• WARD AND NAKA

Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 205
Long-Range UHF Radars for
Ground Control of Airborne
Interceptors
William W. Ward and F. Robert Naka
s The standard Air Force radars available in the early 1950s had major
shortcomings for air-battle management in the face of plausible threats. At that
time Lincoln Laboratory was achieving impressive success in developing UHF
radars for airborne early warning with moving-target indication by changing
from shorter to longer operating wavelengths. It appeared that similar
innovations would also yield major performance improvements for radars
devoted to the ground control of airborne interceptors. Lincoln Laboratory
developed and fielded two different UHF radars that showed that this promise
could be fulfilled. Both had quite large antennas rotating in azimuth. A
narrowband radar operating near 425 MHz was built on Jug Handle Hill near
West Bath, Maine; it became a primary sensor for the Cape Cod System and the
Experimental SAGE Subsector. A broadband radar operating across the 400-to-
450-MHz band was built atop Boston Hill near North Andover, Massachusetts.
This radar was designed as a test bed for the development of techniques to
combat active electronic jamming and passive countermeasures such as chaff
dispensed by hostile aircraft. These radars paved the way for subsequent Air
Force efforts to achieve frequency diversity in its air-defense network.
S
raxoaio aii ioici i-naxo axo s-naxo radars
provided the primary input data for ground
control of intercepts (GCI) in the Cape Cod
System, which was Lincoln Laboratory’s early dem-
onstration of air-battle management by a central
computer. By 1954, it became apparent that these
sensors, only marginally more capable than radars de-
veloped by the end of World War II, displayed an un-
acceptable amount of clutter on their plan position
indicators (PPIs). The circuits intended to cancel out
echoes from fixed and slowly moving targets and to
display only those from high-speed targets such as air-
planes in flight were not fully effective. The
uncanceled echoes from mountains, buildings, pre-
cipitation, and occasional flocks of birds produced
numerous false targets that had to be identified and
eliminated (mapped out) before the airborne targets
of interest could be tracked from their digitized coor-
dinates. The proportion of the radars’ coverage that
had to be sacrificed in this way was unacceptably
large. Could anything be done about these problems?
Deficiencies also existed in the vertical coverage
provided by these radars. They had been designed to
detect aircraft powered by piston engines. Such air-
craft do not routinely operate much above 20,000 ft.
The advent of heavy bombers and interceptors pow-
ered by jet engines meant that an airborne threat
would soon be able to escape radar detection by flying
over the radars’ coverage volumes.
As a result of the 1952 Summer Study at MIT in
Cambridge, Massachusetts, Lincoln Laboratory be-
gan developing and testing UHF airborne-early-
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
206 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
warning (AEW) radar systems with airborne-moving-
target-indication (AMTI) capability [1]. These radars
demonstrated in flight tests some of the advantages of
operating at lower frequencies (longer wavelengths),
which are discussed in the article entitled “Displaced-
Phase-Center Antenna Technique,” by Charles Ed-
ward Muehe and Melvin Labitt, in this issue. For ex-
ample, echoes from precipitation and birds were
reduced because the scatterers were smaller in terms
of wavelength. The success of the AEW program led
radar designers to believe that operating GCI radars
at longer wavelengths could solve many of the prob-
lems that radars experienced at higher-frequency L-
and S-bands. Of course, the horizontal aperture of
the rotating radar antenna needed to be wider in pro-
portion to the wavelength ratio to maintain the same
resolution in azimuth. Keeping the vertical dimen-
sion of the antenna about the same meant that the
vertical beamwidth was broader so that the coverage
in elevation angle extended to correspondingly—and
gratifyingly—higher altitudes. The resolution in
range could be preserved by using transmitted pulses
of the same length as before.
Other benefits were associated with the move to
longer wavelengths. Engineers realized by this time
that—all other things being equal—the effectiveness
of a pulsed radar in searching for targets at unknown
positions throughout a given volume is proportional
to the product of the average power of its transmitter
and the aperture area of its receiving antenna, inde-
pendent of the wavelength at which the radar oper-
ates. Thus using longer wavelengths would increase
the effectiveness of a pulsed radar by increasing the
horizontal aperture of the rotating antenna. Transmit-
ters with higher peak and average powers at longer
wavelengths would be easier to build than those at
shorter wavelengths because the physical dimensions
of the radio-frequency (RF) components would be
larger. Increasing the size of the RF components
would reduce the likelihood of breakdown within
them because of high electromagnetic-field strengths.
Longer wavelengths also facilitated efforts to resist
jamming, an unanticipated vulnerability of radar op-
erations. In the early years of World War II, radar de-
velopers such as those at the “RadLab,” MIT’s Radia-
tion Laboratory [2, 3], were elated just to get their
equipment to work properly in the field. They had
enough problems to solve without also considering
that an enemy might try to jam the radars’ operation
with electronic countermeasures. Just as the radar de-
velopers did not initially anticipate jamming tech-
niques, the Axis forces, especially submarine crews,
were initially unaware that they were vulnerable to
detection by airborne radars. Soon enough, on all
sides researchers began to invent measures to counter
or reduce the effectiveness of their opponents’ radars.
For example, the Radio Research Laboratory [3] at
Harvard University worked hard to develop active
electronic jamming and passive countermeasures
(dropping chaff ) to interfere with radars like the ones
under development about a mile down the street at
the MIT Radiation Laboratory. It therefore became
necessary for the radar developers to devise counter-
countermeasures (CCMs) for their equipment.
With substantial electronic jamming of radio
transmissions from the western world already in ef-
fect, there was no doubt that the Soviet Union would
employ radar countermeasures. The standard Air
Force GCI radars that were available in the early
1950s offered little in the way of CCM capability. A
new generation of GCI radars needed the flexibility
to accommodate emerging CCM techniques to oper-
ate in assorted frequency bands. They also required
the ability to burn through wideband noise jamming
and provide coverage on airborne targets of interest
out to a useful extended range. The sidebar entitled
“The Air Force Frequency-Diversity Radar Program”
describes how these requirements were incorporated
in new radars. An incoming bomber force facing an
array of frequency-diverse air-defense radars needed
to carry an equally diverse collection of active and
passive countermeasures, adding to the complexity of
the aircraft and reducing their combined useful pay-
load of bombs.
Lincoln Laboratory’s contributions to this effort
were made at a practical level. Two different large
UHF GCI radars were developed and put into opera-
tion at field sites in New England.
Jug Handle Hill, West Bath, Maine
In 1954 Lincoln Laboratory undertook to cobble to-
gether a demonstration UHF GCI radar in a hurry.
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 207
Its characteristics were spelled out by midyear. The
antenna, 120 ft wide by 16 ft high, was not expected
to be a great construction challenge. Its mechanical
tolerances in terms of wavelength were no more strin-
gent than those of the ≈1300-MHz AN/FPS-3, then
a standard Air Force heavy radar for fixed GCI instal-
lations, which had a 40-ft-wide by 16-ft-high antenna
reflector. Both yielded approximately 1.5°-wide radar
beams and blips on their PPI displays. Rotating the
UHF GCI radar antenna at 6 rpm would make its
data-output characteristics essentially the same as
those of the AN/FPS-3. The new radar promised to
ix iuxi :,ss Rome Air Develop-
ment Center, Griffiss Air Force
Base, New York, let design-study
contracts for six new ground con-
trol of intercepts (GCI) radars,
each to operate in a segment of the
frequency range 214 to 5900
MHz. At that time, Air Force GCI
radars were moving through attri-
tion toward occupancy of only
two frequency bands: the AN/
FPS-7 surveillance and height-
finding radar with stacked beams
operated at 1300 MHz, and the
AN/FPS-6 height-finding radar
operated at 2900 MHz. These two
radars, lineal descendants of ra-
dars developed during World War
II, constituted what amounted to
a single-frequency air-defense ra-
dar system. The frequency-diver-
sity (FD) radar program was to re-
verse that trend.
The spread of operating fre-
quencies to be provided by the FD
radar program promised to make
it more costly in terms of payload
for an airborne intruder to pen-
etrate and survive in the defensive
radar environment, as discussed
in the main text. At the same time,
the new program would enhance
the Air Force’s GCI capabilities, in
particular its ability to feed high-
quality data to the Semi-Auto-
matic Ground Environment
(SAGE) air-defense system. Table
1, on the following page, shows
the characteristics of the Air Force
frequency-diversity radars.
Five of the six proposed radars
were selected for prototype devel-
opment, and four were produced
in quantity. These five systems in
their prototype forms were in-
stalled for testing and evaluation
at operational Air Force sites in
Alabama, Louisiana, and Missis-
sippi, part of the Mobile, Ala-
bama, Air-Defense Sector. Their
test programs began in 1959.
In addition, the AN/GLA-8
signal processing system, built by
Airborne Instrument Laboratory,
was an important common ad-
junct to each frequency-diversity
radar. This equipment included a
special antijamming console used
by the radar’s human counter-
countermeasures (CCMs) opera-
tor. As discussed in the main text,
CCMs such as frequency hopping
and PRF jitter/stagger are useful
in reducing the effectiveness of
both passive countermeasures
(chaff, for example) and active
countermeasures (spot and noise
