INVAR - A FINAL REPORT

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP67B00341R000800150001-6
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 54
• Document Creation Date: December 23, 2016
• Document Release Date: November 21, 2013
• Sequence Number: 1
• Publication Date: June 30, 1965
• Content Type: REPORT

Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
S 21,
Copy of
TECHNICAL REPORT NO. 17
INVAR - A FINAL REPORT
Scientific Engineering Institute
Peter A. Hall
June 30, 1965
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
SUMMARY
INVAR is a program to produce a high-resolution ground mapping
radar to be used as an auxiliary sensor in the A-12 aircraft. The air-
borne equipment is the APQ-93 side-looking radar including a synthetic
antenna. The hologram record is optically correlated on the ground by
the 9015 optical ground processor. The final map is on two parallel
strips of 9-inch roll film and represents a swath 20 miles wide over a
range interval from 20 to 40 slant miles from the aircraft. The present
equipment has a resolution of 25 feet and weighs 950 pounds.
Two complete flyable radar systems with spares, a third experi-
mental system for flight tests in an F-101, ground support equipment,
one continuous optical processor, and two static processors for detailed
work have been produced under the program.
The radar includes a CFA transmitter which provides a 30 nano-
sec pulse with a peak power of 1 megawatt at a prf of 4 kc; a receiver
with a parametric amplifier front end and a noise figure of 6. 5 db; a
recorder with a 1 mil spot size; and a motion compensation system that
steers the antenna and controls the LO frequency in the receiver. The
optical correlator uses a carbon arc source, a wedge interference filter
for range compensation and has a resolution capability of 20 1/mm on
axis.
The radar has been test flown in an F-101 aircraft for a period
of three years. Minor changes were made in the recorder and the
correlator to accommodate a reduced altitude of 45, 000 feet and a speed
of 850 knots. Under these conditions the map is 10 miles wide instead of
20, and under most conditions is about 90 miles long which is the limit
of full speed flight for the F-101. These maps include all types of
- i -
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
targets and are believed to be the best radar maps in existence. Reso-
lution in the range direction is limited by the transmitted pulse length
and the bandwidth of the receiver to approximately 20 feet. Resolution
in the azimuth direction is approximately 10 feet. However, corner
reflectors separated 5 feet on the ground have been resolved in a radar
image.
With the exception of resolution, the present INVAR system has
been able to fulfill most of the specifications which were contained in
the original proposal of 1960. It is now felt that the original resolution
of 10 feet in both directions from full altitude can be obtained by a
combination of available modifications. In addition to a resolution
improvement, there is an urgent need to discover why the present maps
do not reveal the detail that is implicit in the resolution already demon-
strated. Recent theoretical analysis has shown that linear signal pro-
cessing is necessary for high quality images. The original specifica-
tions compromised linear processing in order to accommodate the wide
dynamic range of radar targets. It now appears that additional effort is
necessary to extend the dynamic range of the recording system and to
reduce the background light in the optical correlator. In particular,.
additional testing and experimenting on the optical correlation process
is necessary to evaluate the practical limits of the dynamic range of the
recording system and the minimum detectable signal in a practical
correlator. This problem is further complicated by the fact that both the
radar and the correlator use coherent sources for illumination. However,
information gained from such an investigation is almost essential for an
optimum design of a second generation system and promises the best
chance for improved radar mapping, whether the radar is flown in air-
planes or satellites.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
It is recommended that the radar be given a final test in the
A-12, that the quality of the radar image be improved and that a
shorter transmitt er pulse be incorporated in the present system.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
TABLE OF CONTENTS
SUMMARY
TABLE OF CONTENTS iv
I. INTRODUCTION 1
GENERAL DESCRIPTION OF THE INVAR SYSTEM
1. Original Design 4
2. Present Design 4
3. System Operation 6
4. Description of the Major Components 9
4. 1 Transmitter 9
4. 2 Recorder 10
4. 3 Receiver 31
4. 4 Antenna 12
4. 5 Ground Processor 13
5. The F-101 Flight Test Installation 15
RESULTS OF THE F-101 FLIGHT TESTS 17
IV. POSSIBLE IMPROVEMENTS
1. Limitations of the Present Design 22
2. Need for Linearity 23
3. Need for Reduced Background in Correlator 25
4. Need for a Laser 25
5. Need for Improved Sensitometry 26
6. Need for System Optimization 26
7. Need for Better Signal-to-Noise Ratio 27
V. RECOMMENDATIONS 30
REFERENCES 31
APPENDIX 32
-iv-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
:a
fir
1111!
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
I. INTRODUCTION
INVAR is a program to produce a high-resolution ground mapping
radar to be used as an auxiliary sensor in the A-12 aircraft. The air-
borne equipment is the APQ-93 side-looking radar including a synthetic
antenna. The hologram record is optically correlated on the ground by
the 9015 optical ground processor. The final map is on two parallel
strips of 9-inch roll film and represents a swath 20 miles wide over a
vat range interval from 20 to 40 miles from the aircraft. The present
equipment has a resolution of 25 feet and weighs 950 pounds.
6.1
The APQ-93 operates in all kinds of weather including cloud
Ot_ft cover and darkness and on those occasions when it is not possible to wait
l?
FRI
for good weather conditions it can get radar pictures which can be used
for a "crises evaluation". Another advantage of the radar is that
targets of interest present a different return at X-band than they do at
visible light. Most metal objects and many man-made structures have
an almost specular character at radar wave lengths. In this regard,
radar provides the photo interpreter with additional information that can
be used to increase his knowledge of the mapped areas or to decrease
his time in locating and identifying certain specific targets.
In the original proposal in 1960 the resolution was set at 10 feet.
It was felt at that time that 10 foot resolution would allow photo inter-
preters to discern and identify small installations; to recognize changes
in these installations; determine a buildup or withdrawal of military
concentrations; to spot new installations; new scars on the landscape;
find new fenced areas. However, in 1960 there were no high quality
radar maps made from holograms and it has since been learned that
radar images made from holograms produced by synthetic arrays are
-1-
Declassified and Approved For Release 2013/11/21: CIA-RDP67600341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
not as clean as the images produced by conventional radars of the same
resolution. This effect is associated with the coherent nature of both the
illuminating radar and the optical correlator. Nevertheless, the resolu-
tion of synthetic antenna systems at these proposed ranges is so much
better than conventional radars that the maps made with synthetic antennas
are far Superior. S.E.I. believes that the maps made from the F-101 in
the INVAR program are the best radar maps in existence.
The original 6B proposal was made by S. E, I. and was prepared
in cooperation with Westinghouse Aerospace and Itek Corporations. S. E. I.
has acted as technical supervisor for the government. Westinghouse has
designed and built the radar and conducted the tests in an F-101-B. Itek
has designed and built the optical ground processor, porrelated the
flight test films, and under a separate sub-contract to Westinghouse has
designed and built the recorder. Minneapolis-Honeywell under sub-
contract to Westinghouse has designed and built the single axis platform
for motion compensation.
