Sometime in the not too distant future, a modern SAC bomber, flying a routine mission, reports aircraft status to the command post via UHF satellite communication relay link without complication or problem. But toward the end of the mission the aircraft develops an emergency. A crewman punches the buttons to send a “Mayday” call, with aircraft position information, via the satellite link. The command post receives the message, but for some reason it is unreadable. Where has the system failed? Is it an equipment problem or maybe an operator error? No, the answer is probably “generation gap.”

Each new generation of communications—from semaphores to smoke signals, to the wireless, and now to satellite communications—brings on a set of unique problems. While the smoke signal provided greater communication distance than the semaphore, it was susceptible to wind and rain effects. The wireless overcame these obstacles but was affected by multipath and static. Satellite communication solved multipath and static but generated a set of its own problems. With the advent of the Air Force Satellite Communication System and the Navy’s FLEETSATCOM system, the problems of satellite communications are being recognized as operational limitations of a new generation of communication technology.

One of the new problems that must be contended with is “ionospheric scintillation.” Scintillation of a star is the twinkling that results from light rays bending as they pass through the inhomogeneous atmosphere. Scintillation of a radio signal is the “twinkling” that results from radio waves passing through an inhomogeneous ionosphere. This twinkling can cause erratic reception of a radio signal and can disrupt vital communications for hours.

In order to develop techniques that circumvent the problems associated with ionospheric scintillation, an effective model must be developed. Much previous work has been done in this area, especially by radio astronomers. But, in general, the models fail to describe fully all parameters associated with observed scintillation.

The particular type of scintillation observed is due to the behavior of the ionosphere. More specifically, ionospheric scintillation appears to be due to irregularities in the F layer of the ionosphere. This layer, which extends from 60 to 500 miles above the earth’s surface, does not appear to be completely uniform. Shortly after sunset small irregularities in the ionosphere are amplified as the ions interact with the magnetic lines of force. It is believed this interaction causes the ions to concentrate in cylindrical shapes that align along the magnetic line of force. These cylinders are typically 100 to 1000 feet in diameter and 10 to 100 miles long.

The behavior of ionospheric irregularities is a very complex function of a number of variables. The gross factors that appear to influence the behavior include location on the earth, sun spot activity, season, and time of day. A host of minor factors also affect these irregularities. The ionospheric scintillation model becomes more complex and less predictable as a result of the large number of variables.

While ionospheric scintillation occurs all over the globe, scintillation associated with the equatorial region has received the most attention for a number of reasons. First, the depth of fading caused by equatorial ionospheric scintillation is generally greater than for mid-latitudes or polar regions. The fading often reduces the signal to one one-thousandth of its unfaded value. A second reason is that there is more opportunity to observe the effects in the equatorial region, as it is more densely populated than the polar region. Since the satellite is starting to play a progressively greater role in military communication and navigation, the Air Force is funding equatorial testing. Also, since the effects of scintillation are less in the mid-latitude and polar regions, it is generally agreed that, once the scintillation problem is solved in the equatorial region, these solutions can be used in other regions.

Researchers have learned much from the host of experiments that have been performed. It is known that equatorial scintillation activity increases during the spring and fall equinox periods and exhibits a broad decrease during the summer and winter solstice periods. The scintillation is primarily a nighttime effect.

The onset of scintillation fading normally begins abruptly one to two hours after sunset. As the radio waves penetrate the disturbed region, they are focused by the high discontinuities resulting from ionospheric irregularities. (Figure 1) These irregularities tend to drift in an east-west direction, causing the fading to drift by a ground or airborne terminal. A disturbed ionosphere area as small as 25 miles in diameter or as large as 2000 miles in diameter drifts at speeds approaching 100 miles per hour. Within this disturbed area are many discontinuities or fade-causing irregularities. A ground terminal sees the effect as slow fading, during which the station may be blanked out for several seconds each time a discontinuity goes by. An airborne terminal generally is moving several times as fast as the irregularities and therefore sees a fade that is much shorter in duration. Since the disturbed area varies greatly in size, the periodic fading may occur for minutes or hours as the irregularities pass by. The effect of scintillation is most pronounced for those frequencies normally used for aircraft communications (HF, VHF, and UHF). The scintillation fading characteristically begins abruptly and ends abruptly. (Figure 2)

Figure 1. Ionospheric scintillation fading model. The equatorial ionosphere (200-mile altitude) contains high ion contractions in horizontal cylinders 500 feet in diameter and tens of miles long, aligned in a north-south direction. The signal from a communication satellite are focused as they pass through the ionosphere. Fade and enhanced areas occur on the earth as a result of the focusing. The ionospheric irregularities—and consequently the fading —drift in an east-west direction.

Figure 2. A recording of the signal received in an airborne terminal from a satellite shows the abrupt end of UHF ionospheric scintillation fading. Times starts at the right and increases to the left. At the start of the segment shown, the fading is greater than 25 db (from —115 dbm to —140 dbm). Duration of an individual fade is about one second. Toward the left side of the chart the fading goes from maximum to none in a period of ten seconds, demonstrating the abrupt end of the fading condiotn.