jamming, and signal repeaters).
The wise use of the many features
of a highly flexible FD radar re-
quired special skills and sophisti-
cated technological support.
The Jug Handle Hill radar be-
gan operation in October 1955.
Without question, the measure of
success it achieved despite the
bearing problems of its gargan-
tuan rotating antenna paved the
way for the three lower-frequency
FD radars.
The Boston Hill radar, which
began operation in 1959, was con-
temporary with the AN/FPS-35
but had quite a different design.
They should not be confused.
Four AN/FPS-35s were procured
by Rome Air Development Cen-
ter under a prototype contract for
early installation at field sites and
operation by Air Force crews. Bos-
ton Hill, on the other hand, was
intended to serve as Lincoln
Laboratory’s long-term test bed
for development and evaluation
of CCM techniques, hence its for-
mal name, CCM Radar Mark I.
THE AI R F ORCE F REQUENCY- DI VERS I TY
RADAR PROGRAM
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
208 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
be a prime input sensor for the Cape Cod System and
later for the Experimental Semi-Automatic Ground
Environment (SAGE) Subsector.
The design of the electronics for the new radar was
taken directly from that of the AN/APS-70, Lincoln
Laboratory’s UHF AEW radar with AMTI, which
was then undergoing development and testing (see
the article entitled “Displaced-Phase-Center Antenna
Technique,” by Charles Edward Muehe and Melvin
Labitt, in this issue). The moving-target-indicator
(MTI) circuitry of the UHF GCI radar was simpler.
Adopting the magnetron-based transmitter design
of the AN/APS-70 was a compromise decision. Al-
though the transmitter and its associated circuitry
could be readily copied from the AEW-radar equip-
ment, the transmitter was not very powerful. Any
magnetron has a random start-up phase on each
pulse, making the radar’s signal processing equipment
more complicated. We would have preferred to build
a fully coherent radar of the master-oscillator/power-
amplifier family (see the article “Early Advances in
Radar Technology for Aircraft Detection,” by Donald
L. Clark, in this issue). That approach would have re-
quired using a high-power amplifier such as a triode,
tetrode, klystron, or amplitron, but no suitable tube
was immediately available. Furthermore, the 2% in-
stantaneous RF bandwidth of the magnetron could
not accommodate the rapid changes of transmitted
frequency—perhaps even pulse to pulse—required
for some radar CCM techniques.
Work on this UHF GCI radar, ultimately desig-
nated the AN/FPS-31 (XD-1), got under way in the
fall of 1954. Figure 1 shows the transmitter and QK-
508 magnetron for the AN/FPS-31 (XD-1) radar. A
site near the Maine coast on Jug Handle Hill, West
Bath, was selected to serve as a counterpart to the
shoreline GCI radars at South Truro on Cape Cod,
Massachusetts, and Montauk Point on Long Island,
New York, that were already integrated into the Cape
Cod System. The radar began operation in October
FIGURE 1. Transmitter of the AN/FPS-31 (XD-1) UHF
ground control of intercepts (GCI) radar. The QK-508 mag-
netron (middle) fits into a pulse transformer and is powered
by the modulator cabinet on the right. The high-power RF
output pulses travel from the magnetron through a section
of 3-1/8-in flexible coaxial transmission line (at the left) to
the vertical waveguide run (see also Figures 2 and 3). Note
the windlass (essential when changing magnetrons) and the
arrangements for liquid cooling of the magnetron.
Table 1. Characteristics of the Air Force Frequency-Diversity Radars
Function Frequency Range (MHz) Equipment Designator Contractor
Surveillance 214–236 AN/FPS-24 GE
Surveillance 400–450 AN/FPS-35 Sperry
Surveillance 510–690 AN/FPS-28* Raytheon
Surveillance and 2320–2680 AN/FPS-27 Westinghouse
height finding (stacked beams)
Height finding (nodding beam) 5400–5900 AN/FPS-26 AVCO Manufacturing Co.,
Crosley Division
*Not produced in quantity
Windlass
Flexible coaxial
transmission line
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 209
FIGURE 2. The AN/FPS-31 (XD-1) UHF radar at Jug Handle
Hill, West Bath, Maine. The vertical waveguide runs from the
equipment building to the rotary-joint housing beneath the
antenna atop the tower. Note the stairway for scale.
1955. The 120-ft-wide antenna shown in Figures 2
and 3, painted with broad vertical white and interna-
tional-orange stripes, was an impressive sight when
rotating at 6 rpm.
The Jug Handle Hill site was ultimately equipped
with two standard Air Force AN/FPS-6 S-band nod-
ding-beam height finders to give it full GCI capabil-
ity. A dual-channel AN/FST-1 Slowed-Down-Video
(SDV) system, later replaced by an AN/FST-2 fine-
grained-data system, was installed at the site to relay
data from the three radars and the Mark X identifica-
tion-friend-or-foe (IFF) equipment to the Experi-
mental SAGE Subsector’s central computer at Lex-
ington, Massachusetts.
The fabrication, installation, and operation of a
full suite of transmitting, receiving, and MTI signal
processing circuitry for the AN/FPS-31 radar was a
straightforward task. There were some interesting as-
pects to it, however. Ignition noise from vehicles driv-
ing by the radar site, which was just seaward of U.S.
Route 1, jammed the radar. AN/APN-1 FM radar al-
timeters, carried by military aircraft passing through
the radar’s coverage volume, caused interference.
There was pulsed RF interference from AEW air-
craft—carrying experimental UHF AEW radars built
by Lincoln Laboratory—when they operated within
line of sight of the AN/FPS-31. All of these disrup-
tions had to be mitigated.
The radar interference was eliminated by replacing
the fixed-tuned magnetron in the transmitter of Fig-
ure 1 by a tunable one. Of course, the radar receiver
had to be tuned to match the frequency of the magne-
tron. An arrangement for “one-knob” tuning control
of the complete radar was developed.
The unique subsystem in the AN/FPS-31 radar
was its large antenna, together with the tower to sup-
port it and the bearing arrangements and drive ma-
chinery to rotate it in azimuth, as shown in Figures 2
and 3. The bearing caused many headaches. The
original design called for the heavy rotating mass to
be carried on sets of bogie wheels at the ends of a
three-armed spider that rolled on a smooth, level cir-
cular track at the top of the tower. This installation
gave trouble from the start. The track had not been
made sufficiently smooth to begin with, and the
wheels soon wore out.
Further design studies and tests showed that this
bearing arrangement could be perfected. However,
the pressure from the SAGE development schedule to
get the AN/FPS-31 radar into full operation speedily
led to the decision to abandon the original design and
go to a large central ball bearing upon which the en-
tire rotating assembly would ride. This modification
proved to have its own problems. There was a shut-
down of several months while the bearing was re-
FIGURE 3. Another view of the tower and antenna assembly
of the AN/FPS-31 (XD-1) UHF GCI radar. The feed (in the
center at the top of the picture) is at the focus of the para-
bolic-cylinder antenna reflector.
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
210 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
worked. These mechanical problems were eventually
solved to achieve reliable operation of the large rotat-
ing antenna assembly. The experience that Lincoln
Laboratory gained in solving such problems was
shared with others and led to subsequent successful
designs of the Counter-Countermeasure (CCM) Ra-
dar Mark I at Boston Hill, Massachusetts, the Mill-
stone Hill radar, the AN/FPS-49 Ballistic Missile
Early Warning System (BMEWS) tracking radars,
and other radars.
Although the performance of the AN/FPS-31 ra-
dar was impressive, it did not meet expectations es-
tablished by scaling from the demonstrated perfor-
mance of UHF AEW radars operating at lower power
and with smaller antennas. Improper orientation of
the feedhorn proved to be the source of the problem.
The peak of the approximately 18°-vertical-width
main beam was 8° above the horizon. For best cover-
age, the 3-dB-down point of the vertical lobe should
have been on the horizon, putting the peak 4° above
it. This point was proved convincingly with the aid of
antenna patterns measured at sunrise and at sunset as
the rotation of the earth moved the antenna beam
across the disk of the sun, as shown in Figure 4 [4]. A
new feedhorn was ultimately procured and installed,
with gratifying results.
In April 1956 the AN/FPS-31 radar was found to
display clutter of an unexpected sort, shown in Figure
5. Echoes resembling returns from storms were ob-
served, but they had unusual characteristics: high
scatterer velocities, sharply defined azimuth bound-
aries, and consistent occurrence in the same general
azimuth direction—magnetic north. Consultation
with personnel from the Communications and Com-
ponents division yielded the suggestion that the AN/
FPS-31 radar was receiving echoes from the aurora
borealis. This surmise was verified when it was pos-
sible to correlate these 425-MHz observations in
Maine with those from a 50-MHz radar located at
Ottawa, Canada. Correlation of the radar data with
the occurrence of solar flares and sudden ionospheric
disturbances led to the conclusion that auroral clutter
showed up on the AN/FPS-31 radar about 48 hours
after a solar flare.
Despite the rare occurrence of auroral activity in
New England skies, the AN/FPS-31 radar was power-
ful enough to produce pulse echoes that backscattered
from the actual aurora (high above the atmosphere
and far to the north) and reached the radar at the
same time as did echoes from later pulses returned by
the much closer targets of interest. This auroral clut-
ter could overlie any part of the radar’s unambiguous
range. The velocity distribution of the ionized par-
ticles comprising the aurora was so broad that there
FIGURE 5. Auroral echoes on the AN/FPS-31 (XD-1) radar
at Bath, Maine. The range of the echoes was seen the “sec-
ond time around.” The distance between range marks is 50
miles. The clock face and grease-pencil notes on the white
tablet represent how test data were recorded at that time.
FIGURE 4. Vertical pattern of the AN/FPS-31 (XD-1) an-
tenna using the sun as a source of RF noise.
16
18
14
12
10
8
6
4
2
0
0
–2
–4
–6
–8
–10
–12
–2 0 2 4 8 10 12 14
07:09 06:39 06:09
(Sunrise)
18 6
L
e
v
e
l

o
f

s
i
g
n
a
l

a
b
o
v
e

t
h
e
r
m
a
l

n
o
i
s
e

(
d
B
)

