em_data.html


Passive/Active Radar: Electromagnetic Scattering Data (Restricted Access)














Main

Links

Papers


Research

Project Only

UIUC Only

FISC CAD models

  • t38a_aligned.facet is a
    model of a T-38A training aircraft adapted from a file purchased from
    Viewpoint Digital

  • falcon100_aligned.facet is a
    model of a Dassault Aviation Falcon 100 adapted from a file purchased from
    Viewpoint Digital

  • falcon_20.facet is an approximate model of the
    Dassault Aviation Falcon 20. To make this, we simply scaled the
    Falcon 100 model to be the same dimensions as the real Falcon 20 (which is
    quite similar to the Falcon 100, but larger.)

  • aaron_sphere.facet is a sphere with a
    diameter of 15 meters

  • standing_square.facet is a flat square
    plate square, 15 meters along each side standing vertically, lined up along
    the x-axis.

  • lying_square.facet is a flat square
    plate, 15 meters along each side lying horizontally (looks completely thin
    when viewed at 0 degrees elevation)

Lower VHF FISC data

The following MATLAB files each contain four variables (containing
complex far-field values): fisc_hh, fisc_hv,
fisc_vh, and fisc_vv, each corresponding to a different polarization.
Each variable is a 5 by 360 by 37 array (expect for aaron_sphere_vhf.mat,
which is just a 5 by 360 array). The first dimension corresponds to the
five frequencies associated with NTSC TV channels 2 through 6
(55.25 to 79.25 MHz, with a 6 MHz spacing.) The second dimension corresponds
to the observation azimuth angle, ranging from 0 to 359 in 1 degree
increments. The final dimension corresponds to the incident azimuth angle,
ranging from 0 to 180 in 5 degree increments. For the aircraft, 0 degree
is looking at the nose, 90 degree at the side, and 180 degree at the tail.
The elevation angle (both observed and incident) is zero degrees.

The sphere file only needs one incident angle due to symmetry.

The above files were created using a special version of FISC that can
loop through both incident and observed angles in one run. The input files
were

After reading the *.field file output by FISC using read_fisc_field_file
(given below), fisc_vv, etc. were created by entering commands like this
into MATLAB:



num_freqs = 5; num_obs = 360; num_inc = 37;

fisc_hh = reshape(hh,num_freqs,num_obs,num_inc);
fisc_vv = reshape(vv,num_freqs,num_obs,num_inc);
fisc_vh = reshape(vh,num_freqs,num_obs,num_inc);
fisc_hv = reshape(hv,num_freqs,num_obs,num_inc);

save vfy_vhf fisc_hh fisc_hv fisc_vh fisc_vv

30 Degree Elevation, Lower VHF FISC Data

These are FISC data files, for channels 2-6,
in the same MATLAB format described above,
except at 30 degrees elevation (either plus or minus). They were created
by Michael Brandfass.

Higher VHF FISC data

Like above, except each variable is a 7 by 360 by 91 array.
The first dimension corresponds to the
seven frequencies associated with NTSC TV channels 7 through 13
(175.25 to 211.25 MHz, with a 6 MHz spacing.) The second dimension corresponds
to the observation azimuth angle, ranging from 0 to 359 in 1 degree
increments. The final dimension corresponds to the incident azimuth angle,
ranging from 0 to 180 in 2 degree increments.
The elevation angle (both observed and incident) is zero degrees.

The input files were

The MATLAB files were made from the field file as above, except here

num_freqs = 7; num_obs = 360; num_inc = 91;

FM radio FISC data

Like above, except each variable is a 4 by 360 by 61 array.
The first dimension corresponds to radio stations at frequencies
88, 94, 100, and 106 MHz. The second dimension corresponds
to the observation azimuth angle, ranging from 0 to 359 in 1 degree
increments. The final dimension corresponds to the incident azimuth angle,
ranging from 0 to 180 in 3 degree increments.
The elevation angle (both observed and incident) is zero degrees.

The input files were

The MATLAB files were made from the field file as above, except here

num_freqs = 4; num_obs = 360; num_inc = 61;

2-D Helmholtz Data

Moving away from FISC for a moment, the data sets found here are from
a 2-D scenario where the scatterer is assumed to be an infinitely long
cylinder with a cross-section profile given by some 2-D shape. If the
incident wave is polarized so that the electric or magnetic field runs
along the axis of the scatterer, Maxwell’s equations reduce to the
much simpler scalar Helmholtz equation.

Each MATLAB file here contains four variables: tm_100, tm_200, te_100,
and te_200. These correspond to the transverse magnetic (TM) and
transverse electromagnetic (TE) cases, and with incident frequencies
of 100 and 200 MHz. The data was simulated using Karl Warnick’s
fantastic 2-D scattering code. Each variable is a 120 by 120 array
containing the complex far field for different incident and observed
angles, with an angular spacing of 3 degress (yielding full aperture
data). The different files correspond to different objects, which
are the top-down profiles of a B-2
artifically scaled to have a wingspan of 15.5 meters, an F-15
(19.5 meters long), a VFY-218 (15.5 meters long),
and a YF-23 (20.5 meters long).
The associated *.txt files give the 2-D coordinates of the
vertices of the objects.

MATLAB Code


Last updated 3/5/01. Send comments or questions to
lanterma@ifp.uiuc.edu.