Document text
A HETERODYNE DETECTION FM-CW
LASER RADAR USING A 10.6 Jj, SOURCE
Maurice F. Fraunfelder
:ry
L POSTGRADUATE SCHOOL
Monterey, California
jurqic
A HETERODYNE DETECTION FM-CW
LASER RADAR USING A 10.6 y SOURCE
by
Maurice F. Fraunf elder, Jr.
December 1974
Thesis Advisor:
T. F. Tao
antiwMwuw um
ii an iiihi M
Approved for public release; distribution unlimited.
T165958
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (Wh*n Data Ent*r»d)
REPORT DOCUMENTATION PAGE
READ INSTRUCTIONS
BEFORE COMPLETING FORM
REPORT NUMBER
2. GOVT ACCESSION NO.
3. RECIPIENT'S CATALOG NUMBER
4. TITLE find Subline)
A Heterodyne Detection FM-CW Laser
Radar Using a 10.6 u Source
5. TYPE OF REPORT a PERIOO COVERED
Electrical Engineer
Thesis ; December 1974
6. PERFORMING ORG. REPORT NUMBER
7. AUTHORf*;
Maurice F. Fraunfelder, Jr.
S. CONTRACT OR GRANT NUMBERC.)
9. PERFORMING ORGANIZATION NAME AND AOORESS
Naval Postgraduate School
Monterey, California 93940
10. PROGRAM ELEMENT. PROJECT. TASK
AREA 4 WORK UNIT NUMBERS
II. CONTROLLING OFFICE NAME AND ADDRESS
Naval Postgraduate School
Monterey, California 93940
12. REPORT DATE
December 1974
13. NUMBER OF PAGES
114
T*. MONITORING AGENCY NAME a ADDRESSf// dlttertnt horn Controlling Olllcc)
Naval Postgraduate School
Monterey, California 93940
!S. SECURITY CLASS, (ot thl* ra>o«;
Unclassified
15a. DECLASSIFI CATION/ DOWN GRADING
SCHEDULE
16. DISTRIBUTION ST ATEMEN T (ol thl* Rtporl)
Approved for THiblic release- distribution unlimited
17. DISTRIBUTION STATEMENT (ol (he tbttrrct *nl*r*rl In Block 20. II rllllmrmt from Report)
18. SUPPLEMENTARY NOTES
19. KEY WORDS (Contlnua on i.vfru aid* II nec****ry and Identity by black number;
Lidar
Optical radar
FM-CW radar
Range and velocity measurement
Acousto-optic modulation
20. ABSTRACT (Conllnu* on r*v*r*» lid* II r<*^**e*j? end Identity b/ block ma4«r)
The feasibility of a heterodyne detection FM-CW laser
radar capable of providing simultaneous range and velocity
information was investigated. Linear triangular frequency
modulation was accomplished with an acoustooptic modulator.
Two separate optical configurations were investigated. Ranges
to 2400 ± 30 yards were measured using a two- inch retro-
reflector as a target, with a 7000-yard maximum predicted
DD | jam "3 1473 EDITION OF 1 NOV 65 IS OBSOLETE
(Page 1) S/N 0102-014-6601 I
UNCLASSIFIED
SECURITY CLASSIFICATION OF "SHIS KAOE (Whin D*te gnfrcd)
UNCLASSIFIED
CliCUTITY CLASSIFICATION OF THIS P»GEC>V>ien D*l« Enf.r«;T
range. Velocities as low as .37 m/sec were easily measured,
and the system was estimated to have a velocity resolution
of approximately .05 m/sec. The measured receiver sensitivity
was approximately 30 db below the theoretical limit. Possible
causes and remedies for this reduced sensitivity are presented,
An unexplained zero velocity return was observed from a
moving target with the second optical configuration. A signal
contamination of the local oscillator beam was also observed.
Investigations of the cause of this contamination and possible
explanations are presented.
Simultaneous results of this project are reported in
FM-CW Laser Radar at 10.6 Microns, a thesis bv Lieutenant T.H.
Chance [Ref . 7] .
DD Form 1473 (BACK) UNCLASSIFIED
1 Jan 1 3 , —
S/N 0102-014-6601 SECURITY CLASSIFICATION OF THIS PAGEr*»»n Dmlm Enttfa)
A Heterodyne Detection FM-CW
Laser Radar Using a 10.6 u Source
by
Maurice F. Fraunfelder, Jr.
Lieutenant, United States Navy
B.S.E.E., Purdue University, 1969
M.S.E.E., Naval Postgraduate School, 1974
Submitted in partial fulfillment of the
requirements for the degree of
ELECTRICAL ENGINEER
from the
NAVAL POSTGRADUATE SCHOOL
December 1974
c./
Library
Mavai Postern
Monterey, California S ■
ABSTRACT
The feasibility of a heterodyne detection FM-CW laser
radar capable of providing simultaneous range and velocity
information was investigated. Linear triangular frequency
modulation was accomplished with an acoustooptic modulator.