R
e
l
a
t
i
v
e

d
B
Elevation angle in degrees (sunrise = 0°)
Local time
–14
–16
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 211
FIGURE 7. Aerial view of Boston Hill, North Andover, Mas-
sachusetts, showing the experimental Counter-Counter-
measure (CCM) Radar Mark I and its facilities.
was no hope of eliminating the backscattered signals
by the techniques of moving-target indication. It had
to be mapped out when it occurred.
It had not been generally believed beforehand that
auroral echoes could be observed above 200 MHz.
The AN/FPS-31 detected strong auroral echoes at
425 MHz, and the Sentinel radar, the AN/FPS-30,
did so at 600 MHz. This surprise is reminiscent of
something that happened at the MIT Radiation
Laboratory during World War II. The newly devel-
oped microwave radars at 3-cm wavelength were so
successful that researchers decided to develop systems
at 1.25-cm wavelength, providing finer angular reso-
lution for a given antenna aperture. When they did
so, they discovered that the new radars, which oper-
ated near the peak of the curve of water-vapor absorp-
tion in the atmosphere, had disappointing perfor-
mance. In the Radiation Laboratory incident, the
cause of the problem was obvious by hindsight. The
auroral-backscatter problem was less obvious.
Ultimately, the MITRE Corporation, incorpo-
rated on 21 July 1958, took over responsibility for the
Jug Handle Hill site along with everything else in the
Experimental SAGE Subsector. They closed the site
in November 1962. Figure 6 shows what was left of
this radar in the summer of 1998. The rotating an-
tenna assembly is long gone. The tower still stands,
festooned with assorted communication antennas for
mobile communications and data links. This old
sword has been beaten into a modern plowshare.
Boston Hill, North Andover, Massachusetts
After the UHF GCI radar at Jug Handle Hill became
an operational element of the Experimental SAGE
Subsector, it could no longer be available for the de-
velopment and testing of new radar techniques. Con-
sequently, Lincoln Laboratory undertook to build an
improved version of it, dubbed the Experimental
CCM Radar Mark I. It was installed atop Boston
Hill, west of Route 114 in North Andover, Massachu-
setts. A comprehensive description of the so-called
Boston Hill radar has been published [5].
The aerial view of Boston Hill in Figure 7 shows
the radar and its associated facilities. The reflector of
the rotating radar antenna was 120 ft wide and 30 ft
high. The low building to the right of the radar tower
housed the AN/FST-2 fine-grained-data signal pro-
cessing equipment needed to transform the analog
output signal from the radar receiver into a digital
data stream suitable for transmission to the AN/FSQ-
7 SAGE central computer.
Figure 8 shows the Boston Hill radar. The L-band
FIGURE 6. The remains of the Jug Handle Hill radar, circa
1998. Courtesy of Harold Heggestad and Chester Kurys.
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
212 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
FIGURE 8. UHF GCI/CCM (Boston Hill) radar. The principal
subsystems of the radar are housed on separate floors in
the tower building beneath the antenna. The feed is at the fo-
cus of the parabolic-cylinder antenna reflector. Note the
“hog-trough” identification-friend-or-foe (IFF) antenna atop
the reflector at its center.
IFF antenna is mounted atop the reflector at its cen-
ter. The ball bearing that carries the 55-ton rotating
load is 13.5 ft in diameter. The radiated E-field polar-
ization of the radar is horizontal. Both horizontal and
vertical polarizations can be received for study of the
depolarization characteristics of aircraft, chaff, pre-
cipitation, and aurora. At UHF, linearly polarized sig-
nals reflected from targets such as sounding rockets,
missiles, and satellites at long range within or above
the ionosphere are almost certain to have undergone a
significant amount of Faraday rotation, so polariza-
tion diversity is essential for their best reception.
It was originally planned to build a two-frequency
radar, the feedhorn and reflector serving at both 200
and 400 MHz. The lower frequency was not imple-
mented, however. All subsystems of the radar were
housed within and atop the tower.
This radar was designed to provide broadband op-
eration over the range 400 to 450 MHz [6]. The
transmitter was a klystron amplifier, shown in Figures
9 and 10, rather than a magnetron oscillator. It pro-
duced higher output power and provided more con-
trol over the transmitted waveform. It allowed fully
coherent operation, since the receiver’s local oscilla-
tors were derived from the same frequency source that
powered the transmitter. The first klystrons, VA-
812s, had 2% instantaneous bandwidth. The klystron
vendor, Varian, later produced VA-812B tubes with
12% instantaneous bandwidth.
The antenna was designed to have low sidelobe
levels to minimize the enemy’s ability to conceal air-
craft by sidelobe jamming. Figure 11 shows a full azi-
muth cut of the radar antenna pattern at 430 MHz, at
an elevation angle of about 0°. A central pedestal
about 200° wide has peaks ranging from 19 to 27 dB
below the observed beam peak. This observed beam
peak lies about 4.5° in elevation angle below the true
peak, which had a gain of about 32 dBi [7]. The re-
mainder of the azimuth scan beyond this 200° pedes-
tal is about 37 dB below the observed peak.
The Boston Hill radar was about twenty miles
from the Millstone Hill radar, and no significant ter-
rain obstructions existed between the two facilities. In
that era Millstone Hill was operating near 440 MHz.
It was decided to restrict the broadband operation of
the Boston Hill radar to frequencies suitably distant
from 440 MHz. That measure minimized RF inter-
ference (RFI) to Millstone Hill.
FIGURE 9. Varian VA-812 klystron for the transmitter of the
UHF GCI/CCM radar. Its rating is 8-MW peak power, 28-kW
average power, 2% instantaneous bandwidth. The cathode is
on the left and is operated below ground potential. The an-
ode is on the right along with the cylindrical RF output win-
dow. The klystron measures nearly ten feet long.
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 213
There was justifiable concern about placing a UHF
ground/air communication terminal (an important
part of the SAGE system concept) at the same site as a
large VHF or UHF radar such as the AN/FPS-24 or
AN/FPS-35. The Boston Hill radar provided a good
experimental facility for the investigation of these po-
tential RFI problems.
A number of radar-evaluation, antijam, and CCM
techniques were tested at Boston Hill. We briefly dis-
cuss seven of them.
Determining a Radar’s Detection-Range Capability
Directly measuring the performance of a high-capa-
bility radar against a small airborne target can be diffi-
cult because of horizon effects and the target’s altitude
limitations. Tests involving an F-86 fighter aircraft
had to be run at reduced transmitter power and with
a 16-dB attenuator in the receiver line in order to de-
termine an experimental value for the detection
range. These results were then scaled to the condition
of full transmitter power and no receiver attenuator.
Figure 12 shows the radar’s coverage diagram.
Adapting Sea-Clutter-Cancellation Techniques
Consider a cloud of chaff in an environment of con-
stant-velocity winds, observed by a ground-based ra-
dar. The echo signals from this cloud behave in some
ways like the sea-clutter returns seen by an airborne
radar. The Time-Averaged Clutter-Coherent Air-
borne Radar (TACCAR) AMTI system succeeded in
reducing sea clutter by causing the zero-response
notch of the IF velocity filter to track the radial com-
ponent of sea-surface velocity relative to the airborne
platform (see the article “Displaced-Phase-Center
Antenna Technique,” by Charles Edward Muehe and
Melvin Labitt, in this issue). The sliding-notch IF
canceler (SNIFCAN), developed for the Boston Hill
radar, was an application of the same idea to reduce
echoes from chaff, and it was tested at the Boston Hill
radar.
Chaff-Canceling Techniques
A fully coherent frequency-hopping radar can over-
come the frequency sensitivity of the motions of a
chaff cloud by making the radar echoes noncoherent.
Just after the chaff bundle is dispensed by an enemy
aircraft the echo from it looks like that from a strong
point target, but as time passes the chaff slows down
and disperses in position and in velocity. The echo
from it in each of the radar’s resolution volumes be-
comes weaker and noiselike. The radar’s problem
then becomes that of detecting the echo signal from
aircraft targets immersed in the noisy echoes from the
chaff cloud.
FIGURE 10. Dummy RF load for the transmitter of the UHF
GCI/CCM radar. The transmitter is on the floor below; the
rotary joint and antenna mount are on the floor above. A
waveguide switch allows for operating with the dummy load
(right) or the antenna (upstairs). For scale reference, each
floor tile measures 9 × 9 in.
FIGURE 11. An azimuth cut at an elevation angle of about 0°
through the 430-MHz antenna pattern of the Boston Hill ra-
dar. A central region about 200° wide contains peaks rang-
ing from 19 to 27 dB below the observed beam peak. This ob-
served beam peak lies about 4.5° in elevation angle below
the true peak, which had a gain of about 32 dBi.
0
–70
–60
–50
–40
–30
–20
–10
R
e
l
a
t
i
v
e