In the course of the design of the equipment it became apparent
that the desired 10 foot resolution could not be achieved within the limi-
tations of weight and space imposed by the vehicle, and the requirement
upon resolution was formally reduced in January 1963 to 25 feet. Also
at that time other changes to the transmitter and the motion compensation
system were planned which increased the total weight of the equipment
by about 50%.
Delivery of the first radar set was accepted by the government
in January 1962. The ground processor had been accepted the previous
Decerrber. The first flight test that produced satisfactory data occurred
in May, 1962. During 1962 and much of 1963 various modifications to the
-2-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
radar and the processor were invented and implemented so that by December
1963 very good radar maps were made in flights S-86, S-87 and S-90. The
complete radar and processor were shipped to the Western site in March of
1964 for flight testing in the A-12. That opportunity did not occur and the
equipment was therefore sent to Baltimore in November 1964 to await a
later opportunity for final testing and to serve meanwhile in the F-101 flight
test program.
The radar and ground processor are now located at Westinghouse,
Baltimore, where flight tests are continuing in the F-101. Itek has just
completed a noise study on the optical processor utilizing the spare glass-
ware from the main instrument. S. E. I. has been engaged in experiments
and analysis to improve the image quality of the radar pictures.
The flight test program at Westinghouse is expected to continue
until the end of the year. Various minor modifications of the system may
still be made, but apart from final tests the original program has been
essentially completed. S. E, I. is therefore concluding its active partici-
pation in the program.
This is a final report on the phases of the program supervised by
S. E. I, and is intended to give a general account of the INVAR system and
an appraisal of its capabilities and future prospects. It describes the
major components within the system stating specific parameters and the ?
measured performance for each component and indicates those limitations
which are most serious. A few examples of flight test results in the F-101
show explicitly the present performance and serve as a guide for the pre-
dicted performance in the A-12. Finally, there is a section on recom-
mended future developments which could improve the resolution and the
map quality. An appendix includes photographs of the delivered equipment
which indicate its general size and shape.
-3-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
II. A GENERAL DESCRIPTION OF THE INVAR SYSTEM:
1. Original Design:
The original design of the INVAR system, as described in
Technical Proposal 6B of April 19601, specifies a high resolution ground
mapping radar with an overall resolution of 10 feet over 18. 5 N. M. on 9"
film and an expected weight of 600 pounds. The optical processor is a
separate unit on the ground. In an effort to accomplish these design goals,
the following developments were undertaken:
a. A transmitter to provide a 10 nanosec pulse with a peak power
of 1 megawatt and a prf of 4 kc.
b. A compatible receiver with a noise figure of 7 db.
c. A recorder with a mil spot size, i. e. 80 1/mm. capability.
d. A motion compensation system such that the overall phase
stability of the radar and the vehicle would allow utilization of
the synthetic antenna for 3/4 of a second.
e. A ground processor capable of producing useful information
over 9 inches of film at a resolution of 40 1/mm.
These developments fell short of the goals for resolution by a
factor of 2- but were otherwise largely successful.
2. Present Design:
The present design as described in a status report2 dated
January 1963, has an overall resolution of 25 feet and a total weight of
950 pounds including the frame and truss. The delivered equipment has
the following features:
a. A transmitter that provides a 30 nanosec pulse with a peak
power of 1 megawatt at a prf of 4 kc.
b. A receiver with a noise figure of 6. 5 db.
c. A recorder with a 1 mil spot size
d. A motion compensation system that has not been flight tested.
e. ?An optical correlator with a resolution capability of 20 1/mm,
on axis.
-4-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Although the resolution that is claimed has not been demonstrated
from design altitude, a resolution of 5 feet in azimuth and 20 feet in range:
has been demonstrated in an P-101 from one-half design altitude. In order
to reach the original goals the resolution should be improved by a factor
of 2, and in order to improve image quality, the SIN ratio and the dynamic
range of the system should probably be increased. Such an improvement
requires that each component and sub-system performance be examined
and that some or all of the following be accomplished:
(1) better signal to noise ratio, (2) increased dynamic range,
(3) improved stability, (4) wider bandwidths, (5) smaller recorder spot
size, (6) smaller final map area, (7) less grainy films, (8) less stray
light in the correlator. Therefore, as the description of the system and
its components is read it is also helpful to bear in mind what the requirement
for improved resolution might mean for each particular part of the whole.
SKS.ET ,14:.(10FNCY - -
"SQL;!NTED FiC;LOGR AM"
,F10.M pow!' lbRGET
, ; s , .?;
H H h
!.;
; i;{ ;t1
'
rup.N AROLMO
YR ACE.
Fig. I - Schematic of APQ-93 System Operation
-5-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
3. System Operation:
As shown in Figure 1, the radar, represented by an
antenna, is designed to fly at an altitude of 90, 000 feet and map a swath
on the ground of 18. 5 N. M. with a maximum slant range of approximately
40 N. M. The beam sweeps over the ground at 3000 ft/sec. with the
center of the beam directly broadside to the flight path. Under these
circumstances a target at maximum range would remain in the beam for
approximately one second and half that time for a minimum range target.
Returns from a single target are stored on film by an intensity modulated
CR beam and have the appearance shown as a "squinted hologram" except
that the drawing is exaggerated as there should be about 200 cycles in
the pattern instead of the 13 shown.
ENVELOPE OF
RECORDED PATTERN
RED LIGHT---.S"
IMAGE PLANE
." MAX RANGE
MIN RANGE
INPUT FILM
BLUE LIGHT
CORRELATION
Figure 2. Schematic of Reconstruction Process
-6-
Declassified and Approved For Release 2013/11/21: CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
The recorder uses a single CR. tube with a folded trace that is
unfolded optically to produce a two strip map, as shown, on 9 inch
film. In this way 1 inch of range data represents 2 miles on the ground
while every single point target on the ground is spread out as a hologram
pattern over nearly 2 inches of film in the "azimuth" direction.
The optical correlator accepts either the near or far range strip
of raw data in its input platten and, as diagrammed schematically in
Figure 2, collapses the recorded hologram into a reconstruction of the
original point target that produced it. A different color is used for each
range interval to compensate for the different focal lengths of the hologram.
In order for the radar system to provide holograms that focus
properly it is necessary either'that the antenna travel in a perfectly
straight line or that the phase of the recorded signal be varied to com-
pensate for any irregularities in the flight path. Uncompensated lateral
motion of the aircraft will cause image defects such as lateral shift,
improper focus, side lobes, and ghosts. Therefore, the specifications
for system stability are dictated by the requirements for image quality3.
The stability of the system is maintained in accordance with the
diagram shown in Figure 3. The INS by Honeywell steers the antenna
broadside to the flight path with an absolute accuracy of less than a
beamwidth and a relative accuracy of a small fraction of a beamwidth.
When the antenna is nearly broadside the frequency of the retur-
ning signals will lie close enough to the frequency of the transmitted
"stab"o frequency so that the beat signal between the two will fall in
the bandpass of the DFT. Under this circumstance the DFT monitors
the average doppler return and senses any deviation from zero doppler.
The sensed error is used as a fine control on absolute steering.
-7-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
1_11Elfgation Systernia
2
I STABILIZES ANTENNA BEAM
ORTHOGONAL TO FLIGHT PATH
2 ERECTS AND POINTS ACC.
3 CONTROLS FILM SPEED
SUBTRACTS DOPPLER?"
