINGLE-
PORT mode selected, it will watch the status of the DSR line, and whenever
DSR goes low, it will then process all incomming data as DF data!. (this last
function is not yet implemented, Im waiting on someone that needs it).

