METEOR SCATTER PROJECT UPDATE: DECEMBER 1991
by John Nordlie
Work on the meteor scatter automated receiver system continues in
the design phase. Design decisions to date have been to use the
FM commercial band and an out-of-sight commercial station for
transmission. Work of others using this concept has yielded
significant results.1 A method of determining the velocity of an
incoming meteor by signal process of the reflection is also being
studied for possible incorporation into the system.2 Total cost
of the system has not yet been determined, since it is
unclear whether funds will be available for a dedicated
microcomputer and a commercial data acquisition board. Another
key component, the FM radio receiver, has yet to be selected.
Difficulties of the selection come from finding a receiver with
the necessary options for the project (synthesized digital
tuning, an analog signal strength meter, and an automatic gain
control (AGC) circuit that can be switched out or otherwise
disabled). However, other components have been selected, and are
readily available on the market. These include a directional
gain antenna of the Yagi design (Radio Shack #15-1636, $16.95), a
mast mounted low noise pre-amplifier (Hamtronics GaAsFET pre-amp,
$59), and hardware to mount the antenna and route the signal to
the receiver (estimated ~$50). If the microcomputer and
commercial data acquisition board are added the cost increases by
about $1300. This may not be necessary if a Zenith microcomputer
can be borrowed from Computer Science, and a simple data
acquisition circuit built that provides the needed data
resolution and range.
To summarize:
System costs (not including receiver)
With PC & D/A board $1425.00
With Zenith & homebrew A/D circuit (~$30) $160.00
Using the Zenith based system will present some special problems
in terms of data storage. The Zenith computers have no expansion
slots, and are only equipped with two 360 kilobyte disk drives.
Using some kind of efficient data reduction and compression would
be necessary with these computers. For example, using a data
sampling rate of 10 Hz (ten samples taken per second), using one
byte to store each data point, will generate 843.75 kilobytes of
data in one 24 hour period of operation. This is clearly
unacceptable in terms of data storage. Instead of compiling a
data file of signal amplitude versus time, the program might
determine the speed of the meteor, the time it fell, the
amplitude of the signal it reflected, and the length of the
signal. This would require about 14 bytes per meteor, or about
16.4 kilobytes per 24 hours if the rate of meteor detection was
slow (50 per hour), to 164 kilobytes per 24 hours if the rate is
high (500 per hour). This would be easily in the range of the
two 360 kilobyte floppy drives. Other schemes of waiting until
the computer detects a meteor event, then doing direct radio
signal amplitude measuring are also possible, though that would
take more data storage space than the previous method (but not
nearly as much as the first). Some experimentation will, of
course, be necessary to determine the best tradeoff between
maximum data retention and practical storage of the data
generated. Some data compression algorithms used in computer
science would be of great help here, if the computer used is fast
enough to do all the necessary calculations without interfering
with data acquisition.
To summarize, I think that significant
progress can be made on
the project in a short amount of time after an appropriate
receiver is chosen and purchased. All other equipment is readily
available and reasonably priced. Software development should, in
theory, present no serious difficulties.
REFERENCES
[1] Pilon, Kenneth V. "Radio astronomy with a home computer".
Journal of the British Meteor Society, Vol. 14 No. 2.
[2] Mackenzie, Robert A. "More about meteor radio scatter".
Journal of the British Meteor Society, Vol. 14 No. 2.
Back to Papers and Publications.