SIGNAL PROPORTIONAL
TO LATERAL VELOCITY
3
_?___FPTGPPLER FREQUENCY TR?AC-KEIT]
VERNIER CONTROL TO MAINTAIN
AVERAGE DOPPLER AT ZERO
A ? IGNORES ROTATION OF AIRCRAFT
. BUT
13 ? CANCELS OSCILLATION
ROTATION OF AIRCRAFT,,?
FLIGHT PATH
I
T
\ L
\, A
..; ACT1JAL
B iji -----1.--
Li
FRECE IV ER
RECOIRDEFE
DESIRED ANTENNA POINTING
Figure 3. - Schematic of Motion Compensation System for
APQ-93 Radar
_1 The time constants of the INS and DFT are different so that rapid
rotations about the yaw axis as in Drawing A will be removed by the INS
and ignored by the DFT. But slow oscillations, magnified in Drawing B,
will be compensated for by the DFT and are presumed to be beyond the
limit of detection by the INS.
P
,
In addition to steering the antenna, the INS supplies a signal to
erect the accelerometer which detects the lateral motion of the aircraft.
The output from the accelerometer controls the phase of the recorded
-8-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
hologram by adjusting the frequency of the variable frequency oscillator
(VFO) in the receiver. In this way, the Motion Compensation System
maintains the radar antenna in almost straight and level flight and cancels
the residual perturbations electronically before the radar hologram is
recorded.
The space available for the antenna permits stabilization about
only one axis, namely the yaw axis. Partial corrections for pitch or
changes in angle of attack can be made, however, by using the single
axis control to stabilize the direction of the central axis of the fan beam.
This maneuver maintains the correct position for the center of the beam
pattern but causes. some degradation at the edge of the map. If the air-
craft's ground speed changes, the INS will modify the film speed in the
recorder accordingly.
From analyses on the present equipment4 and from flight test
data in the F-101 that is similar in performance, it is concluded that the
motion compensation is more than adequate for a final map with the
present 25 foot resolution.
4. Description of the Major Components:
The equipment produced under this program comprises two
complete flyable radar systems with spares, a third experimental
("brassboard") system for flight tests in the F-101, ground support equip-
ment, one continuous optical processor, and two static processors for
detailed work. The equipment is located at Westinghouse Aerospace,
Baltimore, and is currently producing radar maps for an F-101 aircraft.
Photographs of the major components are included in Appendix A.
4. 1 The Transmitter:
Three models of the present transmitter5 (Figure A-1)
-9-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
have been delivered. It is enclosed under pressure in a semi-spherical
tank 40 inches in diameter. It weighs 264 pounds. It delivers a 30
nanosec. pulse with a peak power of 1 megawatt at a PRF of 4 kc. from
a crossed-field amplifier (CFA) and a TWT driver.
The original transmitter employed a resonant
ring6 developed at S. E. I. which produced a 10 nanosec pulse of 250 kw
peak power. This transmitter represented an advance in the state of the
art and achieved the pulse width that was originally specified, but the
power output could not be increased and the ring had to be replaced.
The present CFA does have adequate power for the
present operation but the 30 nanosec. pulse is three times as wide as
originally planned. However, when used with a 1 mil recorder to map a
20 mile swath, the 30 nanosec. pulse will not degrade range resolution
appreciably. It represents a state-of-the-art development for short
pulse, high power, airborne transmitters. If the pulse width is reduced,
or if linear signal processing can be successfully maintaned more power
will be required. More is said on this topic later.
4. 2 The Recorder
Four models of the present recorder7 (Figure A-2)
have been delivered. The package is 47 inches long, 14-iinches wide, and
19-i- inches high. It weighs 185 pounds and holds 500 feet of thin base
(3- mil) or 250 feet of thick base (5. 5 mil) Eastman Kodak Plus-X film.
It has a single Westinghouse WX4903 CRT with a p-11 phosphor which dis-
plays a rectangular. pattern 4i- inches long by wide. The pattern is
produced by the combined action of a triangular sweep and step deflection
circuits so that when it is unfolded by two parallel optical channels the
two ends of the sweep trace are joined end to end on the nine inch data film.
-10-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
The video signal is applied to the CRT grid to produce a corresponding
variation in density on the recording film. The film is transported past
the imaging slit at a speed that is precisely synchronized with the air-
craft's ground speed.
In the original design the unfolding of the CRT
sweep was to be done by means of a fiber array. The required array of
optical fibers was built and achieved a resolution of 700 lines/inch, but
it produced severe streaking. A second fiber-optic CRT8 with only a
single trace was built and tested in excess of 1000 lines/inch with no
streaking but it was never incorporated into a recorder. It is presently
being flown by Conductron Corp. in their experimental system. Also, in
the original design the CRT was intended to have a ,12-- mil spot everywhere
along the 4i inch trace. In practice the best tubes currently available
produce a spot approximately 0. 65 mil at the center of the trace. Such
measurements are made using about 11 cubic feet of laboratory power
supplies with adequate shielding of either space or mumetal or both. The
present recorder in the F-101 environment performs at 40 1/mm or 1000 -
lines/inch.
Other recorder problems that required considerable
attention were the development of the high-voltage power supply for the
CRT, the design of a variable speed film drive, the elimination of stray
magnetic fields which perturb the CRT beam, the design of adequate
vibration isolators, and the design of a flat optical image field.
4. 3 The Receiver
The present receiver5 uses a parametric preampli-
fier, a TWT rf amplifier, two IF amplifiers with a band-width of 60 Mc,
centered at 120 Mc, a synchronous detector and a bi-polar video amplifier
-11-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
WN
with a bandpass of 45 Mc. The response to a 30 nanosec. half-sine wave
pulse stretches the pulse approximately 10%. The receiver uses a stable
oscillator (STALO) as a reference signal to determine the phase of the
received signal; a variable frequency oscillator (VFO) to maintain the
desireduoffset" and to adjust the phase of the recorded hologram in res-
ponse to signals from the accelerometer; a separate channel for the DFT;
and other specialized circuits. It has the required bandwidth and shaping
NE
for the present transmitted pulse; and the stab o is stable to a few parts in
1011. If a new transmitter were to be built with a shorter effective trans-
mitted pulse, then a new receiver with commensurate increase in band-
width would have to be built also.
4. 4 The Antenna
The present antenna5 (Figure A.-3) has a rectangular
aperture 100 inches by 13 inches. It is composed of slotted waveguide
rigidly supported by a honeycomb structure and fed from the center so
that the radiation is broadside to within 15 minutes of arc. The beam
width is 0. 75 degrees between half-power points in azimuth while the
vertical beam has a csc28 cos ?2 pattern of approximately 20 degrees in
width. The maximum gain is 31. 5 db. This antenna is?some 3 feet .
shorter than the original proposed design because of space limitations in
on the aircraft. The reduction in length has broadened the beam somewhat
and reduced the gain by a db.
The length of the synthetic antenna is equal to the
? width of the antenna beam which is 3100 feet at 40 miles and is half that at
20 miles. The doppler spread at Mach 3 and X-band radar for this beam
amounts to 760 cycles and is independent of range.
