I use aircraft scatter at VHF/UHF and microwaves to make long-distance
non LOS contacts. It is a bit of hit and miss and I'm trying to make a
more scientific approach to the problem.

Programs like Airscout tell when an aircraft lies in a zone on the
direct path between myself and the distant station in real time. So
that removes some of the guesswork.

The system is a form of bistatic radar .Equations exist to estimate the
received signal level and hence it is possible to calculate the SNR from
knowledge of both stations equipment e.g. antenna gain, power and
effective noise figure. I'm including ground gain in antenna gain.

Now the unknown parameter is the reflection characteristics of the
aircraft. Some generic info exists on the monostatic cross-section e.g.
A 747 is about 100m^2

The bistatic equivalent does not, because it is highly dependent on
relative angles.

The case I'm interested in is however a small subset i.e the forward
scattering x-section over relatively well defined angles.The unknown is
the orientation of the aircraft. However this can be derived from
Airscout ,since the direction of travel is known.

To try to get the range of figures, I've been looking at 4NEC2, which
for a starting point has 747 wire model. To simulate the bistatic
radar case, it is possible to replace the voltage excitation , by a
plane wave at a specific angle.

When the model is run, you get a set of complex polar plots with
strongly peaked lobes in the forward direction. 4NEC2 gives the relative
gain dBi as a function of angle.

What I have done is to turn the gain figures into an equivalent bistatic
radar cross-section by multiplying them by lamda^2/4*pi , since this is
the RCS of an isotropic radiator.

What I want to know is if this approach is valid ? Also has anyone else
tried something similar ?

I'm not looking for precision, but only an indication that a particular
QSO is possible or a no-hoper.

Brian GM4DIJ





--
Brian Howie

On 27/09/2018 16:46, brian wrote:
I use aircraft scatter at VHF/UHF and microwaves to make long-distance
non LOS contacts. It is a bit of hit and miss and I'm trying to make a
more scientific approach to the problem.

Programs like Airscout tell when an aircraft lies in a zone on the
direct path between myself and the distant station in real time. So
that removes some of the guesswork.

The system is a form of bistatic radar .Equations exist to estimate the
received signal level and hence it is possible to calculate the SNR from
knowledge of both stations equipment e.g. antenna gain, power and
effective noise figure. I'm including ground gain in antenna gain.

Now the unknown parameter is the reflection characteristics of the
aircraft. Some generic info exists on the monostatic cross-section e.g.
A 747 is about 100m^2

The bistatic equivalent does not, because it is highly dependent on
relative angles.

The case I'm interested in is however a small subset i.e the forward
scattering x-section over relatively well defined angles.The unknown is
the orientation of the aircraft. However this can be derived from
Airscout ,since the direction of travel is known.

To try to get the range of figures, I've been looking at 4NEC2, which
for a starting point has 747 wire model. To simulate the bistatic
radar case, it is possible to replace the voltage excitation , by a
plane wave at a specific angle.

When the model is run, you get a set of complex polar plots with
strongly peaked lobes in the forward direction. 4NEC2 gives the relative
gain dBi as a function of angle.

What I have done is to turn the gain figures into an equivalent bistatic
radar cross-section by multiplying them by lamda^2/4*pi , since this is
the RCS of an isotropic radiator.

What I want to know is if this approach is valid ? Also has anyone else
tried something similar ?

I'm not looking for precision, but only an indication that a particular
QSO is possible or a no-hoper.

A couple of points...

The azimuthal RCS of an aircraft results in what's sometimes known as a
'fuzzball' which can have high peaks and deep nulls. I suspect the
stated RCS is some average of this.

If you're targeting large transport aircraft having two or four large
turbofans, then perhaps your best best is to choose those aircraft that
are approaching along a vector that results in roughly equal angles
between your station, the aircraft itself, and the distant station. and
bounce your signal from the fans, which in themselves will have high
RCS. Masking the fans from radar is a critical requirement for stealth
or low RCS aircraft.You'll probably be able to recover the Doppler from
the fan rotation, if you wanted to.

It is said that during WWII Wurzburg radar operators could hear the
Doppler caused by the rotation of the target aircraft's propellers.


--
Spike

"Nearly all men can stand adversity,
but if you want to test a man's character,
give him an internet group to manage"

In message , Spike
writes
On 27/09/2018 16:46, brian wrote:
I use aircraft scatter at VHF/UHF and microwaves to make long-distance
non LOS contacts. It is a bit of hit and miss and I'm trying to make a
more scientific approach to the problem.

Programs like Airscout tell when an aircraft lies in a zone on the
direct path between myself and the distant station in real time. So
that removes some of the guesswork.

The system is a form of bistatic radar .Equations exist to estimate the
received signal level and hence it is possible to calculate the SNR from
knowledge of both stations equipment e.g. antenna gain, power and
effective noise figure. I'm including ground gain in antenna gain.

