Thanks to the Dutch Activity Contest I am convinced that aircraft scatter
is the best way to make nice DX contacts on 23 cm.
 

I have made in the last few years many QSO's on 23 cm with Ray GM4CXM using planned aircraft scatter using aircraft flying through
our common scatter area. Almost all our tests were successful. Mind you, we're talking about
a distance of 729 km, which for many 2m station is not possible or a very beautiful DX contact.

In the past two years I have approximately made fifteen
times a QSO with GM4CXM using aircraft scatter. It occasionally failed,
 at that time the aircraft flew exactly through the scatter area but there was also good tropo propagation.
Possibly that a temperature inversion weakened our signal reaching the aircraft or completely
not arriving at the aircraft.
During the other contests I succeed contacts at 23cm often at greater distances.
Usually I make distances up to about 850 km with stations from OK, like OK5Z from
JN89ak about 854 km or with OK1KUO JO80ff about 849 km.
 

During the VERON Dag voor de radio amateur I had a live demo in the VHF-cie stand to demonstrate aircraft scatter.
I was receiving the signal of the beacon DB0AJA from JN59as, 457 km away, I could show that the signal became much stronger if a plane flies through the
scatter area. Since the beacon is not that far away Troposcatter also played a role in the reception. During the demo we many
times saw an increase of the signal with more than 20 dB.
Even more striking is when the distance is larger. Without a plane you really copy nothing from a DX station, and if there is a plane in the common
scatter area you will then hear the signal of your counterpart rise and usually a minute or two at usable strength to make a QSO.
 

Equipment and method
Requirements to be able to work with aircraft scatter.
Firstly, on both sides a good working radio system and an experienced operator
with the skills to complete a QSO within 30 to 60 seconds.
Depending on the flight path of the airplane a QSO sometimes must be realized in only 30 seconds.
If the flight path is better than there is more time, up to more than four minutes.
 

Planning of a QSO.
Using your own position and the position of the dx station the scatter point can be calculated.
This can be done using a tool built by SM7LCB, available on the Internet.
 In this tool, the own locator and the locator of the DX station are entered, and the
opening angle of the two different antennas.
In the figure, you see the common scatter point between me and GM4CXM.
|

Now it is important to know where the aircraft fly, to determine at what moment the QSO is possible.
There are a variety of on-line radars available that shows where aircraft are flying. Beware though of
arbitrarily choosing a  on-line radar website, because most websites are delayed, for safety and sometimes are fifteen minutes behind.
That kind of radar is obviously of no use, because using a delayed radar you can only see that you could have been able to make a QSO fifteen minutes ago.
A radar without delay and which I use myself is: http://www.flightradar24.com/

With the use of the radar one can see which aircraft is approaching the scatter area and when
a test can begin.
Using a microwave chat like ON4KST is actually essential, because you want to know if
your DX station is QRV and at the right time and frequency transmitting or listening.
You agree with each other who at what time and at what frequency will transmit and who is going to listen.
With GM4CXM most of the times he gives CQ in CW and I respond as soon as I can copy his signals.  This works fine for the tests we have done up to now.
The transmitting station gives short periods CQ or calls the opposite station for short periods and switches to RX to listen for
the other station to return. At the moment that you copy the calling station, you answer when he goes on RX,  and
you can then quickly make the QSO through reports and locator exchange and finally confirm everything with Rogers and 73's.
The fastest QSO what I made in this way is within twenty seconds.
I sometimes copy GM4CXM four minutes and can still hear him calling CQ.
In the beginning I was surprised that nobody else in the Netherlands tried a QSO after me,
but if you look in the scatter maps the area were the reflection arrives is not that big.
A station more than 30 km away from my home probably cannot hear anything
and the reflection from the plane which I use to make a QSO will not work for the other station.
 Each amateur must make a calculation, for its own situation, of the scatter area.
For stations that are within 30 km it should be possible using the same scatter area to make a QSO a connection to the DX station.
However both stations should be working fast, because the reflection often is short.
On my website I have a audio recording of one of the contacts between me and GM4CXM, who can be heard for more than 4 minutes
and on the end his SSB signal comes in calling CQ contest.
During the DAC of June 2012 I made a QSO scatter on the KLM49J from Amsterdam to Glasgow,
flying through the scatter point at more than 11km altitude.
The picture is made just after the QSO with GM4CXM.
After my first article I started to better understand aircraft scatter and try to make QSO's with more stations.
One problem that I had trouble was that the scatter area was not simultaneously on the radar map.
After some searching I found a piece of software that both maps placed over each other could be visible
 by making the upper map transparent. With this tool it is much more clear how aircraft in fly the vicinity of a scatter area.
The program is called Acer GridVista and be downloaded from here
After installing it needs some time to figure out how things work, but that is reasonably fast to understand.
Now put the map with the calculated scatter area below the map of the aircraft radar and be sure that both maps have the same scale.
If all that matches you can click the icon GridVista (almost top right of your screen) click on the button transparent.
Then you have to align the two maps exactly, by using the mouse on the radar screen and move the map with the left button of the mouse pressed, gently
moving the map until it fits exactly above the map with the scatter area.
 To align you can use coastlines, or sometimes place names, which often are placed on same way in the two maps.
It is a bit trying, but the result is a picture in which you see the scatter area and the positions of the aircraft.
In this figure you see a preview.