-12-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
4.5 The Ground Processor
The present ground processor10 or correlator
shown in Figure A-4 is an optical system 12 feet long mounted on a
steel frame approximately 4 feet wide, 6 feet long and 6 feet high. A
schematic of the processor is shown on the next page. Light from
the carbon arc is condensed on the knife slit, which is variable in
width from 10-120 microns. Light from this line source is rendered
parallel in the azimuth plane by the collimator. The data film, con-
tained in a liquid gate, modulates the light beam. The field lens images
the knife slit onto the stop at the frequency plane, inside the relay lens.
The field lens also converges the real image produced by the hologram
in the azimuth direction to a line somewhere ahead of the frequency
plane. This real image is relayed to the image film by the cylindrical
lenses which are, in effect, one lens made in two parts because of the
redesign forced upon the correlator when the film speed of the recorder
was increased. The range dependence of the hologram is compensated
for by a variation in the color of the light. The range filter at the output then
accepts the appropriate color for that range interval. The relay lens images
the data film in the range direction onto the output film. The image roller
tracks the input film through a close tolerance gear and rim drive mech-
anism. A TV monitor is available to view the output image with the aid
of a mirror as shown.
The input film to the processor is 9i-inches wide
and up to 500 feet long with two parallel data strips, the near range and the
far range (see Figure 1). The chemical development of the data film is
done on the ground under carefully controlled co nditions. The output is on
two 91 inch films, one for each strip. Normal processing speed is 2
inches/minute of input data or . 45 inches/minute of output film due to the
-13-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
am am gm gm moi NM NM MI MN 11?111 NM MI II= AIM, Es , ANL am
r-, r Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
KNIFE EDGE
COLL1M et\-i 0
LIQUID GKTE
? P U\NE - ?
/
?7
miRiRoR
/\/
rt ELD LEN5
.601-
N
C 0 IND D t--.)
'iTE INA
IMGE OL.LER
OUT PLYT
RiL\NIGE V-\L7ER
-TV CAME RP.
DESIGINIP-NTE
G
DIRECTION
OPT\CAL PROCESSOR
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
stretch-out factor which scales to approximately 30 miles/hr of
final output map.
The processor has a resolution of 20 1/mm. in
both range and azimuth which means that it can resolve targets separ-
ated by 15 feet on the ground when the data film is taken from full
altitude or by 7i feet on the ground when the data is taken at half-
altitude as in the F-101.
There is also a Detail Correlator which uses the
spare optics from the big machine and a neon gas laser. It can examine
any selected area of raw data amounting to 3 inches in azimuth and
4 inches in range. The detail machine allows variation of the individual
parameters of focus, squint, tilt, and exposure in order to optimize
the correlation for selected data. In addition there is a second detail
correlator available in Baltimore primarily as a research tool to inves-
tigate and optimize system parameters.
5. The F-101 Flight Test Installations:
Flight tests have been conducted at Baltimore in an F-101
bailed to Westinghouse and flown by Westinghouse personnel. The first
recognizable map was made over Annapolis, Md. in May 1962. As of
? June 1965 there have been 169 flights and it is expected that the flight
test program will continue into fiscal 1966. The equipment installed
in the F-101 is the same equipment that was delivered for the final
vehicle with the following exceptions: (1) the accelerometer is of different
design and manufacture, (2) the INS is replaced by an APN-102, (3) the
antenna is mounted in an external pod and stabilized only in pitch instead
of yaw. Also, for the F-101, the recorder is temporarily modified to
take account of the lower altitude and lower speed. (1) the sweep speed
-15-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
1
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
of the CRT is doubled, because the range interval covered by the antenna
beam is only 10 miles instead of 20. (2) The film speed is reduced to
1. 25 inches per second to maintain the proper focal lengths for the
holo-
grams. With these changes the holograms recorded at 45, 000 feet and
at 850 knots can be processed in the same system that will process the
final data. An incidental consequence of this change in recorder para-
LJ
meters is that the recorder spot size and the processor resolution limit
are less degrading on the present maps than they will be in the final
version. Figures A-5 and A-6 show respectively the complete F-101
assembly mounted on the bomb bay door and the antenna pod suspended
fl
fl
I
beneath the airplane.
-16-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
III. RESULTS OF F-101 FLIGHT TESTS
All of the experimental results for the INVAR system to date have
been obtained in an F-101-B aircraft at either half-altitude or quarter-
altitude. By January 1965 the maps being obtained at half-altitude had
achieved substantially the quality that they now have and are believed to
be representative of the results to be expected at full altitude in the A-12.
However, when the system is flown at full altitude there will be some
reduction in resolution. Specifically, it is expected that the resolvable
interval in range will increase from the present value of 20 feet to 25
feet. ? This increase will occur because neither the recorder nor the
processor is quite able to handle twice the present amount of range infor-
mation. The degradation in azimuth resolution is expected to increase
from 10 feet to 15 feet because of increased speed and altitude. With
regard to the signal-to-noise ratio, flights at quarter altitude have
demonstrated that there is 15 db to spare in S/N ratio and the quality of
the map is expected to remain about the same as it is now.
The pictures that are included here were chosen to show typical
f--i
, _
renditions of good areas, to allow a subjective evaluation of the resolu-
IT tion and image quality obtainable, to show how signal-to-noise ratios
affect the maps, and to document the resolution obtained against a corner
reflector test range. It should be pointed out that photo interpreters
prefer to work with the original negatives and that different contrasts in
the printing will affect the appearance of any map.
Flight S-137, flown 6 January 1965 over southeastern Pennsylvania,
as shown in the accompanying overlay, Figure 4, has been selected as a
typical 45, 000 foot flight. The area mapped lies to the right of the air-
craft and covers a strip approximately 10 miles wide and 80 miles long.
- 17
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
? Snyderto
rg
mr.=7?/-- _ 1
.:1
NumIdial Declassified and Approved For Release 2013/11/21 :
inther Res.
Home.
town
42 -In llingtown Brandonvilla.
Delano
se? te, aha4by ?
Shenandoah cit
Cost Cr. Shaft
c4Centralia9
Pazinos
Fishers
Ferry
H. er 19 G
shalls___s,uen CUd
Dornsife ahon:tyt
Cr. 6
Rebirth tam, /6.
Red Cross
Flerndo
dat
Urban
Hebe
Lock
Kill
- 0llirord
-a-Mand2.:0e _
.60?=?' ?
GoirlOn Dark
H ckscher. Water
Buck ville
Run
Q,c
Barn .s
vine
4
9
CIA-RDP67B00341R000800150001-6
Mabel
St
Y Glair
Andreas
Normal
?9,gIP
Bowrnans-
town
Ashfield cP
.?.:7Tunnel
Cr. Slatingt n
Mat:Amli oh
almerf
Little
Gap Kunkletosert.;!:'
n' Isville
Moore tow
Crony
'ors
o?uuro
stsurg
?saw*
rardle
pillos,C, Erdman
Grata
amine
Glen
17696
iComsco Williamstown Ci
-
ReineTton
N. Pine Gwve
Woe-
Pine throve
'Suldburg 11 II
2,2arsii
Forestville
Mine sville
nBranchdale 9
'11 *ue"A"Pottvi
Newtown
rZerbe P.O., Cressona
/
_Tremont
Friedensburg
RoedersvIlle
F urtgin
ona Id s
fWanamakers
E?ld,??
