316 RADIO WAVE PROPAGATION EXPERIMENTS 
on short-range targets. The author considered this to 
be a form of receiver saturation. Another group had 
been troubled by the same phenomenon and had at- 
tributed it to receiver saturation in which there was 
blocking of the i-f amplifier during a portion of the 
time. 
RADAR SCATTERING OVER 
CROSS-SECTION AREA? 
It is of great interest to determine the cross-section 
values of aircraft, not only in order to attempt predic- 
tion of ranges on these aircraft, but also to make pos- 
sible the design of radar equipment which will utilize 
these factors a little better. The instantaneous pattern 
of reflection properties of an airplane is very complex. 
It depends upon the frequency, type of aircraft, and 
certain other factors, such as propeller rotation. The 
pattern has an extremely complex lobe structure which 
depends essentially upon the lengths of the plane’s 
structure in terms of wavelength and upon the areas 
of specular reflection, that is, reflection from fairly 
large, flat, mirror-like surfaces found in most aircraft, 
such as the sides, bottoms, or wing surfaces. 
It would be possible to define the instantaneous 
cross-section area as a function of the angle from the 
airplane, but this kind of thing would be purely aca- 
demic, since actually the airplane is moving. In the 
early part of this work an attempt was made to derive 
a cross-section number which would apply to the actual 
radar performance on an airplane in flight. The scat- 
tering cross section may be calculated from the re- 
lation 
_ (4n)8P,R4 
7 P:G72) 
where the quantities are measured in free space. The 
symbols are defined on pp.312-316 . They are all easily 
measurable except P,, the received power. This was 
measured by injecting into the system, with a signal 
generator, an artificial echo which was matched to the 
size of the airplane echo. 
In pratice, o is necessarily a function of time, and 
for lack of a better criterion the following procedure 
was adopted. The signal generator reading was con- 
tinuously matched to the size of the aircraft echo and 
recorded for successive 3-see intervals. The signal 
measured in decibels above receiver noise power was - 
plotted against range. On a logarithmic scale such a 
plot should be a straight line whose variation is 40 db 
for a factor 10 in range. This is actually found, pro- 
vided one draws a line through the average of the 3-sec 
interval points. From moment to moment the fluctua- 
tion is rather high, but nevertheless a good average 
line can be drawn. 
It is now possible to define a cross section by the 
condition that its value is exceeded in one-half of 
these 3-sec intervals, and this appears to be an easy 
4By J. L. Lawson, Radiation Laboratory, MIT. 
operational way of obtaining cross sections. However, 
this still does not represent what could be called the 
average value for each 3-sec interval. It was found very 
early that it was very difficult to adjust a signal gen- 
erator to the average value of the signal. It is much 
easier to adjust to the top value that has occurred dur- 
ing an initerval. The reason for this is that the signal 
is quite often fuzzy and filled in by propeller modula- 
tion. Therefore, the figures represented here are in 
general the highest values that occur during the 3-sec 
interval. For this reason we have attempted to see how 
the value of o depends upon the interval timing and 
whether or not it is permissible to put this value into 
range formulas in the usual way. A rough working 
model is the following: If these cross-section values 
are reduced to 60 per cent, they may be used in the 
range formula presented in the previous paper to ob- 
tain the correct operational radar range, The cross 
section averaged over the lobe structure in the front 
aspect or tail aspect of a plane would be lower than 
these values by probably 50 per cent. 
Some representative figures are as follows: Fighter 
aircraft usually vary from 1 to 200 sq ft; medium 
bombers, B-18, Beaufighter and similar aircraft range 
from 4 to 600 sq ft; and heavy bombers, B-17, 800 
sq ft. The larger bombers such as the B-29 have not 
been measured but are estimated to be of the order 
of 1,200 sq ft. 
Discussion 
To a question regarding the wavelength dependence 
of aircraft cross sections, the reply was that such a 
dependence was a function of the structure of the air- 
craft. Outside surfaces having rounded structures such 
as wings, wires, and similar members have a cross 
section which is essentially independent of wavelength 
and produce random scattering, provided the frequen- 
cy is high enough. As the frequency is lowered, reso- 
mances in the structure of the airplane and differences 
in the wings may appear. This might possibly cause 
differences with regard to polarization. At S-band and 
higher frequencies there seems to be little dependence 
upon frequency. These figures have been checked at 
S and X bands with essentially the same results. No 
sensible dependence on polarization was observed, in- 
dicating that at S-band or higher frequencies, this sort 
of cross-section value will apply. 
Ohio State University is conducting an extensive 
program of cross-section measurements on various 
types of aircraft for a variety of frequencies up to 500 
me. Measurements are made for all aspects of the air- 
eraft and for both vertical and horizontal polarization. 
The procedure used is to scale the aircraft down to a 
convenient model size and to use a correspondingly 
higher frequency. 
The results of these measurements exhibit a confus- 
ing lobe structure. In order to give an overall descrip- 
tion of the behavior of the cross section, a reasonable 
