Home‎ > ‎Search Patterns‎ > ‎Aural Null‎ > ‎

Radio Propagation

This page will provide information on Radio Propagation intened for Search and Rescue volunteers performing electronic searchs for Emergency Locator Transmitters (ELT) and Emergency Position Indicating Radio Beacons (EPIRB) that transmit a continuous VHF homing signal. For a general radio propagation there are many other fine sources of information on the World Wide Web.

One migth rightfully ask if a Search and Rescue volunteer need to know about radio propagation, and if so how much? Frankly if a volunteer is willing to adhere faithfully to the published search protocols, and never encounters anomalous conditions, then the answer is little or none. However, a volunteer who wishes to really understand electronic search to be able to deal with anomalous conditions should be familiar with at least what is published here. You can read the story
of what can happened when search crews mislocated an ELT based on signal information in the Transportation Safety Board of Canada's report A01W0261. You may also be interested in reading what Paul D Turner, a member of the Civil Air Rescue Emergency Service (a member unit of the Civil Air Search and Rescue Association of Ontario), had to say about what it takes to perform an electronic search for an ELT and the difference between locating an UNSAR ELT activation and a distress ELT activation.

Radiation Pattern

Most discussions of electronic searchs concentrate on the radiation pattern of the emergency transmitter. Indeed the radiation pattern is important but it is not the whole story. The radiation pattern is a description of the radio energy envelope the transmitting antenna produces. It is usually presented as a polar or shperical graph with the azimuth, elevation, or both plotted against the amount of energy radiated in a particular direction. The amount of energy is normally represented on the graph by the distance of the point from the origin of the graph rather than the distance of the point along a single axis. This form of display makes the radiation pattern easier to visualize, but it may be confusing to people who expect the amplitude, or strength of the signal to be displayed on Y axis (for two dimensional graphs) or the Z axis (for three dimensions). It is also very important to remember that there are no spacial values (distances) depicted on a ratiation pattern graph.

To begin there are two radiation patterns that we need to be familiar with: isotropic; and monopole. An isotripic radiation pattern (on the left)
is produced by an isotropic radiator which is a theoretical (and thus imaginary) antenna with no physical dimension that radiates energy equally in all directions in three dimensional space. A Google (or other search engine) querry for isotropic radiation will result in many good references and images for isotropic radiation patterns. The graph here shows a hemi-sphere because an emergency transmitter is most likely to be located on or very near the surface of the earth when volunteers are looking for it. The bottom half of the sphere would be underground.

A monopole antenna, more commonly known as a whip antenna (on the right)
, is the most common antenna used with emergency transmitters. The perfect monopole radiates energy at the same amplitude for all azmuths, but in varying amplitude depending on elevation. This results in a radiation pattern that is a torroid (or doughnut shapped) centred on the antenna. Again Google will provide many examples of this radiation pattern, and discussion of the antenna. You may click on any of the graphs on this page to see the full size plot. These two graphs themselves are linear, but the colour gradients are in decibels.

Antennas also have the property of reciprocity which states that the receiving pattern of an antenna (sensitivity as a function of direction) is identical to the far-field radiatiion pattern when used for transmitting.

As stated above, the radiation pattern of the transmitter is important, but other factors have equal if not greater importance. The radiation pattern of the receiver can be more important. For example directiion finding (DF) equipment such as the El-Per depend heavily on the radiation pattern of the two or four element annena array. We will deal with the radiation pattern of the receiver in greater detail later, but for now, because the search receiver is usually in motion with six degrees of freedom we will remove some complexity by assuming it is isotropic. Other factors that must be taken into account are: path loss, polarization factor, line of sight and fading. If you aren't familiar with these terms don't worry, each will be discussed as we progress. When all these factors are applied to the radio signal the result is a pattern that is unlike either the radiation pattern, or the receiving pattern. Let's call it the Reception Pattern.

Decibels

We should probably talk a bit about the decibel (abreviated dB) before we go any further. When viewing, or discussing signal strength numbers we often must deal with values that vary over a factor of a million to one, or more. For example it is common for a receiver to have to properly receive a radio signal that will result in a few microVolts at the antenna input, and shortly afterwards properly receive a signal that will result in a few Volts. Using the decibel makes thinking about, visualizing and handling these number far easier. Wikipedia has a very good article, so you should read that if you are unfamiliar, or we get out of your comfort zone along the way.