After that it is to wait until the moment that an airplane is approaching the calculated scatter area,
and then it's time to quickly agree with the other station to start testing
After I projected both maps over each other, I was interested why for some stations I needed to plan a QSO and why for some other stations this was not or hardly necessary.
I had for several years the experience that a QSO with OK2KKW actually always works out without planning
- sometimes stronger then a bit less - .
The explanation I found by looking on the radar maps was simple:
The scatter area needed for that QSO is located in an busy airway most of the time there were usually two or more aircraft flying in the scatter area.
For some other stations the scatter area is way outside an airway and then you need to guard
 and plan a QSO and keep a look on aircrafts approaching the scatter area.
For example the scatter area for my QSO with GM4CXM has about three flights in the evening when the DAC contest is held.
If you will miss these planes the chance of making a QSO is lost.
Once accustomed to this practice, how airplanes fly and which flights are suitable, it is not very difficult.
So now I know that very often the KL49J crosses very well the scatter area.
I monitor the departure of the KL49J from Schiphol and then I know that about 30 minutes later,
the flight approaches the scatter area.
Ten minutes before the flight enters the scatter area I chat with GM4CXM using the ON4KST chat box to warn the him to be alert for the upcoming possibility.
 Then it's most of the times pretty simple to make the QSO, in which more than 90% of the cases we tried actually succeed.
With SP1JNY from JO73gl a distance of 686 km, I have made 3 times a planned aircraft scatter test so far which all were successful.
The first time I experienced a QSO with SP1JNY as very special, because I still remember that my first QSO with SP is not that long ago needing good tropo propagation.
In using aircraft scatter now it is a matter of proper planning and good stations on both sides and to work methodically,
and then we can always make a QSO
 In Figure 4 shows how the plane flew, and which route it has followed some time after I the connection was made.

 The diameter of the scatter area is user set.
In this picture you can see the plane actually flying just on the edge of the area.
My impression is that as the distance between the two stations is larger also the diameter of the scatter area becomes larger.
Also I made several QSO's with GM0USI from IO75uv about 722 km, using planned aircraft scatter.
Furthermore, there are many QSO's I made which probably were successful using aircraft scatter but being unaware,
because I was not trying to find if an aircraft was flying in the scatter area.
So the contacts I made with OK5Z from JN89ak about 854 km were made with great certainty using aircraft scatter.
The last few months I have been trying to test with IZ4BER with if we can make an aircraft scatter QSO. 
In the tree times in which we now tested we have not been able to succeed.
Perhaps this distance is just too far.
It is obvious that a large part of the real DX contacts on 23cm is made with help of aircraft scatter.
The signals are usually not very strong, but success rate is very high.
As technical parameters of the equipment needed for these contacts you need to have more than 22 dB antenna gain
 and more than 100 watts output. With a free take off this allows DX contacts over 800 km reasonably reliable.
If the DX station has a very nice QTH, for instance many OK stations are high up in the mountains,
then the maximum distance of the DX will be more, because he can have line of sight contact with a plane over a much larger distance.
An airplane flying on an altitude of more than 10 km is visible over a range of about 400 km away.
If both stations have a free take of to the horizon and be able to see the aircraft on a large distance a DX of more than 800 km is possible.
When one is on a high mountain then he could see the plane even a 100 km further and the we can make the 900 km range.
Distances of 1000 km are not realized using aircraft scatter.
 

