FIGURE 1.1
Meteoroids of all sizes are continuously raining down upon the Earth. The visible phenomenon of a meteor is due to incineration of a sand-sized particle by friction with the upper atmosphere. Although they are among the smallest bodies in the solar system, meteoroids are important because they are related to comets and asteroids. Comets are considered to be the most pristine remnants of the original solar nebula, while asteroids are thought to have come from an early planet that fragmented or one that never coalesced. Thus, meteors and meteoroids are associated with the early history and evolution of our solar system.
A few massive meteoroids survive their passage through the atmosphere and wreak destruction on the ground. Some large basins, such as Sudbury in Canada, are thought to be the scars of meteoroid impacts. Other lesser depressions, like the Meteor Crater in Arizona, are known to be the result of meteoroids because of the presence of material from the impacting body. Meteoroids have become a important topic in biology since scientists from diverse fields reached consensus that a very large impact about 65 million years ago caused global climate changes that led to the extinction of the dinosaurs.
Some exo-biologists have proposed that life originally came to the Earth on a meteorite. This theory has taken on new importance since the discovery of possible fossils in a Martian meteorite.
Meteor study is not a very old science. In the early years of the twentieth century observations were carried out visually under the direction of C. P. Olivier. Later, photographic results were obtained by a team led by F. L. Whipple. These data provided a basic understanding of the orbits of meteors.
With the development of radar during World War II, much more precise and continuous meteor observation and orbit analysis became possible. A good summary is given by W.J. Baggaley (Earth Moon and Planets, vol 68, pp 127-139, 1995). From the late 1940's through the 1960's, systems operated by A.C.B. Lovell at Jodrell Bank Experimental Station in England, Z. Sekanina at Havana, Illinois, and others made extensive records of meteor activity. Daylight observations were possible for the first time, and led to the discovery of several intense new streams of meteors.
During the period 1970 to 1975 radar observation was carried out in Russia. Baggaley does not list any further northern hemisphere observation since that time. However a southern hemisphere monostatic radar was operated by L.M.G. Poole from the late 1980's until the mid-1990's. A full scale southern hemisphere radar has been run by Baggaley from 1990 onward.
There is an observational method similar to conventional radar, but much simpler in concept and easier to implement. Forward scattering (FS), like radar, depends upon on the well known phenomenon that frictional heating of meteoroids produces a trail of ionized air that can reflect radio waves. In radar the transmitter and receiver are in the same locality, while in FS the transmitter is beyond the receiver's horizon. FS does not require a dedicated transmitter, since there are already enough sources of radio propagation, as explained in the next section. While FS provides comparatively little information on meteor trajectories, it is orders of magnitude less expensive than radar.
Ham radio operators have used radio reflections for many years in order to receive signals from distant transmitters. Directed communication of short messages can be accomplished by FS from meteor trails. Thus some secure communications also use meteor trail reflections. However, scientific content regarding meteors is unavailable from this data because of its proprietary nature. FS has also been used for a number of years to investigate the properties of individual meteors, though there is no mention of systematic meteor counting in the literature.
The best portion of the spectrum for radio meteor observation is in the Very High Frequency (VHF) range from about 40 to 70 MHz. Below 30 MHz the ionosphere reflects all radio waves, so meteors have no effect; while above 80 MHz the reflectivity of meteor trails declines. Fortunately, most countries have powerful commercial television stations broadcasting continuously in this frequency range, and they can be used in meteor echo work.
The video carrier signals for channels 2, 3, and 4 in the United States were assigned to the frequencies 55.25, 61.25, and 67.25 MHz in the early days of television. Later on, two more frequencies were added for each channel at 10 KHz above and below the original. Thus stations in nearby cities all broadcasting on channel 3, for example, would be assigned to 61.24, 61.25, and 61.26 MHz. This reduced interference between stations, some of which was due to meteor echoes.
Since the signal from each transmitter has an audible tone associated with it, one can hear the presence of a video carrier. Even a station beyond the horizon will deliver some signal through the troposphere the so-called ground wave. A good candidate frequency for meteor detection can be determined by listening to the tones. Local stations will deliver very loud tones, and are not suitable for registering meteor echoes. Distant stations will produce a very faint tone, which can be greatly amplified by the passage of a meteor between the transmitter and receiver. Thus the presence of a weak audio tone indicates a candidate transmitter suitable for meteor echo observation. Channel 3 at 61.24 MHz was chosen for the system described here. The nearest transmitter is about 200 km away.
In order to monitor the flux of meteors continuously and economically, an Automated Meteor Counting Apparatus (AMCA) has been developed. This radio/computer combination makes use of the signal enhancement associated with an ion trail reflection in order to count meteors, as shown in Figure 1.1. The receiver is connected to a personal computer (PC) by means of an analog-to- digital (A/D) board. The AMCA receiver has a jack connected to its automatic gain control (AGC), which gives access to the voltage that drives the signal meter. Thus a change in signal intensity alters the voltage across the AGC jack, and this voltage change is measured on one channel of the A/D board. Voltage is sampled at 10 Hz and a sudden increase by several dB or more alerts the system of a possible meteor echo in progress.
Another channel of the A/D board is connected to the receiver's audio output. This interface serves to distinguish true meteor echoes from other types of signal enhancements based upon analysis of the audible tone. Spurious electrical activity such as lightning and circuit switching would otherwise be counted along with the meteor echoes and would invalidate the results. The AMCA system is unique in combining signal intensity with audio tone analysis for meteor echo detection.
The audio signal is sampled at 9000 Hz. After 0.1 seconds the samples are autocorrelated in order to find the frequency and power of the video carrier's tone. Correlations of the data spaced from 12 to 24 samples apart are computed, thus covering the audio frequency range from 375 (=9000/24) to 750 (=9000/12) Hz. Tones at high frequencies can also be recognized due to aliasing. The receiver's beat frequency oscillator can be adjusted to accommodate tones at lower audio frequencies. Since the ground wave is always present to some degree, the correlation coefficient is normally between 10 and 60 percent. This sampling and analysis is repeated at intervals of 10 seconds to update the reference correlation which drifts with time.
A signal enhancement during the subsequent 10 seconds triggers an additional audio sampling and analysis. If the signal increase is due to a meteor echo, the faint video carrier tone rises above the noise background and the correlation coefficient will increase usually exceeding 90 percent. On the other hand, a signal enhancement due to electrical noise reduces the correlation. Thus, triggering due to a meteor may be distinguished from that due to noise by a comparison of the triggered audio correlation coefficient with that of the reference correlation.
The variable intensity echoes from an airplane overflight can also trigger audio sampling, and they may pass the autocorrelation test. However, overflights can be recognized because they begin with a small intensity echo, then build to larger strength and greater variability. Thus AMCA can be set to eliminate airplane echoes by requiring one second of quiet signal before allowing a trigger. Running in this quiet mode greatly reduces the number of false triggerings as well.
The remainder of this document describes AMCA in more detail. Section 2 elaborates on the hardware components and their interconnections. Section 3 describes the software as seen from the perspective of a programmer. Section 4 tells how to use the program. In Section 5 a sample of the data obtained during system testing is presented.