Reception Pattern

Of the four factors we are going to look at, perhaps the most dramatic, from a search and rescue perspective, is path loss. Radio signals, just like visible light, are subject to the inverse square law. This means that if you are receiving a signal from a transmitter at some distance d away, and you then move to a distance 2d twice as far away, you will receive only one quarter 1/4 of the energy as you did at the first location. In practice this results in the signal strength dropping off very rapidly as the receiver moves away from the transmitter. Radio and antenna engineers use the Friss transmission equation. For example the reception pattern for an isotropic transmitter and receiver as the reciver moves in a grid pattern at a fixed altitude over the transmitter, taking path loss into account would look like the image to the left. This is a plot of the reception pattern (radiation pattern with path loss factored in) for an isotropic antenna at ranges from 0 to 50 km on the X and Y axis, signal strength in dB on the Z axis. The range from 0 to 50 km was chosen to highlite the profile along the axis as the receiver approaches, or leaves the transmitter. It is easy to see that the reception pattern is not spherical. The other three quadrants are symetrical around the Z axis.

We can look at the same plot for a monopole antenna.
The image to the right is an idealized 1/4 wave monopole with an infinite ground plane but gain adjusted to be on the same scale as the isotropic antenna above. This antenna has the zone of maximum radiation at an elevation of zero degrees. There isn't much difference between the reception patterns of the isotropic antenna compared to the idealized 1/4 wave monople. This is because over the distances plotted here, most of the time the receiver would be at an elevation of a few degrees or less. This keeps the receiver in the part of the radiation pattern that is nearly the strongest. The difference from an isotropic antenna in that region is too small to notice on this scale, especially since the monopole gain has been adjusted. If you look closely at the peak of the signal, near X = 0, Y = 0 you will see that it is lower than the peak of the isotropic plot. This is because the centre of the toroid (the hole in the doughnut) causes a significant drop in signal streangth directly in line with the axis of the antenna. This signal drop is often called an Aural Null. It is not certain that is where the name of the search pattern comes from, as the term aural null is used to refer to a number of phenomena that cause signals to drop out. The aural null can be more clearly seen at shorter ranges, like the image below on the left, ploted with the same parameters as the first monopole graph, but on a range of 0 to 1 km.


It may appear to you that the 'hole' is much smaller than one might assume from the depictions of the radiation pattern. This is because the path loss equation is dominated by the distance of the receiver from the transmitter. For an aircraft flying around a transmitter, for any given altitude, the distance from the transmitter is going to be the least when directly overhead. Unless the aircraft is a space craft, or perhaps a U2 spyplane, the distance overhead due to altitude will always be small when compared to the distance when some number of miles or kilometers away horizontally. The next image at the right gives a better view of the aural null in the centre of the pattern. The X and Y axis are in the range of +/- 500m, the altitude of the simulated search platform above the transmitter for all these graphs is 600m so it is easy to see how small the aural null really is.


The final graph in this series, centred below, shows the reception pattern out to 500km. This shows that although the decrease in signal strenth as the receiver moves away from the transmitter falls off sharply at the beginning, the rate of decline becomes less the farther away the receiver gets. Of course at some point, depending on how much power the transmitter puts out, and how sensitive the reciver is, the signal will become so weak that the receiver will not be able to detect it. Emergency transmitters are not particularly powerful, but modern receivers are very sensitive. To put this in context, on a clear night with the naked eye one can see stars that are very far way. Now stars are many orders of magnitude more powerful than an emergency transmitter, but stars are also many orders of magnitude further away than it would be possible for a search platform to be from an emergency transmitter. Also, the human eye is not an especially sensitive receiver. So an emergency transmitter may be detectable by a search platform from many hundres of kilometers away. In fact, when the SARSAT/COSPAS system was listening to 121.5 MHz, the satellites were able to detect the signals at ranges of over 1000 km. How then is it possible to locate them using the Aural Null search patterns in a reasonable amount of time? That is where line of sight comes in, which we will tackle next.

Line of Sight

If path loss is the most striking factor resulting in the observed reception pattern, perhaps the most useful, at least for searchers without special equipment, is line of sight. Very High Frequency or VHF signals (those in the range of 30 MHz to 300 MHz), and above travel on line of sight paths. They don't go around corners. So they won't travel around the corners of buildings (though they may go through depending on the material in the building structure), over or around hills, or around the curvature of the earth.
This gives rise to the property of a radio horizon that is analogous to the visual horizon that we are all familiar with. Just as the visual horizon is further away the higher the observer is, the radio horizon also varies with altitude. The diagram to the left shows an ELT on the surface of the earth. Airplane A is able to hear the signal because there is a line-of-sight path from the ELT to that airplane. As the airplane flies away, maintaining the same altitude, by the time it gets to position B the signal won't be detectable because there is not a line-of-sight path. If the airplane then climbs up to position C it will again be able to detect the signal because there is a line-of-sight path. Notice that even though airplane C is further away, and at a higher altitude than airplane A they are both at the same anlge above the horizon when viewed from the ELT.