Aircraft scatter and Doppler shift
During my demonstration at the VERON DvdA several questions were asked about
Doppler shift. When we calculate the scatter area we always see that its located exactly halfway the distance.
If an RF signal is blasted on the aircraft the reflected signal will have a Doppler shift in accordance with the speed of the aircraft.
The RX station than receives the shifted signal but this signal is shifted back just on the same amount as the signal shift on the aircraft.
During scattering we are dealing with two Doppler shifts, from the transmitter to the aircraft and from the aircraft to the receiver.
When the aircrafts flies midway, the Doppler shift of the transmitter to the aircraft is just as large as that of the
plane to the receiver. Because the two shifts are opposite the net Doppler shift is zero.
If the aircraft leaves the center of the path the two Doppler shifts start to differ, and
the frequency starts to drift away.
But in addition, the signal strength is usually much less than that the middle of the scatter area and a contact is no longer possible.
Occasionally you can indeed hear signals that encounter Doppler shift, but usually this is very short and useless.

The height at which the aircraft is flying of interest. If the plane flies low then the signal from your antenna cannot reach the plane
because the earth comes in between. As the airplane is flying higher the bigger distances can be reached.
On the radar internet site you can select a height filter.
For distances up to 500 km I usually set the height filter so that I don't see aircraft flying below 7 km altitude.
For distances above 700 km I set the filter so that I only see aircraft flying at an altitude of more than 10 km.
In the November 2012 DAC I made a test with GM0USI.
We tested using an Boeing 747 flying well in the scatter area at an altitude of 9 km height, but that test failed.
Shortly thereafter a Boeing 737 entered the scatter area at 11 km altitude, and then we were successful in our QSO.
The height of the aircraft is certainly important, and one could say that for distances above 700 km the
airplane must be flying above 10 km altitude.
 

Also on the other bands, both above and below 23cm, its possible to make aircraft scatter contacts.
Even at 10 GHz aircraft scatter contacts are possible, but due to the poorer reflecting
characteristics of the aircraft and the faster occurrence of a significant Doppler shift
the distances are generally less than at 23cm.
Distances up to 600 km at 10 GHz are certainly possible.
 

Calculating aircraft scatter
On the website of SM6FHZ I found an Excel spreadsheet to calculate the signal strengths which is helpful in understanding what to expect.
The station parameters data from the two stations as antenna gain, output power, and so on must be entered.
Based on radar equations the  spreadsheet calculates how strong the signals can be.
I have been playing with this tool and come to the conclusion that this calculation gives a good impression.
In practice I have the feeling that the signals are somewhat stronger than the spreadsheet calculates,
I estimate a difference up to 10 dB
stronger.
A possible explanation could be that the radar equation is based on back scatter, whereas we are scattering forward.
The determination by the spreadsheet of the reflecting surface of the aircraft may be an other source for the difference.
Nevertheless, the spreadsheet gives a good impression what is possible.
The starting point for aircraft scatter contacts must be a well working technical system with decent
antenna gain and as much as possible output power. Also a free take off is very important:
you need to see the aircraft without obstacles between.
 

Procedure using in contests
During a contest with high activity with many possible DX stations you actually want to have an overview of the various scatter areas
to see if aircrafts are approaching interesting scatter area's.
I use the again the above-described radar and the tool from SM7LCB, but now for a much larger area.
On the map to calculate the scatter areas I enter all interesting DX stations with their locator,
and this will then project the scatter areas for these stations.
Then I put the aircraft radar in the same scale above and I make the radar display transparent.
Provide the same scale of the maps and both maps neatly aligned you see all flights and scatter areas.
When an aircraft approaches an scatter area on the map, then
quickly using ON4KST make a sked request to the related DX station and agree on timing and frequency.
Than it is a matter of good operating technique and listening to be able to make a QSO.
The mechanism aircraft scatter is very interesting to use and one can further investigate using this technique.

I myself, was a year or two ago, not very convinced of the possibilities, but since I have been trying and studied further into this,
I am more and more convinced that using aircraft scatter brings us the DX opportunities at the
higher frequencies sometimes even better than on the two meter band.
It is interesting to further investigate and study the size of the scatter area.
As stated above, I think that the area is smaller when the DX distance is smaller.
It is also interesting to examine how the scatter area must be defined for the higher frequencies.
Obviously also nice too study the relationship between the signal strength of the echo
compared to the airplane type. The most experience I have is with the usual Boeing 737, simply because the most aircraft flying are 737's,
but probably signal strengths and perhaps the DX maximum attainable will increase with larger
aircraft such as the Boeing 747 or the modern Dreamliner.
 

It is fortunate that the passengers in the aircraft do not know we use their plane to realize a DX contact.