s
i
g
n
a
l

l
e
v
e
l

(
d
B
)
Angle (deg)
Peak of beam
–200 –160 –120 –80 –40 0 40 80 120 160 200
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
214 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
The broadband characteristics of the Boston Hill
radar made it practical to demonstrate the efficacy of
pulse-to-pulse frequency hopping in the minimiza-
tion of echoes from distributed targets such as
weather and chaff.
More about Pulse-to-Pulse Frequency Hopping
Pulse-to-pulse frequency hopping has the further ad-
vantage of transforming ground-clutter echoes into
noiselike signals also, unless a particular piece of clut-
ter corresponds to a large physical point target. Of
course, frequency hopping adds complexity to the
radar’s MTI circuitry.
The incorporation of frequency coding in the fre-
quency-hopping pattern made pulse-interval expan-
sion possible at the Boston Hill radar. A target (a mis-
sile, for example) in a distant and specific interpulse
range interval could be detected without its having to
compete with echo signals from targets in other inter-
vals. The addition of instantaneous-frequency-corre-
lation (IFC) constant-false-alarm-rate (CFAR) cir-
cuitry to the radar’s frequency-hopping receiver
greatly reduced or eliminated the echoes from
weather and chaff. The echoes from auroral ioniza-
tion were also reduced, but it was found that pulse-to-
pulse frequency hopping was not necessary; the rela-
tively simple IFC CFAR circuitry sufficed.
Jittered Pulse-Repetition Frequency
The Boston Hill radar was capable of jittering its
pulse-repetition frequency (PRF). That CCM tech-
nique can be employed to prevent a pulse-repeater
jammer, carried by an aircraft, from jamming echoes
from targets that are closer to the radar than it is.
Outside that range the repeated pulse signals fall into
the same range box as the authentic signals when re-
ceived at the radar. Inside that range they fall into
randomly distributed range boxes, depending on how
the PRF jitter is programmed. They do not simulate
echoes from a nonexistent aircraft, so they cause less
confusion to the radar signal processing circuitry.
Sidelobe Cancellation, Jammer-Strobing Systems
A system installed in the Boston Hill radar could in-
dicate unambiguously the azimuth (or “strobe”) of a
jammer, even when the latter was within its self-
screening region. The operating principle involved
comparing the signal received by an omnidirectional
antenna with the signal received by the main radar
antenna. The output of the system was a PPI strobe,
of angular width roughly equal to the antenna beam-
width at its sidelobe level, pointing directly toward
the azimuth of the jammer. This system was an out-
growth of Lincoln Laboratory’s Project Cross Over. It
was satisfactorily tested in the course of several U.S.
Air Force jamming exercises.
Another jammer-strobing method was developed
in the course of Lincoln Laboratory’s program to de-
velop electronic counter-countermeasures for AEW
radars. This method requires only a Clark/Dicke-Fix
IF channel in the radar receiver and provisions for in-
serting pulsed RF signals ahead of it. For a discussion
FIGURE 12. Boston Hill UHF radar coverage diagram on an
F-86 fighter aircraft for 50% blip-scan ratio (the radar opera-
tor on average sees the blip every other scan). The radio ho-
rizon is 4/3 the radius of the earth.
Slant range (nmi)
H
e
i
g
h
t

(
1
0
3

f
e
e
t
)
Radar

0.5°

3
°
4
°
5
°
6
°
7
°
8
° 9
°
1
0
°
1
2
°
1
5
°
2
0
°
6
0
° E
le
v
a
tio
n
a
n
g
l
e
(
d
e
g
)

3
0
°
80 160 240 320 400 480 0
0
40
80
120
160
200
240
280
320
360
400
440
480
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
VOLUME 12, NUMBER 2, 2000 LINCOLN LABORATORY JOURNAL 215
of the Dicke Fix, a counter-countermeasure, see the
article entitled “Early Advances in Radar Technology
for Aircraft Detection,” by Donald L. Clark, in this
issue. Comparison of the two jammer-strobing meth-
ods revealed essentially the same basic limitations for
both.
Observing Objects in Space
An interesting experiment was carried out on 29 Oc-
tober 1959, when NASA launched a 100-ft-diameter
metallized-plastic balloon called Shotput 1 on a
sounding rocket from Wallops Island, Virginia, some
400 to 500 miles south of Boston Hill. The balloon
was inflated after launch. This preliminary test was
followed by the successful launch to orbit of the Echo
1 balloon from Cape Canaveral, Florida, on 12 Au-
gust 1960. In the Shotput 1 test, the balloon rose to an
elevation angle of about 25°, as seen from the radar,
and it could be seen with the naked eye. The echoes
from it were strong. The signals dropped out during
the higher-altitude portions of the balloon’s flight,
probably because it had then risen above the main
lobe of the radar’s antenna pattern. The signals reap-
peared a few minutes later, when the balloon fell back
into the antenna beam. The Boston Hill radar also
supported NASA’s Shotput 2 test on 16 January 1960.
This radar was not well suited to the observation of
orbiting satellites; attempts to detect them were un-
successful.
The Boston Hill radar reached its full operating
capability in late 1959, just about the time when
Lincoln Laboratory changed the thrust of its radar
programs. Although by no means had all of the prob-
lems presented by airborne threats been solved,
Lincoln Laboratory’s efforts in radar research and de-
velopment were to be concentrated on ballistic mis-
sile threats until the late 1960s. At that time Lincoln
Laboratory began its FAA-sponsored program in air
traffic control. Also at that time there was resurgence
of interest in tactical radar applications, engendered
by the Vietnam War. These two disparate influences
led to the broad range of radar technology that is
chronicled in other articles in this issue of the Lincoln
Laboratory Journal.
On 1 April 1960 responsibility for the Boston Hill
radar was transferred to the MITRE Corporation,
just as was done earlier for the Jug Handle Hill radar.
Several years later the antenna of the Boston Hill ra-
dar was demounted and used to replace an AN/FPS-
35 antenna that had been damaged by high winds at
an operational Air Force site. One of the authors, vis-
iting Boston Hill in the early 1990s, found the tower
still standing. It, like the tower of the Jug Handle Hill
radar (Figure 6), had become an antenna farm.
Acknowledgments
Much of the information for the sidebar was drawn
from “History of the Rome Air Development Cen-
ter,” Griffiss Air Force Base, New York, 1 January to
31 December 1959, Document 60-14627, pp. 51–
55. This document and others that were graciously
provided by Thomas W. Thompson, Chief Historian
at the Rome Research Site, New York, were invalu-
able. Frank Mastrovita, archivist at the Bedford, Mas-
sachusetts, operation of the MITRE Corp., helped
the authors follow the trails of the Jug Handle Hill
and Boston Hill radars after they were transferred
from Lincoln Laboratory to MITRE. It is a pleasure
to acknowledge their generous assistance.
REF ERENCES
1. J.R. Zacharias and A.G. Hill, Final Report of 1952 Summer
Study Group, vol. 2, Lincoln Laboratory (10 Feb. 1953).
2. “Five Years at the Radiation Laboratory,” 1991 IEEE MTT-S
Int. Microwave Symp., Boston, 10–14 June 1991 (originally
published 1946).
3. R. Buderi, The Invention That Changed the World: How a Small
Group of Radar Pioneers Won the Second World War and
Launched a Technological Revolution (Simon & Schuster, New
York, 1996).
4. J. Aarons, “Antenna and Receiver Measurements by Solar and
Cosmic Noise,” Proc. IRE 42 (5), 1954, pp. 810–815.
5. P.A. Northrop, “Boston Hill UHF Radar,” Technical Report
243, Lincoln Laboratory ( 6 July 1961), DTIC #AD-325634.
6. The Jug Handle Hill radar was narrowband but tunable over
the same frequency range.
7. The nose of the antenna beam was designed to be about 4.5°
above the local horizontal at the radar, to improve its coverage
of airborne targets.
• WARD AND NAKA
Long-Range UHF Radars for Ground Control of Airborne Interceptors
216 LINCOLN LABORATORY JOURNAL VOLUME 12, NUMBER 2, 2000
wiiiiax w. wann
was born in Texas in 1924.
During World War II, he
served in the U.S. Army Signal
Corps, where he installed,
maintained, and repaired
cryptographic equipment in
the Pacific Theater of Opera-
tions. He received a B.S.
degree from Texas A&M
College, and M.S. and Ph.D.
degrees from California Insti-
tute of Technology, all in
electrical engineering. In 1952,
he joined Lincoln Laboratory,
where his first thirteen years
were devoted to radar system
engineering, including air-
borne-early-warning and
ground-based surveillance
radars, and space tracking and
range instrumentation for
NASA’s Project Mercury and
for ballistic missile testing. In
1965 he switched from strug-
gling to solve problems that
involve (range)
–4
to working on
more tractable problems involving
(range)
–2
. That work has been
in space communication,
primarily in the development
of systems that serve the di-
verse needs of the military and
civil user communities by
means of reliable links through
satellites. He has helped to
design, build, test, and operate
in orbit Lincoln Experimental
Satellites 5, 6, 8, 9, and two
EHF Packages carried by host
satellites FLTSATs 7 and 8. He
has also contributed to the
development of the operations
centers associated with these
satellites. Being blessed with a
retentive memory, and having
the collecting habits of a pack
rat, he helped to prepare MIT
Lincoln Laboratory: Technology
in the National Interest, an
illustrated history of the Labo-
ratory published in 1995. He
retired from Lincoln Labora-
tory in 1994 after long service
as manager of Satellite Opera-
tions (“Keeper of Old Satel-
lites”). He now putters around
with a few old satellites that
refuse to die, consults, writes,
lectures, and raises vegetables
in the summertime. He is a
registered professional
engineer in Massachusetts, a
member of several professional
societies, and currently a
Distinguished Lecturer for the
IEEE Aerospace and Elec-
tronic Systems Society.
v. nonrnr xa×a
joined Project Lincoln in June
1951 after completing his
Doctor of Science degree
(electron optics) at Harvard
University. In 1954 Dr. Naka
became associate leader of the
Special Radars group. In 1956
he became the leader of the
Heavy Radars group, where he
led the development of the
Boston Hill radar and was a
member of the Air Force’s
Frequency Diversity Advisory
group. In 1959 Bob joined the
MITRE Corporation, which
had been established a year
earlier to head the Radar
Systems and Techniques de-
partment. Subsequently he
became associate technical
director, then technical direc-
tor of MITRE’s Applied Sci-
ence Laboratories. In 1969
Bob became chief scientist of
the MITRE Corporation.
That same year, he reported to
the Pentagon to become
deputy director of the Na-
tional Reconnaissance Office.
In 1972 he joined Raytheon to
be director of Detection and
Instrumentation Systems, and
in 1975 returned to the Penta-
gon to become Air Force chief
scientist. In 1978 he joined
Science Applications, Inc. as
corporate vice president, and
in 1982 he joined GTE Gov-
ernment Systems Corporation
as vice president, Engineering
and Planning, from which he
retired in 1988. He now runs a
small business, CERA, Inc.,
specializing in electromagnetic
technology, as its president. He
is a registered professional
engineer in Massachusetts.
Bob’s honors include Member
of the National Academy of
Engineering, Fellow of the
Explorers Club, and member
of the honorary societies
Sigma Xi, Tau Beta Pi, and the
Druids (Omicron Delta
Kappa) of the University of
Missouri. He has received the
U.S. Air Force’s Exceptional
Service Award four times. The
University of Missouri be-
stowed an Honor Award for
Engineering and the Faculty
Alumni Award.

