Doppler Radar

Home ] Up ]


Click HERE for the Amazing Online Ham Radio Flea Market

Click to Buy and Sell Ham Gear Online





6.7m fibreglass poles for sale click ***HERE***

  Search G3CWI's site

Summits on the Air - click here


What's it all about?

Ever wanted your own radar station? Well maybe not, but it is fun to see what is up in the skies around your area. The following plots shows Doppler returns from aircraft. The straight line in the centre is an HF carrier (about 26 MHz) and the curves are Doppler returns from aircraft - look how many there are!

scat1.jpg (62436 bytes)

In the trace above there are 4 aircraft trails (crossing the carrier at 20s, 135s, 180s and 195s). The trails are produced by signals from the carrier transmitter being reflected from the aircraft and then Doppler shifted due to the motion of the aircraft. Read about the Doppler effect here.

So, what can we tell from these trails? The first thing we can tell is the relative speed of the aircraft - that is how fast they are heading towards or away from us. This is done by looking at the maximum Doppler shift. I will explain this with reference to another plot.

overhead pass.jpg (39882 bytes)

This plot was produced late at night (local midnight) when the maximum usable frequency was well below the frequency of the transmitter. As above, the fixed transmitter carrier is the horizontal line in the centre of the plot (at an audio frequency of about 745Hz). The aircraft plot is the heavy curve which crosses the carrier at about 200 seconds. The other curves which cross the carrier are thought to be due to the effect of the gain control in the receiver. The aircraft passed low over my QTH (and woke me up) on its approach to Manchester airport. Looking at the aircraft plot we can see that it appears first at about 150 seconds, the reflected tone at that point is about 715 Hz. The tone then rises, crosses the carrier and disappears at around 250 seconds. At the point when it disappears, the tone is about 767 Hz.

Working out the speed

In the trace above, the aircraft speed can easily be calculated. As it travels away from me (at 235 seconds), the doppler shift is about 22 Hz. The observation is at 26.6 MHz. At this frequency a half wavelength is 5.77 metres. The speed is the doppler shift multiplied by the half wavelength (i.e. 5.77 x 22 = 127 m/s).

For people in the South Manchester area using the carrier in 26615kHz, here is a conversion table. Note that this table only applies to this specific carrier frequency.

Hz km/hr mph
0 0 0
5 101 63
10 203 126
15 304 189
20 406 252
25 507 315
30 609 378
35 710 441
40 812 504
45 913 567
50 1014 630
55 1116 693
60 1217 756
65 1319 820
70 1420 883
75 1522 946
80 1623 1009
85 1725 1072
90 1826 1135
95 1927 1198
100 2029 1261

But that is not all that can be deduced as Peter G3PLX explains:

The Doppler curve shows two sections and they are symmetrical. This says that the aircraft is flying along the line between one station and the other (you cannot resolve which station is the transmitter and which is the receiver). When the aircraft is flying between the two, the Doppler shift is essentially zero. So we can measure the time interval between the aircraft flying over one station and flying over the other,. It's about 38 secs, measured between the two points of maximum slope. This puts the transmitter 5km away from you. If you know which direction the aircraft was flying and whether it flew over you first or second, you can then get the direction to the transmitter and attempt to locate it.

All this is a bit approx. but you can begin to see how you could, if you knew the tracks of aircraft, locate unknown transmitters (or vice versa) from the Doppler trace. I did spend some time on this at one point, writing software to place a computed Doppler trace over the top of an observed one while adjusting the position of the aircraft track or the position of the transmitter.

Other effects...

double cross 29082002 2240L.jpg (41011 bytes)

This plot shows a "double cross" and I think that it shows an aircraft circling prior to its approach into Manchester.

crossover 29082002 2230L.jpg (51661 bytes)

The kink in the trace above is probably the result of an aircraft making a turn.

Note that the mirror images here are caused by the AGC in my radio it is not really there!

Do Try This At Home!

There is no rocket science here. Anyone with a stable receiver, by which I mean any modern sythesised ham radio receiver, can produce plots like this. First download the spectrogram software here. Install it on your PC. Now using the same connections that you use for PSK31 etc, connect the PC to your radio so that audio from the radio is going to the line input of the PC. Now you are set.

Experiment a little by tuning in a few carriers on HF and observing them on the display.

wpe8.jpg (32004 bytes)

Above are the settings that I use for my spectrograms. This will give you the right resolution to see aircraft scatter.

Now the tricky part. To get aircraft returns you need to find a fairly local transmitter to observe. By local I mean within about 100km. To get good Doppler effects a higher frequency will be best. To find something useful, I suggest that in the first instance you listen above 16 MHz late in the evening. This will mean that the bands are closed for most forms of ionospheric propagation, thus any stable carriers will be local transmitters. If you find nothing then you will need to see if there are any shortwave broadcast sites near to your QTH and listen for them. Here in Macclesfield, UK I use a carrier on 26615 kHz (not on all of the time). But I can also get returns from the BBC World Service transmitter near Penrith on 15485 kHz (see below).

BBC.jpg (98681 bytes)

If you do find a local transmitter, the chances are high that you will see aircraft returns within a few seconds - you have then started exploring this interesting phenomena.

What else can be done?

With a highly stable receiver all sorts of interesting things can be observed. Below is an example sent to me by Peter Martinez G3PLX together with his explanation.

ST8BRSW2.bmp (16950 bytes)

I attach ST8BRSW2.BMP which shows such a scattergram taken on 8MHz last thing at night. The vertical scale is height or time delay up to 1930km or 13mS and the horizontal scale is Doppler shift +/-4.6Hz. Bottom centre is a black dot which is the groundwave at zero height and zero Doppler. Above and to the left is the F layer at about 4-5mS and -3Hz. Above that and further to the left is the backscatter mixed-in with some two-hop F. Across the top of the trace in a broad band stretching from +4.6Hz and beyond to -4.6Hz and beyond (behind the backscatter) at 11mS, is some field-aligned F-layer backscatter, better known to us as Aurora. Finally, just to the right of centre at the bottom at +0.28Hz, is groundwave backscatter, a sharp dot at almost zero range (but not quite zero). This is scatter from sea waves in Morecambe bay and the Irish Sea. It's a sharp dot because although the waves are essentially random in direction, wavelength, and speed, there is a causal relationship between wavelength and speed and that gives a Doppler-shifted diffraction effect. This dot is at +0.28Hz because the waves are essentially blowing towards me. When the wind is NE this dot is at -0.28Hz.

I have done lots of these, including sea scatter via Sporadic E, etc.

Peter is the expert on this subject and it is thanks to his articles over the years in the RSGB magazine Radcom, that I be came aware of the techniques. Producing scattergrams like the one above is not a simple task however. It requires a receiver that is much more stable than thge average ham radio setup!

So, what is next? I am keen to explore gravity waves in the ionosphere and hope to start some test transmissions in the next few months to allow this to be done.





Home ]