The original scintillation testing was performed at ground stations. More recent experiments have involved specially instrumented aircraft working in conjunction with ground stations to uncover problems unique to the mobile terminal. By coordinating the data at the airborne and ground stations, we have developed a clearer picture of the ionospheric model.

In a recent joint Air Force/Navy test in the western Pacific, airborne and ground data were collected from the 250 mega-hertz (MHz) downlink beacon signal from the TACSAT satellite. Severe fading was recorded on 12 of the 17 night test periods. During about 7 percent of the total time, the fading was severe enough to disrupt normal satellite communications. This averaged more than 1½ hours a day, which could constitute a serious operational problem. During the severe fading period the signal did not always remain in a faded condition but went through periodic fade and enhancement at such a rate that normal communication could not be carried on. During the fading period the signal amplitude followed a curve, depicted in Figure 3. This plot shows that the signal was enhanced by 8 decibels (db) over its average value about 1 percent of the time and was faded 10 db below its average value 3 to 5 percent of the time.

Figure 3. Comparison of amplitude distribution for airborne and ground data.

Figure 3. Comparison of amplitude distribution for airborne and ground data. The figure is a plot of the amplitude of the signal received during a scintillation fade, similar to the one shown in Figure 2. If the amplitude is sampled about 100 times per second for the five-minute fade period, the cumulative percentage of the time the signal was at various levels follows the plot of this figure. The curve for the airborne data taken in the same area and at the same time as the ground data is statistically the same as for the ground data.

In order to confirm the ionospheric scintillation model and to determine the length of time an individual irregularity remained identifiable, data were recorded simultaneously in a ground and airborne terminal in the same vicinity. As predicted, the data showed the fading in the aircraft to be identical to the fading occurring on the ground if the effect of the aircraft velocity was considered. To do this airborne data were digitized and “slowed down.” With computer analysis, various amounts of “slowdown” were tried as the airborne data were compared with ground data taken in the same area. Finally, the correct “slowdown” was determined, and good correlation of the airborne and ground data was obtained by comparing the original ground recording with the “slowed down” airborne data. (Figure 4) These results showed that the irregularities that caused the fading remained “coherent” for a period of more than 10 minutes as they drifted some 10 to 20 miles over the earth’s surface.

Figure 4. Comparison of airborne and ground scintillation fading data. The airborne fade data, taken in the vicinity of the ground station, display a much faster fade rate than the ground fading because of the aircraft velocity. If the effect of the aircraft velocity is removed from the data, the exact fade that occurred on the ground can be seen in airborne data. The two sets of data correlate very well on the left-hand side of the plot but start becoming uncorrelated at the right side of the data.

A more complete model of the ionosphere has been developed with the results of this testing. However, there are still many unknown factors in the model, which will continue to surprise users of communication satellites. Using the new model, how do we propose to overcome the effects of ionospheric scintillation fading? It appears that frequency diversity, the technique of transmitting the same information on two frequencies, will not improve the communication reliability. Likewise, using two antennas to receive the signal would require antenna separations of several thousand feet to improve the reception. These distances are clearly impractical on an aircraft. One technique that does offer promise is message repeating or very long error correction coding. When a communication terminal is in a disturbed area and experiencing periodic fading, the communication signal is lost for only 10 to 25 percent of the time. The remainder of the time the signal is normal amplitude or enhanced. A simple repeat technique would be to store the message and transmit a small segment, possibly a one-second portion, for 10 to 20 times in a row. The next segment of the message is then sent and so on until the complete message has been transmitted. The receiver, likewise, stores the segments and looks for a repeated segment that would get through during the unfaded portion of the time. These recognized portions are recombined and represent the complete message. Such a technique slows down message transmission rate by the number of times the segments are repeated but should provide a reliable means of overcoming the fading.

In the future the satellite will play a more prominent role in military and civilian communication and navigation systems. Air traffic control satellite communication systems will provide a reliable means for monitoring aircraft status and location during overwater flights. Military systems such as the Air Force Satellite Communication System will allow positive control of the airborne forces worldwide. With satellite communications come the phenomena of ionospheric scintillation fading. Since satellites will play a dominant role in future communication systems, problems such as ionospheric scintillation will continue to receive considerable attention.

Allen L. Johnson (M.S.E.E., Northeastern University; M.B.A., Ohio State University) is a Group Leader, Air Force Avionics Laboratory, AFSC, responsible for flight-test evaluation of the airborne communication systems developed by AFAL. During the past 13 years at Wright-Patterson, he has been responsible for developing and testing several UHF and microwave air-to-air and air-to-satellite communication systems. He was previously employed by the Bell Telephone Laboratories. He received the Scientific Achievement Award for the 1971 Ionospheric Scintillation Test.

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

The Effects of Ionospheric Scintillation on Satellite Communications

The Effects of Ionospheric Scintillation
on Satellite Communications