an unanticipated vulnerability of radar operations. For example. independent of the wavelength at which the radar operates. which are discussed in the article entitled “DisplacedPhase-Center Antenna Technique.” MIT’s Radiation Laboratory [2. Keeping the vertical dimension of the antenna about the same meant that the vertical beamwidth was broader so that the coverage in elevation angle extended to correspondingly—and gratifyingly—higher altitudes. The resolution in range could be preserved by using transmitted pulses of the same length as before. Of course.” by Charles Edward Muehe and Melvin Labitt. on all sides researchers began to invent measures to counter or reduce the effectiveness of their opponents’ radars. 2000 equipment to work properly in the field. A new generation of GCI radars needed the flexibility to accommodate emerging CCM techniques to operate in assorted frequency bands. radar developers such as those at the “RadLab. The standard Air Force GCI radars that were available in the early 1950s offered little in the way of CCM capability.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors warning (AEW) radar systems with airborne-movingtarget-indication (AMTI) capability [1]. were initially unaware that they were vulnerable to detection by airborne radars. For example. In the early years of World War II. Thus using longer wavelengths would increase the effectiveness of a pulsed radar by increasing the horizontal aperture of the rotating antenna. were elated just to get their 206 LINCOLN LABORATORY JOURNAL VOLUME 12. Other benefits were associated with the move to longer wavelengths. Lincoln Laboratory’s contributions to this effort were made at a practical level. Longer wavelengths also facilitated efforts to resist jamming. Two different large UHF GCI radars were developed and put into operation at field sites in New England. An incoming bomber force facing an array of frequency-diverse air-defense radars needed to carry an equally diverse collection of active and passive countermeasures. They had enough problems to solve without also considering that an enemy might try to jam the radars’ operation with electronic countermeasures. Engineers realized by this time that—all other things being equal—the effectiveness of a pulsed radar in searching for targets at unknown positions throughout a given volume is proportional to the product of the average power of its transmitter and the aperture area of its receiving antenna. the Axis forces. It therefore became necessary for the radar developers to devise countercountermeasures (CCMs) for their equipment. 3]. Jug Handle Hill. The success of the AEW program led radar designers to believe that operating GCI radars at longer wavelengths could solve many of the problems that radars experienced at higher-frequency Land S-bands. echoes from precipitation and birds were reduced because the scatterers were smaller in terms of wavelength. the horizontal aperture of the rotating radar antenna needed to be wider in proportion to the wavelength ratio to maintain the same resolution in azimuth. Just as the radar developers did not initially anticipate jamming techniques. adding to the complexity of the aircraft and reducing their combined useful payload of bombs. NUMBER 2. . Increasing the size of the RF components would reduce the likelihood of breakdown within them because of high electromagnetic-field strengths. The sidebar entitled “The Air Force Frequency-Diversity Radar Program” describes how these requirements were incorporated in new radars. With substantial electronic jamming of radio transmissions from the western world already in effect. Soon enough. These radars demonstrated in flight tests some of the advantages of operating at lower frequencies (longer wavelengths). Maine In 1954 Lincoln Laboratory undertook to cobble together a demonstration UHF GCI radar in a hurry. They also required the ability to burn through wideband noise jamming and provide coverage on airborne targets of interest out to a useful extended range. West Bath. especially submarine crews. the Radio Research Laboratory [3] at Harvard University worked hard to develop active electronic jamming and passive countermeasures (dropping chaff ) to interfere with radars like the ones under development about a mile down the street at the MIT Radiation Laboratory. Transmitters with higher peak and average powers at longer wavelengths would be easier to build than those at shorter wavelengths because the physical dimensions of the radio-frequency (RF) components would be larger. in this issue. there was no doubt that the Soviet Union would employ radar countermeasures.

Its characteristics were spelled out by midyear. The wise use of the many features of a highly flexible FD radar required special skills and sophisticated technological support. lineal descendants of radars developed during World War II. let design-study contracts for six new ground control of intercepts (GCI) radars. 2000 LINCOLN LABORATORY JOURNAL 207 . and signal repeaters). in particular its ability to feed highquality data to the Semi-Automatic Ground Environment (SAGE) air-defense system. These five systems in their prototype forms were installed for testing and evaluation at operational Air Force sites in Alabama. was not expected to be a great construction challenge. each to operate in a segment of the frequency range 214 to 5900 MHz. CCM Radar Mark I. Griffiss Air Force Base. At that time. the AN/GLA-8 signal processing system. Air-Defense Sector. part of the Mobile. hence its formal name. then a standard Air Force heavy radar for fixed GCI instal- lations. and four were produced in quantity. built by Airborne Instrument Laboratory. In addition. These two radars. Alabama. CCMs such as frequency hopping and PRF jitter/stagger are useful in reducing the effectiveness of both passive countermeasures (chaff.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors THE AIR FORCE FREQUENCY-DIVERSITY RADAR PROGRAM    Rome Air Development Center. was intended to serve as Lincoln Laboratory’s long-term test bed for development and evaluation of CCM techniques. As discussed in the main text. Its mechanical tolerances in terms of wavelength were no more stringent than those of the ≈1300-MHz AN/FPS-3. Without question. Five of the six proposed radars were selected for prototype development. Table 1. 120 ft wide by 16 ft high. The Boston Hill radar. and Mississippi. The antenna. the new program would enhance the Air Force’s GCI capabilities. for example) and active countermeasures (spot and noise jamming. New York.5°-wide radar beams and blips on their PPI displays. At the same time. Four AN/FPS-35s were procured by Rome Air Development Center under a prototype contract for early installation at field sites and operation by Air Force crews. The Jug Handle Hill radar began operation in October 1955. was contemporary with the AN/FPS-35 but had quite a different design. Louisiana. They should not be confused. The spread of operating frequencies to be provided by the FD radar program promised to make it more costly in terms of payload for an airborne intruder to penetrate and survive in the defensive radar environment. Both yielded approximately 1. and the AN/FPS-6 height-finding radar operated at 2900 MHz. which began operation in 1959. The frequency-diversity (FD) radar program was to reverse that trend. on the other hand. NUMBER 2. Rotating the UHF GCI radar antenna at 6 rpm would make its data-output characteristics essentially the same as those of the AN/FPS-3. constituted what amounted to a single-frequency air-defense radar system. which had a 40-ft-wide by 16-ft-high antenna reflector. Their test programs began in 1959. Boston Hill. shows the characteristics of the Air Force frequency-diversity radars. This equipment included a special antijamming console used by the radar’s human countercountermeasures (CCMs) operator. the measure of success it achieved despite the bearing problems of its gargantuan rotating antenna paved the way for the three lower-frequency FD radars. The new radar promised to VOLUME 12. Air Force GCI radars were moving through attrition toward occupancy of only two frequency bands: the AN/ FPS-7 surveillance and heightfinding radar with stacked beams operated at 1300 MHz. as discussed in the main text. on the following page. was an important common adjunct to each frequency-diversity radar.