Again, Wikipedia has an article describing line of sight propigation and the radio horizon that is worth reading. If we apply the radio horizon formula from Wikipedia to our reception pattern we get the graph on the right. This shows the reception pattern (signal strength taking path loss and radio horizon into account) of a monopole on the surface received by an airplane at 600m giving a radio horizon at 87.5 km. The graph on the left is derived from all the same parameters except the airplane altitude is 300m giving a radio horizon of 61.8 km. This is why search crews are often encouraged to perform the aural null search patterns at the lowest safe altitude at which the transmitter can be received.


Unfortunately it is not that simple. Most radio horizon distance equations are for formal radio links, and asume that reasonable steps have been taken to provide the best signal path possible. The antennas are normally located on high terrain, placed on towers or building tops, or otherwise away from obstructions that may interfere with signal propagation. An ELT at a crash site wont benefit from any such intentional action unless a survivor is able to remove the transmitter from the airplane and take it to a better transmission location. An ELT may be surrounded by forrest. Trees and foliage can significantly reduce the strength of radio signals that travel through them. As the search aircraft flies at greater distances from the transmitter, the angle of elevation will approach zero degrees. This, in turn, means that the signal will have to travel further through any trees, or other obstructions. So the radio horizon for an ELT may, depending on terrain, may be closer than the formula used above would indicate.

All these factors have been taken into account when designing the Aural Null Simulator Java applet. So you may go to that site and experiment.

You may also be wondering why a search crew has to fly all over the reception area of the signal looking for the edges. Cleary the signal gets continuously stronger as the search platform approches the ELT. Why can't search crews use a simple hill-climbing algorithm (an electronic version of you're cold, you're hot), or like a blind person could find the top of a hill by always walking up slope until all directions lead down hill? Well, on one hand the crew may not have equipment that tells them how strong the signal is. Standard aviation radios installed in light airplanes don't provide that information. The volume of the aural signal the crew hears is controlled by the radio Automatic Gain Control circuit, so it is not helpful. On the other hand, even if the crew has access to equipment that does provide signal strength, there are other factors that make this strategy unusable. We will discuss some of those in the next section.

Fading

Fading is a term used in radio communications to describe the condition when the signal varies in strength over a short period of time, or within a short distance. There are a number of causes of fading, we will start with Multi-path Induced Fading caused by reflections. If you are old enough to remember watching analog TV using rabbit ears, or an antenna on a tower you will be familiar with a type of multi-path that used to plague TV viewers. A television signal would be broadcast from the TV station and, since TV is broadcast on VHF and UHF, would travel line-of-sight to a viewer's antenna. Some times the signal would also bounce off a mountain, hill or large building and then be picked up by the viewer's antenna as well as following the direct path from the station. The bouncing paths are always longer than the direct path. Even though radio waves travel at the speed of light, reflected paths would not have to be very much longer before there was a detectable delay between them and the direct path signal. The horizontal frame rate of an NTSC television signal is 15.75 kHz which only takes 63 microseconds to draw. So the bounce path would only have to be around 2 km longer than the direct path to form an easily seen 'ghost' image.

The type of multi-path that affects search crews operates on an even finer scale. Radio signals behave like waves. This means that two radio signals of the same (or nearly the same) frequency received by the same receiver will interfere with each other producing a new signal which is the result of combining the two. The resulting signal will depend on the relative strengths of each signal, and the relative phases (where the peaks and troughs of the waves are with respect to the other signal). If the signals are in phase with each other, the peaks are lined up, the resulting signal strength will be the sum of the two original signal strengths. If the signals are out of phase, the peaks of one signal are lined up the the troughs of the other, the resulting signal strength will be the difference between the two original signal strengths. The phase of the signal at the receiver for each path is determined by the remainder of dividing the path length by the wave length of the signal. The ratio of the remainder of the two paths will give the relative phase: 0 when the signals are in phase; 0.5 when out of phase. The wave length for the frequency of an ELT is about 2.5 m, so it doesn't take much of a difference in one of the paths to change the relative phase of the two signals from in phase to out of phase and back. So, if there are any reflections present, as the search platform moves about the reception area, or changes altitude the strength of the signal received will vary between the sum of the direct and reflected signals to the difference between the direct and reflected signals. If the reflected signal is nearly as strong as the direct signal, then the crew will observe the signal fading completely away to nothing, then back to twice as strong as the direct signal alone. When the signals are in phase the addition is called constructive interference. When the signals are out of phase it is called destructive interference. This situation is further complicated by the fact that while there can be only one direct signal, there can be a very large number of reflected signals with different phases and strenghts all interfering with the each other and the direct signal. Our hypothetical blind person could walt to the top of a hill, but being unable to see a higher hill accros a small valley, not know that they haven't found the summit yet. Search crews do well to be careful not to use techniques that will lead them astray from the ELT location; surviors could be depending on them for rescue.