Schuylkill Landing.
Haven sill
11 Auburn
Enders
---)Strau stayn
183
Berne
Shartlesvine
Shoe akersv
Centerport
K GS
Railroad
78 5
enharts.?
ville
rVi ginville
Crystal *
Cave
jordn ville Neff.:
NTerrp o I ?Il."
Lynnville
E H
Claussville
Fogelsvill
Lo'fryth
5 Egypt
cooai
Catasadgua
.lerdangullehn
Northampt ? n
*22 Sf 9
eithwviiieTrexterto%
Montere
tri-1
* Lyons
Fleetwood
en New Huffs
Pricetown Jerusalem Church
Masa-
tawny
Kutztown
Iwjrtau.7
Bowers mertztown
L_Seisholtzville
Fredericks.
ville
cu,"
U P
,er
L yttesto
Rehrers.
burg
Mt. Aetna
-.1 Mt. Zion *
Paztobta
'111SB
a
Mete detail
other side
Union Depos
Hummelst n
Annville
? 89
BA
ntin
?do
Host seehork0
?94?Stouch bu4rg)
eConrarl Weiser StateH o,
Hoer steed
Robesonia
Blandon
Aye; NV
ti+
Naz reth
5
Sc.
cksville
I. r
Long
Valley
ester
Martins
tockerto
Washingt n
11
Newburg
hillipsburg
'' 1VUetnIenem
12; Hill GIL
ltHelleFLOWD Bu,eere *
a13
ilentown m.Los.t R. Caherls
Emmaai
tern Cruz
tee It
mid;row
oyalto
Os
lebroo
Falrnouth
Bellaire
Elizabethtown
Rheems
32
"
land.
ex, 4.44ernel:er wn *Rx
* /.6r
itlie s Lincoln3i ...61 Ephrata 3e
h Go ,,,,,,inci,eiTgirlf
Roth
v Ile fa
b MINH)
Or2n 19
772 A/l
./6E Petersb Penn -
?
41de
""?r11,1.? T.
AU*
1471
Mt. Wolf 141 iLlac. a tW
nchester Marietta
C. Wrightsville 1
lireAlgart
gelst
B,
Spring Swv
Grdve Y
Seve Vall
16
Jefferson
(Coder.
P. Oa
Glen Rock
Violet
Hill
odbecks
Skye
lenville? gm Railrorid ?
Freedo
Dallas
town
neboro
nchester Preltyboy
Rex.
...a\\ 0 \
Washinidon Denville
* 1 { ? ... sZi ? S
* r\\.,, Co
,
: i sor Safe Harborn 'thge
ElE0 cy
? ? ? ? - SA
RedrO 5 .)5\
Brogueville Qs:go. Mt. Nebo
Collinsville
A
Hinkleto
Brownstown B ue
New Holland
? Pleasantville
rm narawny P.0.1
cite(A9 vi Bechtels.
hanes. ville
Boy own
el
PenrenereeA
no 41.516 eyfe tk/30.
1;:onGibr Ith`
1431Birdsboro
ers?
Geiger
Cede
Lane
B arto n
dge
organ.
onocacy
Fgc CH CREEK
C.ileteSEt.nt:,,cedatsrille
" ? 1-14awell Vi lag*
St. Peters*
Joa
Elver ?n
Hereford
Hose
Corn
Pa
? Green sub
Pennkur
E hbach
9 PerNtoem. en
isenruvine4
Gilberts-
vine,*
Cu
* Geryzille
Finland
Delmont G.
Scoot Res.
Rex. \Tylerspoct,
Green ap..?, Surnn y-
whtteat?0"?'
9
Upper
Kintnersville
Fe r nd It%
lley PIVearleynt Uhlerstown
?Applebachs.
ftn:
ville
Kellers
St. P
oftbitchlanndlown Ottsville
Churc
pi:IP teen, t
uake .co rninster. .
town /6h
l',Lu be lite
umume Hagerrille ????
Milford
9 Riag/ Plumstead-
?erkasi * D'u1b"L
? Glen
*Fri:s yles
131wrnO'no owng:
.. M c anicsville
Tel o d
0.nGwaenville
SH"r le B?. Ul C 4K s
?' New Britain 37 *
o
IIIii 'gag 6714t
L*8-
4JOCOJ
Lonl grt
1
tiaNekharnin,y
Prospec ville
Ringoes
6
Stockton
nter Br.
ellers
Ile
'4 *
M." Furlong
Men.
&lie., Pineville
'Irr 4,1 Pie.anrir,
?-?le
'Mom,
Lambertville
Laha
Buckingh
SF4inretil
S rGIO
'.GROvE
Harts-
ville
Warminster4 BM corgi yci.7 RI hboro
Iffi]
aibljOrhO IbSs'ine La ghorne
Yard elf
Honey Brook
Bird in
Hand
TiVee/ ) 9
Gordo vill
Paradi e
Willow
St. tr sbUft.,6,9
Refton er,,%,
HessdaleW
N w Providence
???Felton
Winterstown
Laurel ilk
Cross 'INV
Roads
Norri0fille
Mc Calls
Fy.
Muddy Cr.
Forks
Fawn
Grove
Shaw vill
Bethesda 'cdn
cf1
stlAl. rr
e4-e,
,
Power -A"
? Sta
Delta ? 7
Cardiff
hit ford
arryville*
Buc
Mechanic
Grove
1311"11111
Gap a Sadsburyvilte
arkesbur
Christiana a? ?
Bart Ft
Bartv Ile
Furniss Liute uyteri?
Peach
Bottom
Britain
Wakefield Gly
mittinghain
Noru.shoe ,3
Res.
Flom evil!
VF
uthviiires-
vg
ra4-
to-c:7---gia 6 alvern
Glen.
Em
ji. ,sq,uare
oPoen
MortonvEiIlebreoH.L:74:2.: 7,wer
WEI T wer m
Gardens 11 A'??*11* ts
--'
LOdwigs
Lo_o w 6 i Nov_
Chester.
3
Eagl -
Devasilt
Lynde
Brandpnne 27'. P40Ii
a hington
Cross
Morrisville
33
TRENT
'
? fi
f airles.s Hills 11
x Illi;
.716viL2OWII40 ) r,,
. 1 % ?,. ? &t,,
9?"n61 .. (tor
,
Otto ?vAt;
D .
ver,
Conshohock n - !,,,0
'.N.
r'e,e v
It
\ : e- _11._4 _ qt....,...wpam
.,
's lita/frde11311,,1
? ?:.
eie- 614:1. adda . ,
FUI 8 amsden
--*= l'Ay\kb.?4
s4. t 3 *
!ve,Glocuitcyesteitil _
New
xford London
garnsley
Hickory
Hill
PA.
?
Calvert
Dorian
Unsonvill
ney
Glen
Mill
: .