the transmitter was not very powerful. Clark. Work on this UHF GCI radar. Any magnetron has a random start-up phase on each pulse. A site near the Maine coast on Jug Handle Hill. Furthermore. NUMBER 2.. ultimately desig208 LINCOLN LABORATORY JOURNAL VOLUME 12. Massachusetts. The design of the electronics for the new radar was taken directly from that of the AN/APS-70. Adopting the magnetron-based transmitter design of the AN/APS-70 was a compromise decision. got under way in the fall of 1954. The moving-target-indicator (MTI) circuitry of the UHF GCI radar was simpler. but no suitable tube was immediately available. which was then undergoing development and testing (see the article entitled “Displaced-Phase-Center Antenna Technique.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors Table 1. Figure 1 shows the transmitter and QK508 magnetron for the AN/FPS-31 (XD-1) radar. The high-power RF output pulses travel from the magnetron through a section of 3-1/8-in flexible coaxial transmission line (at the left) to the vertical waveguide run (see also Figures 2 and 3). in this issue). and Montauk Point on Long Island. Although the transmitter and its associated circuitry could be readily copied from the AEW-radar equipment. the 2% instantaneous RF bandwidth of the magnetron could not accommodate the rapid changes of transmitted frequency—perhaps even pulse to pulse—required for some radar CCM techniques. New York. 2000 nated the AN/FPS-31 (XD-1). making the radar’s signal processing equipment more complicated. Characteristics of the Air Force Frequency-Diversity Radars Function Surveillance Surveillance Surveillance Frequency Range (MHz) 214–236 400–450 510–690 2320–2680 5400–5900 Equipment Designator AN/FPS-24 AN/FPS-35 AN/FPS-28* AN/FPS-27 AN/FPS-26 Contractor GE Sperry Raytheon Westinghouse AVCO Manufacturing Co. Lincoln Laboratory’s UHF AEW radar with AMTI. The radar began operation in October Windlass Flexible coaxial transmission line FIGURE 1. tetrode. in this issue). Crosley Division Surveillance and height finding (stacked beams) Height finding (nodding beam) *Not produced in quantity be a prime input sensor for the Cape Cod System and later for the Experimental Semi-Automatic Ground Environment (SAGE) Subsector. or amplitron. . that were already integrated into the Cape Cod System. We would have preferred to build a fully coherent radar of the master-oscillator/poweramplifier family (see the article “Early Advances in Radar Technology for Aircraft Detection.” by Donald L. was selected to serve as a counterpart to the shoreline GCI radars at South Truro on Cape Cod. Note the windlass (essential when changing magnetrons) and the arrangements for liquid cooling of the magnetron. West Bath. That approach would have required using a high-power amplifier such as a triode. The QK-508 magnetron (middle) fits into a pulse transformer and is powered by the modulator cabinet on the right.” by Charles Edward Muehe and Melvin Labitt. Transmitter of the AN/FPS-31 (XD-1) UHF ground control of intercepts (GCI) radar. klystron.

installation. Massachusetts. The feed (in the center at the top of the picture) is at the focus of the parabolic-cylinder antenna reflector. painted with broad vertical white and international-orange stripes. the radar receiver had to be tuned to match the frequency of the magnetron. and MTI signal processing circuitry for the AN/FPS-31 radar was a straightforward task. Further design studies and tests showed that this bearing arrangement could be perfected. receiving. All of these disruptions had to be mitigated. Of course. however. The fabrication. There was a shutdown of several months while the bearing was reVOLUME 12. 1955.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors FIGURE 3. and operation of a full suite of transmitting. There was pulsed RF interference from AEW aircraft—carrying experimental UHF AEW radars built by Lincoln Laboratory—when they operated within line of sight of the AN/FPS-31. Note the stairway for scale. which was just seaward of U. Ignition noise from vehicles driving by the radar site. An arrangement for “one-knob” tuning control of the complete radar was developed.S. West Bath. the pressure from the SAGE development schedule to get the AN/FPS-31 radar into full operation speedily led to the decision to abandon the original design and go to a large central ball bearing upon which the entire rotating assembly would ride. Another view of the tower and antenna assembly FIGURE 2. This installation gave trouble from the start. Hill. The track had not been made sufficiently smooth to begin with. However. carried by military aircraft passing through the radar’s coverage volume. This modification proved to have its own problems. The 120-ft-wide antenna shown in Figures 2 and 3. There were some interesting aspects to it. The original design called for the heavy rotating mass to be carried on sets of bogie wheels at the ends of a three-armed spider that rolled on a smooth. Maine. as shown in Figures 2 and 3. later replaced by an AN/FST-2 finegrained-data system. A dual-channel AN/FST-1 Slowed-Down-Video (SDV) system. NUMBER 2. The Jug Handle Hill site was ultimately equipped with two standard Air Force AN/FPS-6 S-band nodding-beam height finders to give it full GCI capability. The unique subsystem in the AN/FPS-31 radar was its large antenna. jammed the radar. caused interference. was an impressive sight when rotating at 6 rpm. The AN/FPS-31 (XD-1) UHF radar at Jug Handle of the AN/FPS-31 (XD-1) UHF GCI radar. 2000 LINCOLN LABORATORY JOURNAL 209 . Route 1. together with the tower to support it and the bearing arrangements and drive machinery to rotate it in azimuth. was installed at the site to relay data from the three radars and the Mark X identification-friend-or-foe (IFF) equipment to the Experimental SAGE Subsector’s central computer at Lexington. The radar interference was eliminated by replacing the fixed-tuned magnetron in the transmitter of Figure 1 by a tunable one. AN/APN-1 FM radar altimeters. and the wheels soon wore out. level circular track at the top of the tower. The vertical waveguide runs from the equipment building to the rotary-joint housing beneath the antenna atop the tower. The bearing caused many headaches.

Improper orientation of the feedhorn proved to be the source of the problem. The clock face and grease-pencil notes on the white tablet represent how test data were recorded at that time. shown in Figure 5. the AN/FPS-49 Ballistic Missile Early Warning System (BMEWS) tracking radars. In April 1956 the AN/FPS-31 radar was found to 210 LINCOLN LABORATORY JOURNAL VOLUME 12. The peak of the approximately 18°-vertical-width main beam was 8° above the horizon. This point was proved convincingly with the aid of antenna patterns measured at sunrise and at sunset as the rotation of the earth moved the antenna beam across the disk of the sun.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors 18 16 14 12 10 8 6 4 2 0 –2 0 2 4 6 8 10 12 14 18 0 –2 –6 –8 –10 –12 –14 –16 Elevation angle in degrees (sunrise = 0°) 06:09 (Sunrise) 06:39 07:09 Local time FIGURE 4. NUMBER 2. but they had unusual characteristics: high scatterer velocities. These mechanical problems were eventually solved to achieve reliable operation of the large rotating antenna assembly. Consultation with personnel from the Communications and Components division yielded the suggestion that the AN/ FPS-31 radar was receiving echoes from the aurora borealis. Vertical pattern of the AN/FPS-31 (XD-1) an- tenna using the sun as a source of RF noise. and consistent occurrence in the same general azimuth direction—magnetic north. This auroral clutter could overlie any part of the radar’s unambiguous range. For best coverage. The velocity distribution of the ionized particles comprising the aurora was so broad that there Level of signal above thermal noise (dB) FIGURE 5. Echoes resembling returns from storms were observed. putting the peak 4° above it. Maine. the Millstone Hill radar. This surmise was verified when it was possible to correlate these 425-MHz observations in Maine with those from a 50-MHz radar located at Ottawa. Correlation of the radar data with the occurrence of solar flares and sudden ionospheric disturbances led to the conclusion that auroral clutter showed up on the AN/FPS-31 radar about 48 hours after a solar flare. Despite the rare occurrence of auroral activity in New England skies. Canada. The experience that Lincoln Laboratory gained in solving such problems was shared with others and led to subsequent successful designs of the Counter-Countermeasure (CCM) Radar Mark I at Boston Hill. The range of the echoes was seen the “second time around. .” The distance between range marks is 50 miles. as shown in Figure 4 [4]. sharply defined azimuth boundaries. it did not meet expectations established by scaling from the demonstrated performance of UHF AEW radars operating at lower power and with smaller antennas. Massachusetts. A new feedhorn was ultimately procured and installed. Auroral echoes on the AN/FPS-31 (XD-1) radar at Bath. the 3-dB-down point of the vertical lobe should have been on the horizon. and other radars. 2000 Relative dB –4 display clutter of an unexpected sort. Although the performance of the AN/FPS-31 radar was impressive. worked. with gratifying results. the AN/FPS-31 radar was powerful enough to produce pulse echoes that backscattered from the actual aurora (high above the atmosphere and far to the north) and reached the radar at the same time as did echoes from later pulses returned by the much closer targets of interest.