Now the unknown parameter is the reflection characteristics of the
aircraft. Some generic info exists on the monostatic cross-section e.g.
A 747 is about 100m^2

The bistatic equivalent does not, because it is highly dependent on
relative angles.

The case I'm interested in is however a small subset i.e the forward
scattering x-section over relatively well defined angles.The unknown is
the orientation of the aircraft. However this can be derived from
Airscout ,since the direction of travel is known.

To try to get the range of figures, I've been looking at 4NEC2, which
for a starting point has 747 wire model. To simulate the bistatic
radar case, it is possible to replace the voltage excitation , by a
plane wave at a specific angle.

When the model is run, you get a set of complex polar plots with
strongly peaked lobes in the forward direction. 4NEC2 gives the relative
gain dBi as a function of angle.

What I have done is to turn the gain figures into an equivalent bistatic
radar cross-section by multiplying them by lamda^2/4*pi , since this is
the RCS of an isotropic radiator.

What I want to know is if this approach is valid ? Also has anyone else
tried something similar ?

I'm not looking for precision, but only an indication that a particular
QSO is possible or a no-hoper.

A couple of points...

The azimuthal RCS of an aircraft results in what's sometimes known as a
'fuzzball' which can have high peaks and deep nulls. I suspect the
stated RCS is some average of this.

If you're targeting large transport aircraft having two or four large
turbofans, then perhaps your best best is to choose those aircraft that
are approaching along a vector that results in roughly equal angles
between your station, the aircraft itself, and the distant station. and
bounce your signal from the fans, which in themselves will have high
RCS. Masking the fans from radar is a critical requirement for stealth
or low RCS aircraft.You'll probably be able to recover the Doppler from
the fan rotation, if you wanted to.

It is said that during WWII Wurzburg radar operators could hear the
Doppler caused by the rotation of the target aircraft's propellers.



The 4NEC2 model of the 747 x-section certainly is a "fuzzball" , The
high peaks and deep nulls from glint certainly show up when listening.
The trick is to wait for the "lucky" peaks.

Http://www.b-howie.co.uk//brian's%20home%20pages_files/amateur_radio/747_
rcs.jpg

But you can see the strongly peaked forward lobe which is what I'm
interested in.

Airscout identifies the type of aircraft, the bigger ones like Airbuses
and 777s are particularly good. I can guarantee to work a station at
380km in the North of Scotland from a location in SW Scotland on 70cm
pretty well guaranteed when these are about.

This guy here has developed some models based on projected areas

http://www.nitehawk.com/w3sz/vk7mo_2000.pdf

I'm looking to see how the 4NEC2 wire grid models compare.

Brian
--
Brian Howie

In message , brian
writes
When the model is run, you get a set of complex polar plots with
strongly peaked lobes in the forward direction. 4NEC2 gives the
relative gain dBi as a function of angle.

What I have done is to turn the gain figures into an equivalent
bistatic radar cross-section by multiplying them by lamda^2/4*pi ,
since this is the RCS of an isotropic radiator.

Reading further it looks like NEC2 gives crossection gain as a/lambda^2
in dB relative to 1m^2 . I'm underestimating by 4.pi

Brian
--
Brian Howie

On 29/09/2018 16:00, brian wrote:
writes

The azimuthal RCS of an aircraft results in what's sometimes known as a
'fuzzball' which can have high peaks and deep nulls. I suspect the
stated RCS is some average of this.

If you're targeting large transport aircraft having two or four large
turbofans, then perhaps your best best is to choose those aircraft that
are approaching along a vector that results in roughly equal angles
between your station, the aircraft itself, and the distant station. and
bounce your signal from the fans, which in themselves will have high
RCS. Masking the fans from radar is a critical requirement for stealth
or low RCS aircraft.You'll probably be able to recover the Doppler from
the fan rotation, if you wanted to.

It's a long time since I last referred to Skolnik - my interests at the
time centred around travelling and creeping waves, and backscatter
rather than forward scatter, so I wrote the above in those terms. This
is of course a completely different kettle of fish! I had no idea such
forward-scatter enhancements could be useful, although aircraft
reflections have been used for many decades by Radio Amateurs. ISTR some
Amateurs living within ground range of airport/field long-range radar
using the pulses to time a receiver display to show the echo from
aircraft many miles away.

Airscout identifies the type of aircraft, the bigger ones like Airbuses
and 777s are particularly good. I can guarantee to work a station at
380km in the North of Scotland from a location in SW Scotland on 70cm
pretty well guaranteed when these are about.

Is there a station or two located at some distance to one side from the
line between you and the distant station you mention that could check to
see what signals could be recovered? You might have more coverage than
you think!

This guy here has developed some models based on projected areas

http://www.nitehawk.com/w3sz/vk7mo_2000.pdf

That's a very interesting paper, which I'm enjoying reading.

I'm looking to see how the 4NEC2 wire grid models compare.

Keep us posted on this interesting topic!



--
Spike

"Nearly all men can stand adversity,
but if you want to test a man's character,
give him an internet group to manage"