London
Grove
7
Tough
enarnon
Landenberg
ci
Strickersville
Stanton
had ds
Ford
PA
DEL.
ville
More 011 11
oth side
Parkside
Ian
Penns Grove It,
st*
Mullica Hill
Rising Sc
Conowingo272 *n
?.72,':::ng? 276 'Kew
Woodstown
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Jobs-
Mt
Holly
Marlton
Medford
E TRAVEL SERVI
ur routing counselors help plan
r motor trips anywhere in the United
ates, Canada or Mexico. Ask for
quickest" route if you're in a hurry or
"scenic" route for leisurely sight-see.
ing. You
Happy M
with trav
route-re(
ersdispl,
Happy Motorii
HUMBLE TOURING
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
(.7
The maximum slant range from the aircraft is 20 nautical miles which
corresponds to 18. 6 n. m. (112, 000 feet) measured on the ground from
directly beneath the aircraft. The flight lasted 6 minutes. Mapping
was at 1400 feet/second. Five photographs and an Esso road map with
L _J
a flight plan overlay have been selected. The last four photographs are
oriented on the road map.
7
LJ
Figure 5 is a composite of a 25 mile stretch of both the near and
the far range covering both rural and industrial complexes and showing
on 3X reduced scale the kind of coverage expected. Comparing this
picture with the road map of Figure 4 one can pick out metropolitan
areas of southern Philadelphia, Chester and West Chester. Industrial
facilities along the Delaware river, the city airport, and major highways
are prominent features. The separation between the near and far range
is due to the 5 microseconds lost in the turn-around of the CR trace and
amounts to less than one-half mile.
Figure 6, Area #1 is of Chester, Pa. and the Delaware river,
with the swamp lands of New Jersey just above the river. Points of
interest that have been labeled on the overlay are the college and part
of the town of Swarthmore, Spring Haven golf course, Rural cemetery,
the Pennsylvania and the B & 0 railroads, the Billingsport Marine
Terminal at the upper left and finally U. S. Highway 295.
Figure 7, area #2 is about half contained in area #1 but is shown
magnified 2 times. It shows sections of both U.S. 295 and the New Jersey
turnpike, a couple of branch railroads the Pennsylvania, Reading and Sea
Shore, a high tension power line, the corner of a petroleum tank farm,
and a mixture of industrial and rural topography.
-19-
Declassified and Approved For Release 2013/11/21: CIA-RDP67600341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Figure 8, area #3 is directly west of area #1 and magnified 4 times.
This print was chosen to show detail obtained on an apple orchard in
January and was printed to fairly high contrast. The regular spacing
between the trees is estimated at 30 feet.
Finally, Figure 9, area #4 is a more rural area including towns of
Downington and West Chester. The scalp for areas #1 and #4 is the same,
amounting to 3500 ft/inch in range and 4800 ft/inch in azimuth, and is a
contact print from the correlated output. In addition to the map-like
quality of the picture the elevation contours of the South Valley hills and
the East Branch Brandywine Creek give an added effect which almost
amounts to relief.
Flight S-123 was flown October 21, 1964 over Washington, D. C.
at 22,500 feet to test resolution capability on an array of corner reflectors
arranged on Bolling Field. Figure 10 shows the actual layout of the
reflectors and their calculated cross section. In the insert there is a
20X enlargement of the radar correlation. The upper left targets which
are separated 5 feet on the ground are resolved in the radar image.
Flight S-154 flown on April 4, 1965 mapped N. A. S. Oceana from
23, 000 feet as shown in Figure 11. Four days later when the clouds had
cleared a K-45 aerial camera took the photograph shown in Figure 12.
The scale of the radar picture is 1400 ft/inch in both directions and for
all parts of the picture. The photograph which has been printed to about
the same scale and has about 5 foot resolution, has the usual optical
distortion which makes parallel lines meet at infinity. It is interesting
to compare the two renditions of the same target. The radar picture
uses its own illuminating source and hence mirror-like objects tend to
return very small amounts of energy: Flat surfaces like runways, aprons,
-20-
Declassified and Approved For Release 2013/11/21: CIA-RDP67600341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
I
35'
25'
15'
15'
FLIGHT PATH
Figure /C . Layout of corner reflectors at Bolling Field and the radar image
from a correlation of S-123 enlarged 20X. Elongation of the reflectors in
the range ,9irection is drawn to correspond to the pulse length. Strong
targets are stretched more than small ones. Azimuth resolution of 5 feet
is inicate,3 but the 6.8 sq ft reflectors were not d.etecteA.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
fl
7
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
and roofs appear black, whereas vertical surfaces and strong scatterers
appear bright. There is a sharp difference between the two pictures in the
manner in which they portray the different vegetation. By looking care-
fully at the same area in both pictures a better appreciation of the actual
condition at the Air Station can be obtained than from either picture alone.
For instance, both pictures show the differences between the macadam,
new concrete, and old concrete runways, but the differences are different
in the two pictures. ? The same kind of differences are noted for the parking
aprons and the planes parked on them.
' Figure 13 is a radar picture and a K-45 picture of the IL S. S.
America underway at 23 knots taken on April 5, 1965 as part of a Navy
exercise. In this picture the radial velocity of the ship away from the
F-101 shifted the doppler spectrum of the returh signal by approximately
700 cycles/sec and caused the wake of the ship, which is largely stationary,
to be displaced laterally from the image of the ship. Again the radar
highlights the island and certain features of the ship while it misses some
others which are detailed by the camera. Some radar difficulty occurs
over water because of the absence of sufficient signal return to operate
the DFT. Therefore, the antenna steering must be maintained by the
vertical gyro and the APN-102 by themselves.
Finally, Figure 14 is from flight S-169 over Washington on June
2, 1965, and is included to show that there is sufficient power to map
from full altitude. The F-101 flew at 23, 000 feet and the sensitivity of the
receiver was decreased by 15 db.
-21-
Declassified and Approved For Release 2013/11/21: CIA-RDP67B00341R000800150001-6
L
fl
L
L_
F-1
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
IV. POSSIBLE IMPROVEMENTS
1. Limitations of the Present Design:
Before making specific recommendations for future develop-
ments it is helpful to review briefly some of the major factors responsible
for the present design. As in every design certain compromises were
reached as a result of requirements imposed by other components or
events not under the control of the radar designer. In a future develop-
ment some of these compromises would be either different or unnecessary.
After acceptance of the 613 proposal by the customer in
August 1960, a detailed engineering analysis of the system was undertaken
concurrently with the design and construction of the major components.
By the summer of 1961 it was apparent that a system design predicated
upon- the availability of a recorder with a mil spot size would have to be .
revised for a 1 mil spot. A -I-- mil spot could not be maintained under flight
conditions which required light weight chassis and power supplies. This
change required the recorder film speed to be doubled which had the effect
of increasing the focal length of the holograms by four. Consequently, the
optics of the correlator had to be partly re-designed. Also the increased
spot meant more attenuation of high frequency signals which required
tighter tolerances on the antenna steering because of the connection between
antenna beamwidth, doppler spectrum, and pointing error. In the summer
of 1962 the space available for the antenna was firmly established and it
was less than expected. The antenna, therefore, had to be shortened by
3 feet, causing the beam width to increase by 30% and the gain to decrease
by 2. 5 db. one way. The space available for the movement of the stabilized
antenna amounted to ?3? in yaw and none in pitch. A third factor responsible
for the present design was the requirement for additional transmitter power
-22-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
H
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
to overcome an accumulation of small losses in the system and to provide
more power than could be expected from the resonant ring. The extra
power is supplied by the CFA but this also increased the pulse width to
30 nanosec.