The low building to the right of the radar tower housed the AN/FST-2 fine-grained-data signal processing equipment needed to transform the analog output signal from the radar receiver into a digital data stream suitable for transmission to the AN/FSQ7 SAGE central computer.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors Ultimately. This surprise is reminiscent of something that happened at the MIT Radiation Laboratory during World War II. took over responsibility for the Jug Handle Hill site along with everything else in the Experimental SAGE Subsector. A comprehensive description of the so-called Boston Hill radar has been published [5]. NUMBER 2. It was installed atop Boston Hill. The AN/FPS-31 detected strong auroral echoes at 425 MHz. North Andover. and the Sentinel radar. FIGURE 7. This old sword has been beaten into a modern plowshare. The L-band FIGURE 6. The rotating antenna assembly is long gone. VOLUME 12. was no hope of eliminating the backscattered signals by the techniques of moving-target indication. did so at 600 MHz.25-cm wavelength. Figure 6 shows what was left of this radar in the summer of 1998. The newly developed microwave radars at 3-cm wavelength were so successful that researchers decided to develop systems at 1. The reflector of the rotating radar antenna was 120 ft wide and 30 ft high. The auroral-backscatter problem was less obvious. The tower still stands. Boston Hill. had disappointing performance. It had not been generally believed beforehand that auroral echoes could be observed above 200 MHz. Consequently. It had to be mapped out when it occurred. incorporated on 21 July 1958. providing finer angular resolution for a given antenna aperture. Figure 8 shows the Boston Hill radar. the MITRE Corporation. circa 1998. Courtesy of Harold Heggestad and Chester Kurys. it could no longer be available for the development and testing of new radar techniques. 2000 LINCOLN LABORATORY JOURNAL 211 . When they did so. festooned with assorted communication antennas for mobile communications and data links. Lincoln Laboratory undertook to build an improved version of it. west of Route 114 in North Andover. The aerial view of Boston Hill in Figure 7 shows the radar and its associated facilities. the AN/FPS-30. the cause of the problem was obvious by hindsight. The remains of the Jug Handle Hill radar. North Andover. dubbed the Experimental CCM Radar Mark I. they discovered that the new radars. Aerial view of Boston Hill. showing the experimental Counter-Countermeasure (CCM) Radar Mark I and its facilities. They closed the site in November 1962. which operated near the peak of the curve of water-vapor absorption in the atmosphere. Massachusetts After the UHF GCI radar at Jug Handle Hill became an operational element of the Experimental SAGE Subsector. Massachusetts. In the Radiation Laboratory incident. Massachusetts.

chaff. It was decided to restrict the broadband operation of the Boston Hill radar to frequencies suitably distant from 440 MHz.5° in elevation angle below the true peak. 28-kW average power. at an elevation angle of about 0°. The cathode is on the left and is operated below ground potential. The first klystrons.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors IFF antenna is mounted atop the reflector at its center. Varian VA-812 klystron for the transmitter of the UHF GCI/CCM radar. rather than a magnetron oscillator. The klystron measures nearly ten feet long. The remainder of the azimuth scan beyond this 200° pedestal is about 37 dB below the observed peak. It was originally planned to build a two-frequency radar. The klystron vendor. NUMBER 2. which had a gain of about 32 dBi [7]. and satellites at long range within or above the ionosphere are almost certain to have undergone a significant amount of Faraday rotation. FIGURE 9. At UHF. It produced higher output power and provided more control over the transmitted waveform. Both horizontal and vertical polarizations can be received for study of the depolarization characteristics of aircraft. FIGURE 8. The lower frequency was not implemented. UHF GCI/CCM (Boston Hill) radar. linearly polarized signals reflected from targets such as sounding rockets. All subsystems of the radar were housed within and atop the tower. The ball bearing that carries the 55-ton rotating load is 13. later produced VA-812B tubes with 12% instantaneous bandwidth. so polarization diversity is essential for their best reception. In that era Millstone Hill was operating near 440 MHz. This observed beam peak lies about 4. Its rating is 8-MW peak power. since the receiver’s local oscillators were derived from the same frequency source that powered the transmitter. A central pedestal about 200° wide has peaks ranging from 19 to 27 dB below the observed beam peak. the feedhorn and reflector serving at both 200 and 400 MHz. That measure minimized RF interference (RFI) to Millstone Hill. The principal subsystems of the radar are housed on separate floors in the tower building beneath the antenna.5 ft in diameter. The radiated E-field polarization of the radar is horizontal. 2000 . 2% instantaneous bandwidth. missiles. It allowed fully coherent operation. shown in Figures 9 and 10. The transmitter was a klystron amplifier. The feed is at the focus of the parabolic-cylinder antenna reflector. The anode is on the right along with the cylindrical RF output window. The antenna was designed to have low sidelobe levels to minimize the enemy’s ability to conceal aircraft by sidelobe jamming. and no significant terrain obstructions existed between the two facilities. This radar was designed to provide broadband operation over the range 400 to 450 MHz [6]. Figure 11 shows a full azimuth cut of the radar antenna pattern at 430 MHz. Note the “hog-trough” identification-friend-or-foe (IFF) antenna atop the reflector at its center. had 2% instantaneous bandwidth. VA812s. 212 LINCOLN LABORATORY JOURNAL VOLUME 12. Varian. precipitation. The Boston Hill radar was about twenty miles from the Millstone Hill radar. however. and aurora.

NUMBER 2. and it was tested at the Boston Hill radar. developed for the Boston Hill radar. A waveguide switch allows for operating with the dummy load (right) or the antenna (upstairs). For scale reference. the rotary joint and antenna mount are on the floor above.” by Charles Edward Muehe and Melvin Labitt. A central region about 200° wide contains peaks ranging from 19 to 27 dB below the observed beam peak. Just after the chaff bundle is dispensed by an enemy aircraft the echo from it looks like that from a strong point target. The sliding-notch IF canceler (SNIFCAN). but as time passes the chaff slows down and disperses in position and in velocity. and CCM techniques were tested at Boston Hill. There was justifiable concern about placing a UHF ground/air communication terminal (an important part of the SAGE system concept) at the same site as a large VHF or UHF radar such as the AN/FPS-24 or AN/FPS-35. The transmitter is on the floor below. Adapting Sea-Clutter-Cancellation Techniques Consider a cloud of chaff in an environment of constant-velocity winds. Determining a Radar’s Detection-Range Capability Directly measuring the performance of a high-capability radar against a small airborne target can be difficult because of horizon effects and the target’s altitude limitations. each floor tile measures 9 × 9 in. antijam. Tests involving an F-86 fighter aircraft had to be run at reduced transmitter power and with a 16-dB attenuator in the receiver line in order to determine an experimental value for the detection range. We briefly discuss seven of them. An azimuth cut at an elevation angle of about 0° through the 430-MHz antenna pattern of the Boston Hill radar. Dummy RF load for the transmitter of the UHF GCI/CCM radar. FIGURE 11. which had a gain of about 32 dBi. Chaff-Canceling Techniques A fully coherent frequency-hopping radar can overcome the frequency sensitivity of the motions of a chaff cloud by making the radar echoes noncoherent. was an application of the same idea to reduce echoes from chaff. Figure 12 shows the radar’s coverage diagram.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors Relative signal level (dB) 0 –10 –20 –30 –40 –50 –60 –70 Peak of beam –200 –160 –120 –80 –40 0 40 80 120 160 200 Angle (deg) FIGURE 10. The Boston Hill radar provided a good experimental facility for the investigation of these potential RFI problems. in this issue). This observed beam peak lies about 4. The Time-Averaged Clutter-Coherent Airborne Radar (TACCAR) AMTI system succeeded in reducing sea clutter by causing the zero-response notch of the IF velocity filter to track the radial component of sea-surface velocity relative to the airborne platform (see the article “Displaced-Phase-Center Antenna Technique. VOLUME 12. 2000 LINCOLN LABORATORY JOURNAL 213 . The echo from it in each of the radar’s resolution volumes becomes weaker and noiselike. The echo signals from this cloud behave in some ways like the sea-clutter returns seen by an airborne radar. The radar’s problem then becomes that of detecting the echo signal from aircraft targets immersed in the noisy echoes from the chaff cloud.5° in elevation angle below the true peak. These results were then scaled to the condition of full transmitter power and no receiver attenuator. A number of radar-evaluation. observed by a ground-based radar.