Wave Interaction

Thanks to Richard Glenn and Konstantin Lukin and the Earth Science Education Resource Center we can use the Java applet included here to experiment with wave interaction and see constructive and destructive interference.

Reflection Sources

In order to have a reflection there must be something in the environment capbable of reflecting the radio waves. We are all familiar with light reflecting from mirrors, but light will also reflect off of less perfect surfaces as well. Freshly waxed cars, polished metal or stone, glass or ceramic surfaces will all reflect light quite well even if they don't produce a perfect image like a bathroom mirror. Smooth in this instance is relative to the wavelength of the light incident on the surface, but at low angles of incidence light will even reflect off comparatively rough surfaces. Radio waves behave in a similar fashion, but we will have to stretch the concept of smoothness. The wavelength of visible ligth is about 500 nm. The wavelength of the ELT signal is 5 million time longer. So we can expect the ELT signal to reflect off something 5 million times rougher than would reflect visible light, but it would have to be about 5 million times larger than the smallest surfact that would reflect visible ligth. As a rule of thumb we would need an area larger than 0.6 times the area of the projection of the first Fresnel zone with an RMS variance from flat less than one quater of the signal wavelength. These terms may be new to you, so let's take them one at a time.





Fresnel Zones

Fresnel zones are primarily used to determine how much an obstical between the transmitter and receiver will obstruct the radio signal. It is a three dimensional shape that encloses the transmit and receive antennas, as well as the propagation path of the signal. The radius of a Fresnel zone at any point along the path depends on the signal wavelength, the distance of the point from the transmit antenna and from the receive antenna. For an ELT signal where the reflection area is located about 500 m from the ELT and 25 km from the search aircraft the radius of the first Fresnel zone is about 30 m. For a very rough analogy think of a Frenel zone like a search light beam originating from the transmitter and spreading out as it travels. If the beam strikes a surface at 90° it will project a circular pattern of light, in this case with a radius of 30 m. If the beam strikes a surface at a lower angle it will project a nearly elliptical pattern of light. The ellipse will have a semi-minor axis equal to the zone radius and a semi-major axis equal to the zone radius divided by the sine of the angle. For low angles of incidence the projection will be long and narrow growing shorter as the angle of incidence gets larger. You must remember though, it is the area of the projection that is important, not necessarlity the shape of the projection. The table below gives some values for different angles of incidence at 500 m, 100 m and 10m from the ELT. So the areas can be relatively large in normal experience, but small in the context of the reception area which could be 300,000 hectares, and certainly not uncommon. Small lakes and ponds, marshes bogs or flats, out cropings or bluffs could all provide adequate area to support a reflection.

 Angle of incidence
 Area (hectares)
(500 m)

 Area (hectares)
(100 m)
 Area (sq m)
(10 m)
 14  3.6  3500
   6  1.8  1700
   2.4  0.6  660
 10°  1.2  0.4  340

Of course, smaller areas that are made from highly reflective materials (for the frequency of the signal) or have very smooth surfaces may also support reflections. For example, a smooth metal surface, like part of an airplane in the debris field, of only a few wavelengths in diameter could produce a strong reflection.

You may have experienced multi-path induced fading when listening to FM radio while driving in a car in a city or town. Building give rise to a great number of reflections that produce a very complex radio environment. It is quite common to stop at an intersection and find that the radio signal that was strong and clear only a moment ago is now weak and noisy. If this is fading due to multi-path movng the car about 1 m ahead (only if it is safe to do so of course) may move the antenna out of the destructive interference into constructive interference. This will restore the signal to the former level.

Root Mean Square

The Root Mean Square, or RMS is a statistical measure of the magnitude of a varying quatity. It is useful when the values are both positive and negative.

The Effect of Reflections

So, our reception pattern is no longer simple and regular. With only a few reflections (nothing like the number that could exist in a real situation) the reception area begins to look like the image on the left. This graph is shown in a top down view because the perspective view is too busy to really see the effects. It is easy to see how a search crew depending on a signal strength search and expecting a single peak signal location could be fooled by the peak near the 10 nm Easting and 10 nm Northing. Just as our hypothetical blind person walking in from the East along the 10 nm Northing could find the hill top at that location, but not know there are a succession of ever higher peaks to the Southwest, nor have any idea which way to turn to seek out any potentially higher peaks. Our intrepid blind hiker, or the search crew, have no alternative than to systematically cover the entire reception area cataloguing all the peaks, or use a more creative solution. This is where the Aural Null search technique works so well. Rather than scouring the whole area, they need only find three points on the radio horizon and do some geometry. Here at SARMobile.ca we are working on software to take care of the geometry.


Comments