In spite of these difficulties, a system has been produced
which has about one-half the proposed degree of resolution but otherwise
has been able to fulfill the original specifications. On the other hand,
many of the features of the design are now far from being optimum. The
recorder has inadequate resolution, the modified processor has some
optical imperfections, the antenna is shorter than it should be, the
stabilization system has one axis instead of two and the transmitter has
too long a pulse.
2. Need for Linearity.
Apart from the unfavorable features that are the results
of compromise, there are two features that are now considered ques-
tionable even though they conform to the original specifications. One is
that signal limiting occurs at a relatively low signal to noise ratio, the
other is that the S/N ratio for desirable weak background signals will be
small at full altitude.
A technical report9 dated December 7, 1964, shows the
effect on radar maps produced by coherent background and also shows
that coherent background is always produced when clipping or limiting
occurs in the receiver or recorder. From the conclusions in this report
and from experimental evidence in some of the radar maps produced in
the F-101 flight test it appears conclusive that linear signal processing
is necessary for high quality radar maps. In practice this means that
the dynamic range must be made as large as possible, so that most of the
-23-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
targets may be faithfully rendered most of the time.
Also, with regard to image quality, it is probable that
the resolution already obtained is not being fully exploited because of
the present shortcomings in signal processing. The coherent interfer-
ence in the optical correlator between image signals and background
signals of all kinds is a major cause of reduced image quality. These
background signals include: (1) spurious signals produced by clipping,
(2) stray light from the lens surfaces and the interior of the glass itself
in the correlator, (3) light diffracted by the film grains, and (4) clutter
from-unfocused virtual images.
Present signal processing is faced with a serious problem.
The dynamic range of the present system is limited to approximately
20 db.. The dynamic range of radar targets in an average scene is at
least 35 db. and it may extend to as much as 60 db. between large bridges
and open water. In present practice this wide dynamic range of targets
is accommodated by severe clipping in the IF amplifier. If limiting must
be used the best place to use it is in the IF stages11, but any limiting or
other non-linear signal processing will produce spurious images that
focus in the output map and thus reduce image quality.
The linear dynamic range of the system is determined by
the ratio of the largest amplitude signal that can be recorded without
saturation to the smallest signal that can be 'recorded and detected in the
final map.
To utilize fully the existing dynamic range it is necessary
to adjust the receiver-recorder combination so that for the desired range
interval the strongest targets just reach the maximum allowable density
variation of the recording film without limiting. Under these circumstances
-24-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
the smallest targets that will be rendered on the final map will be those
with signals that just exceed the noise of the stray light and film in the
correlator.
7 3. Need for Reduced Background in Correlator
At present there is too much stray light in the correlator.
Holograms recorded at modulation levels that exceed the radar noise
and the noise of the recording film are obscured by this stray light at
the output film plane. In the ground processor tracking of the output
image by the output film over an appreciable range (0. 1 inch) tends to
smear the stray light by destroying the coherence between it and the
images. However, the large number of glass surfaces, the thickness of
the glass, and the interference filter all contribute excessive stray light
which reduces the dynamic range of the system. In the detail correlator
there is no tracking and hence much of the advantage to be gained by the
laser source is lost because stray light adds coherently to the images.
A study recently completed at Itek12 identifies many of the sources of
tills unwanted stray light.
4. Need for a Laser
The present ground processor was designed for a wide
spectrum of colors from'a carbon arc source in order to accommodate
the range dependence of the hologram focal lengths. Since the original
design was completed the laser has emerged as a new scientific tool.
Today a new optical design for the single color of the laser and utilizing
its great intensity could be expected to reduce the stray light and hence
improve the dynamic range by a factor of 3 or more.
-25-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
5. Need for Improved Sensitometry
1
As explained above the dynamic range must be made as
large as possible. One requirement for a wide dynamic range is a long
linear region on the data film response curve. The present film has a
response curve which is linear to within 5% from 95% to 28% trans-
mission when developed under the proper conditions. A second re-
quirement is to have a film whose "noise figure" is very small. The
?
smallest signal that can be recovered is limited by the noise fluctua-
tions in the grains of the recording film. It is known that there are
many films whose noise is less than the film presently in use but few of
them have enough "speed" to qualify.
6. Need for System Optimization
In the first three years of the INVAR development, major
problems such as stray fields and vibration in the recorder, variable
frequency control in the receiver, and antenna steering difficulties all
obscured the degradation caused by spurious clutter. Now that a reliable
system exists and these major problems have been dealt with satisfac-
torily it is time to linearize the system., reduce the background noise,
fl and optimize the sensitometry in order to produce the image quality
we desire.
L As an example of one step in such a program the most
recent improvement in system performance came about when the bias
level and the CRT drive was adjusted to yield the largest linear portion
on the overall response curve of the CRT and recording film Combination.
More work needs to be done in this same general area.
7
The first attempts at linear signal processing conducted in flights S-168,
and S-169 failed to show any substantial improvement in map quality.
-26-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
this failure is attributed to other difficulties. In particular, the stray
light in the correlator, which adds coherently to both images and clutter,-
is large enough to obscure the small targets and the fine detail of the map.
However, there was improvement in linear signal processing. In the
neighborhood of strong targets there was no evidence of spurious targets
which indicates there are no cross modulation products produced in the
modified receiver which was used for these tests. The modified receiver
includes an amplifier whose gain is adjusted so that for some parts of the
flights very few targets saturate the recording film. Under these circum-
stancps the recording is nearly linear over a limited dynamic range. It
is hoped that future tests will be able to correlate such linear recordings
and render the desired fine detail.
7. Need for a Better Signal-to-Noise Ratio
Although the present S/N ratio is more than adequate for
the F-101 flight tests and although it fulfills its promise of performing
adequately according to present standards at full altitude, it now seems
likely, that additional transmitter power will be desired. This apparent
paradox can be explained as follows: the present system has a limited
dynamic range and hard limiting in the IF amplifier. The gain is adjusted
Liso that the smallest discernible signal has a reasonable modulation level
on the recording film and all signals 10 db. greater are limited in the IF
amplifier. The future system should have a wide dynamic range and
linear signal processing. The gain should be adjusted so that very few
targets are limited. Under this circumstance the small targets receive
very low modulation on the recording film and demand a more sensitive
correlator to recover them. Such a correlator would not mask the noise
of the radar and therefore a less noisy, or rather higher power, trans-
mitter will be desired. In addition, when the fine detail does become
-27-
Declassified and Approved For Release 2013/11/21: CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
available in the radar maps, improved range resolution will be very
desirable, and this will also increase the demand for more transmitted
power.