Of course. For a discussion 5° 4° 3° 2° 1° 0.S. from jamming echoes from targets that are closer to the radar than it is. for example) in a distant and specific interpulse range interval could be detected without its having to compete with echo signals from targets in other intervals. Air Force jamming exercises. The broadband characteristics of the Boston Hill radar made it practical to demonstrate the efficacy of pulse-to-pulse frequency hopping in the minimization of echoes from distributed targets such as weather and chaff. The operating principle involved comparing the signal received by an omnidirectional antenna with the signal received by the main radar antenna.5° Radar 0° 0 80 160 240 320 400 480 Slant range (nmi) FIGURE 12. pointing directly toward the azimuth of the jammer. 2000 . That CCM technique can be employed to prevent a pulse-repeater jammer. Another jammer-strobing method was developed in the course of Lincoln Laboratory’s program to develop electronic counter-countermeasures for AEW radars. They do not simulate echoes from a nonexistent aircraft. This system was an outgrowth of Lincoln Laboratory’s Project Cross Over. A target (a mis214 LINCOLN LABORATORY JOURNAL VOLUME 12. Inside that range they fall into randomly distributed range boxes. frequency hopping adds complexity to the radar’s MTI circuitry. the relatively simple IFC CFAR circuitry sufficed.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors 480 440 400 360 320 Height (103 feet) 280 240 12° 10 ° 9° 8° 7° sile. even when the latter was within its selfscreening region. The incorporation of frequency coding in the frequency-hopping pattern made pulse-interval expansion possible at the Boston Hill radar. Sidelobe Cancellation. The addition of instantaneous-frequency-correlation (IFC) constant-false-alarm-rate (CFAR) circuitry to the radar’s frequency-hopping receiver greatly reduced or eliminated the echoes from weather and chaff. Elev atio na ng le ( de g) 6° 60° 30° 20° 15° 200 160 120 80 40 0 Jittered Pulse-Repetition Frequency The Boston Hill radar was capable of jittering its pulse-repetition frequency (PRF). but it was found that pulse-topulse frequency hopping was not necessary. This method requires only a Clark/Dicke-Fix IF channel in the radar receiver and provisions for inserting pulsed RF signals ahead of it. carried by an aircraft. The radio horizon is 4/3 the radius of the earth. More about Pulse-to-Pulse Frequency Hopping Pulse-to-pulse frequency hopping has the further advantage of transforming ground-clutter echoes into noiselike signals also. unless a particular piece of clutter corresponds to a large physical point target. Boston Hill UHF radar coverage diagram on an F-86 fighter aircraft for 50% blip-scan ratio (the radar operator on average sees the blip every other scan). depending on how the PRF jitter is programmed. so they cause less confusion to the radar signal processing circuitry. Jammer-Strobing Systems A system installed in the Boston Hill radar could indicate unambiguously the azimuth (or “strobe”) of a jammer. It was satisfactorily tested in the course of several U. The output of the system was a PPI strobe. Outside that range the repeated pulse signals fall into the same range box as the authentic signals when received at the radar. The echoes from auroral ionization were also reduced. of angular width roughly equal to the antenna beamwidth at its sidelobe level. NUMBER 2.

The echoes from it were strong. some 400 to 500 miles south of Boston Hill. Massachusetts. J. a counter-countermeasure. engendered by the Vietnam War. 1954. Thompson. Clark. J. The signals dropped out during the higher-altitude portions of the balloon’s flight. Although by no means had all of the problems presented by airborne threats been solved. 1953). 810–815. Virginia.A.G. The signals reappeared a few minutes later. archivist at the Bedford. Comparison of the two jammer-strobing methods revealed essentially the same basic limitations for both. just as was done earlier for the Jug Handle Hill radar. had become an antenna farm. The nose of the antenna beam was designed to be about 4.” 1991 IEEE MTT-S Int. Hill.” Griffiss Air Force Base. Acknowledgments Much of the information for the sidebar was drawn from “History of the Rome Air Development Center. IRE 42 (5). Zacharias and A. vol. Florida. 2. Observing Objects in Space An interesting experiment was carried out on 29 October 1959. 51– 55. visiting Boston Hill in the early 1990s. Lincoln Laboratory’s efforts in radar research and development were to be concentrated on ballistic missile threats until the late 1960s. “Boston Hill UHF Radar. It is a pleasure to acknowledge their generous assistance.” Proc.5° above the local horizontal at the radar. R. 6. when NASA launched a 100-ft-diameter metallized-plastic balloon called Shotput 1 on a sounding rocket from Wallops Island. The Invention That Changed the World: How a Small Group of Radar Pioneers Won the Second World War and Launched a Technological Revolution (Simon & Schuster. 2000 LINCOLN LABORATORY JOURNAL 215 . In the Shotput 1 test. on 12 August 1960. probably because it had then risen above the main lobe of the radar’s antenna pattern.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors of the Dicke Fix. Lincoln Laboratory (10 Feb. New York. Document 60-14627. Boston.. Several years later the antenna of the Boston Hill radar was demounted and used to replace an AN/FPS35 antenna that had been damaged by high winds at an operational Air Force site. This preliminary test was followed by the successful launch to orbit of the Echo 1 balloon from Cape Canaveral. helped the authors follow the trails of the Jug Handle Hill and Boston Hill radars after they were transferred from Lincoln Laboratory to MITRE. Frank Mastrovita. New York. Microwave Symp. Buderi.. Aarons. Final Report of 1952 Summer Study Group. 1996). This radar was not well suited to the observation of orbiting satellites. 7. The Jug Handle Hill radar was narrowband but tunable over the same frequency range. 10–14 June 1991 (originally published 1946). Lincoln Laboratory ( 6 July 1961). NUMBER 2. to improve its coverage of airborne targets. At that time Lincoln Laboratory began its FAA-sponsored program in air traffic control. found the tower still standing. DTIC #AD-325634. The Boston Hill radar reached its full operating capability in late 1959. pp. like the tower of the Jug Handle Hill radar (Figure 6). This document and others that were graciously provided by Thomas W. One of the authors. pp. 3. New York. VOLUME 12. P. Northrop. REFERENCES 1. Also at that time there was resurgence of interest in tactical radar applications. Chief Historian at the Rome Research Site. 5. in this issue. The balloon was inflated after launch. 1 January to 31 December 1959. as seen from the radar.” Technical Report 243. “Five Years at the Radiation Laboratory. operation of the MITRE Corp. 2. “Antenna and Receiver Measurements by Solar and Cosmic Noise. It. These two disparate influences led to the broad range of radar technology that is chronicled in other articles in this issue of the Lincoln Laboratory Journal. were invaluable. see the article entitled “Early Advances in Radar Technology for Aircraft Detection. and it could be seen with the naked eye.R. 4. when the balloon fell back into the antenna beam. On 1 April 1960 responsibility for the Boston Hill radar was transferred to the MITRE Corporation. just about the time when Lincoln Laboratory changed the thrust of its radar programs. attempts to detect them were unsuccessful. the balloon rose to an elevation angle of about 25°.” by Donald L. The Boston Hill radar also supported NASA’s Shotput 2 test on 16 January 1960.

Bob’s honors include Member of the National Academy of Engineering. In 1978 he joined Science Applications. He now putters around with a few old satellites that refuse to die. In 1969 Bob became chief scientist of the MITRE Corporation. from which he retired in 1988. Air Force’s Exceptional Service Award four times. He has received the U. . and two EHF Packages carried by host satellites FLTSATs 7 and 8. degree from Texas A&M College. 2000 . then technical director of MITRE’s Applied Science Laboratories. and the Druids (Omicron Delta Kappa) of the University of Missouri. During World War II. which had been established a year earlier to head the Radar Systems and Techniques department. where his first thirteen years were devoted to radar system engineering. He is a registered professional engineer in Massachusetts. a member of several professional societies. He has helped to design.• WARD AND NAKA Long-Range UHF Radars for Ground Control of Airborne Interceptors  . He retired from Lincoln Laboratory in 1994 after long service as manager of Satellite Operations (“Keeper of Old Satellites”).S. He now runs a small business.D. lectures. he served in the U.S. Engineering and Planning. as corporate vice president. CERA. and having the collecting habits of a pack rat. and repaired cryptographic equipment in the Pacific Theater of Operations. and Ph. where he installed. That same year. all in electrical engineering. he helped to prepare MIT Lincoln Laboratory: Technology in the National Interest. Subsequently he became associate technical director. where he led the development of the Boston Hill radar and was a member of the Air Force’s Frequency Diversity Advisory group. In 1959 Bob joined the MITRE Corporation. he joined Lincoln Laboratory. Tau Beta Pi. he reported to the Pentagon to become deputy director of the National Reconnaissance Office. That work has been in space communication. Inc. primarily in the development of systems that serve the diverse needs of the military and civil user communities by means of reliable links through satellites. and in 1982 he joined GTE Government Systems Corporation as vice president. 6. and currently a Distinguished Lecturer for the IEEE Aerospace and Electronic Systems Society. Inc. an illustrated history of the Laboratory published in 1995. Fellow of the Explorers Club. He has also contributed to the development of the operations centers associated with these satellites. In 1965 he switched from struggling to solve problems that involve (range)–4 to working on more tractable problems involving (range)–2.   joined Project Lincoln in June 1951 after completing his Doctor of Science degree (electron optics) at Harvard University. maintained.S.S. and space tracking and range instrumentation for NASA’s Project Mercury and for ballistic missile testing.  was born in Texas in 1924. and raises vegetables in the summertime.. The University of Missouri bestowed an Honor Award for Engineering and the Faculty Alumni Award. and member of the honorary societies Sigma Xi. 9. He is a registered professional engineer in Massachusetts. specializing in electromagnetic technology. Being blessed with a retentive memory. as its president. He received a B. test. 8. In 1972 he joined Raytheon to be director of Detection and Instrumentation Systems. and in 1975 returned to the Pentagon to become Air Force chief scientist. In 1954 Dr. NUMBER 2. In 1952. Naka became associate leader of the Special Radars group. build. consults. and operate in orbit Lincoln Experimental Satellites 5. including airborne-early-warning and ground-based surveillance radars. 216 LINCOLN LABORATORY JOURNAL VOLUME 12. writes. Army Signal Corps. In 1956 he became the leader of the Heavy Radars group. degrees from California Institute of Technology. and M.

Sign up to vote on this title
UsefulNot useful