The effective transmitted power illuminating a specific
target area can be increased in several ways. Two of these ways are
(1) chirping the transmitted signal while maintaining the same peak
power and (2) increasing the antenna gain by increasing either its length
and/or width.
Chirping not only allows an increase in effective peak
power by frequency dispersion but it also increases the efficiency of the
CFA transmitter. In addition, a chirped transmitter allows the effective
width of the transmitted pulse to be reduced and thus allows the original
range resolution goal of 10 feet to be achieved. A chirped transmitter
could improve both the resolution and the image quality of the radar map.
If a longer antenna could be installed, either in the present
aircraft or in some future aircraft, an improved antenna design might
produce the necessary increase in S/N. A narrower antenna beam would
both increase the gain and reduce the clutter from unwanted signals
existing at the edge of the present beam.
As a typical example, the present antenna is 8 feet long
and has a beam width of 0. 0132 radians and a measured gain of 31. 5 db..
This antenna has a resolution limit in a focused synthetic system of 3. 3
feet. A new antenna 14 feet long could have a beam angle of . 008 radians
and a gain of 34 db.. Such an antenna would have a resolution limit of
5. 8 feet. The two-way gain in S/N amounts to 5 db. in this hypothetical
example. If, in addition to a reduction in the horizontal beam pattern,
a reduction in the vertical beam were also made, a narrower strip on the
-28-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
ground, say 10 miles wide instead of 20, would be illuminated with twice
as much transmitted energy. Therefore, an antenna both longer and
wider could conceivably increase the S/N ratio by 8 db.
-29-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
V. RECOMMENDATIONS
S. E. I. recommends that the INVAR program be continued with
emphasis placed on certain specific aspects of the program. First, the
equipment should be given a complete test in the final vehicle; the radar
has been given a thorough engineering evaluation, numerous tests and
analyses have been made which predict satisfactory performance in the
A-12, and everything is ready for a final test. Second, during the
remaining tests in the F-101, there should be continued emphasis on
linear signal processing in conjunction with efforts to reduce background
noise in the correlator and to improve the sensitometry. There is good
reason to believe that better maps can be made with the present equip-
ment, that continued analysis and experiment can determine the parti-
cular limitations of the present equipment, and that the additional
knowledge on this point to be gained would be useful in the design of a
better system. Third and last, the design and construction of a proto-
type chirp transmitter would improve the range resolution and provide
increased power for high quality mapping.
-30-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified
and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
REFERENCES SOURCE
1.
Technical Proposal 6B dated April 1960
S. E. I.
2.
Status Report on INVAR, dated January 1963
S. E. I.
3.
STM-104; System Stability Requirements dated
January 15, 1963
4.
STM-101; Motion Compensation for Side-Looking
7
Radar 12/17/63
n
STM-102; Motion Compensation for SOARD Flight
Test 1/16/63
5.
Final Design Report of AN/APQ-93 (XA-1)
Radar 5/12/65
6.
STM-128; Resonant Ring Transmitter 7/17/63
Li
7.
High Resolution Radar Data Recorder, Final
Report 3/10/64
Itek
?1
8.
Recorder Follow-on Proposal 12/17/63
Itek
9.
Technical Report - Effects of Coherent Background
in Radar Maps produced with synthetic antennas
/
12/7/64
S. E. I.
10.
Final Report Model 9015 Processor 5/1/64
Itek
?J
11.
STM-122; Performance of IF limiting as a form of
automatic gain control 5/28/63
12.
9015 Report for Period 1 January to 30 June 1965
Itek
13.
Project 9015 Final Report 1960-1964 (2 vol.)
Itek
-31-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
APPENDIX
Photographs of Major Components
Figure A-1. CFA Transmitter
Figure A-2. Recorder
Figure A-3. Antenna
Figure A-4. Ground Processor (interior view)
Figure A-5. F-101 Door Installation
Figure A-6. F-101 Antenna Pod
-32-
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
"COMPOSITE MAP RE
DUCED 3X SHOWING
THE ENTIRE COVER-
'AGE FOR THE LAST
-25 MILES OF S-!37.
? ???
.APPROX:'I/2 MIL
Lopt IN THE CEN
,oF :THE STRIP D
'IT TO 'THE TORN A
(- OF THE CRT-TR
,74,090;.! FIT 86,000, FT
Declassified and Approved For Release 2013/11/21 : dIA-RDP67B00341R0008001500016
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Ca1.5.1.GSTIorgs. 01,1ijc S,M ai0g.
t (5.018.13 1110 CI
(111011-0
This document contains
National Defense of the
meaning of Esonc--..?
tions 793 ant! 6.
Of its contents in an*,.
person is prohibited by
information affecting the
United States within the.
-?vs. Title 18, U.S.C.. Sec-
s,ion or the revelatibn
..?-c-r to an unauthorized
taw.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
S5,000 FT. Declassified and Approved For Release 2013/11/21 :_CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
47'1
ti.V.? I ?v`
,
99,000 FT.
F/GURE-7 AREA NO 2
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
4
, Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6 kpr ? ,,,..,
it **N.':
- , -- ev , ? ,,,..4,r. .
4--d, - . ? *.::$0,7' t .i?A.N' et'
ile.*-?4- . .s- '?'
L 'Iwo
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
6000 FT.
85,000 FT.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6 FT
Release 2013/11/21: CIA-RDP67B00341R000800150001-6
Figure 11, N.A:S. Oceana. S-154, 23,000 ft., 585 knots, Scale , 1400 ft/inch
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Figure 12. N.A.S. Oceans. K-45 aerial camera
AG:
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
- -
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
-
Fic;ure 13. U.S.S. America Radar Image and Camera Ima7e.
//C Flirht S-155 date 4/5/65
/C r,round speed 570 knots
',/C standoff from target 7.9 n.mi.
Tar-et speed 23 knots
Raiz' Scale: ^Az. 350 Ft/inch, Ran7,e )405 ft 'inch
Weather; Clear, visibility to 15 miles; sea state
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
9-1-00091-00800011-17?008L9dCl-V10 1-Ziio eseeiej -101 panaiddv pue Pe!PsseloeCI
u()H,imL[mt: ocip [
- 000 ?:F. ".) "(I '1()C71(;--,L,..AA
-
tO
A C
.11"?
' 9-1-00091-00800011-17?008L9dCl-V10 1-Ziio eseeiej -101 panaiddv pue Pe!PsseloeCI lab
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
2
Pt?
,
?
Declassified and Approved For Release 2013/11/21: CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
l' ' A ''
Declassified and Approved For Release2012013/11/211/21 : CIA-R6Pe7B00344R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
HANDLE
BARRIERS
( TYPICAL)
A
ANTENNA RADIATING
FA CE
POWER DIVIDER
AND MANIFOLDS
3 PIVOT
ATTACHMENT
HOLES
HOLES
NOT
USED
4 MOUNTING HOLES
FOR BELL CRANK
AND SLIDE RAIL
ATTACHMENTS
BA CK OF ANTENNA
HOLES
NOT
USED
- - -
Figure A-30 Antenna
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
111111J.
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6
Declassified and Approved For Release 2013/11/21 : CIA-RDP67B00341R000800150001-6