Two separate optical configurations were investigated. ' Ranges
to 2400 ± 30 yards were measured using a two-inch retro-
reflector as a target, with a 7000-yard maximum predicted
range. Velocities as low as .37 m/sec were easily measured,
and the system was estimated to have a velocity resolution
of approximately .05 m/sec. The measured receiver sensitivi-
ty was approximately 30 db below the theoretical limit. Pos-
sible causes and remedies for this reduced sensitivity are
presented. An unexplained zero velocity return was observed
from a moving target with the second optical configuration.
A signal contamination of the local oscillator beam was also
observed. Investigations of the cause of this contamination
and possible explanations are presented.
Simultaneous results of this project are reported in
FM-CW Laser Radar at 10.6 Microns, a thesis by Lieutenant
T. H. Chance [Ref . 7] .
TABLE OF CONTENTS
I. INTRODUCTION ------------------- 11
II. GENERAL THEORY OF OPTICAL DETECTION- ------- 13
A. DETECTION MECHANISMS IN SEMICONDUCTORS - - - - 13
1. Photoconductor - __.__-___ 14
2. Photodiode ---------------- 15
B. SENSITIVITY COMPARISON OF OPTICAL
HETERODYNE AND ENVELOPE DETECTION- ------ 16
1. Envelope Detection Sensitivity ------ 16
2. Heterodyne Detection Sensitivity ----- 18
C. REQUIREMENTS FOR OPTICAL HETERODYNE
DETECTION- ------------------ 20
III. GENERAL THEORY AND ASPECTS OF FM-CW RADAR- ----- 21
A. PRINCIPLES OF OPERATION- ----------- 21
1. Range Information- -- - - - 21
2. Doppler Information- ----------- 23
B. COMPARISON OF FM-CW AND PULSE RADARS ----- 24
IV. OPTICAL MODULATION ---------------- 26
A. GENERAL CONSIDERATIONS - - - - - 26
B. ACOUSTO-OPTIC MODULATOR- ------ 26
V. GENERAL SYSTEM ASPECTS OF A 10.6 MICROMETER
LASER RADAR- ------------------- 29
A. LASER TRANSMITTER CONSIDERATIONS - - ----- - 29
B. ATMOSPHERIC PROPAGATION- - - - - - 30
1. Absorption ---------------- 30
2. Scattering ---------------- 31
3. Beam Distortion- ------- 33
C. RECEIVER OPTICS - 36
1. Unfocused Heterodyne Detection ------ 36
2. Simultaneously Focused Beams with a
Single Aperture- ------- -38
3. Signal Beam Only Focused ---------39
VI. DEVELOPMENTAL HETERODYNE DETECTION FM-CW RADAR - -42
A. SYSTEM DESCRIPTION --------- 42
1. Block Diagram- --------------42
2. Component Description- ----------42
a. Laser- ----- -43
b. Modulator- --------------43
c. Detector ---------------43
3. Theory of Operation- -----------45
4. Optical Configurations ----------45
B. EXPERIMENTAL PROCEDURES- ----------- 47
1. Modulator Measurements ---------- 47
2. Alignment Procedures -----------48
a. First Optical Configuration- - - - - - 49
b. Second Optical Configuration ----- 50
C. EXPERIMENTAL RESULTS ------------- 51
1. Zero Range Error -------------52
2. Range Measurement- ------------54
3. Velocity Measurements- ----------57
4. Receiver Field of View ----------59
5. Transmitter Divergence ----------61
6. Detector Evaluation- ----- -63
7. Receiver Sensitivity -----------65
VII. CONCLUSIONS AND RECOMMENDATIONS- - - 72
A. CONCLUSIONS- ----------- 72
B. AREAS OF IMPROVEMENT AND FURTHER STUDY - - - - 73
C. POSSIBLE SYSTEM APPLICATIONS --------- 75
FIGURES- ------------------------ 80
APPENDIX A - EQUIPMENT LIST- -------------- 107
BIBLIOGRAPHY ------ ________ 10g
INITIAL DISTRIBUTION LIST- --------------- H2
LIST OF FIGURES
Figure Page
1 Linear Frequency Modulation, Stationary
Target --------------------- go
2 Linear Frequency Modulation, Moving Target - - - 80
3 Dual Mode Operation- --------------81
4 Incident and Resulting Wave Numbers of the
Optical and Acoustic Energies- ---81
5 Basic Acousto-optic Modulator- ---------82
6 Bragg Angle Acousto-optic Modulator- - - ----- 83
7 Scattering Area Ratio for Spherical Water
Drops- ---------------------84
8 Lateral Phase Coherence Length for
Intermediate Turbulence- ------------85
9 Standard Deviation in Beam Arrival Angle
due to Intermediate Turbulence ---------86
10 Spatial Misalignment of Local Oscillator
and Signal Beams ---- -------87
11 Heterodyne Detection with Focused Beams- - - - - 88
12 Simple Photoconductor Circuit- ---------89
13 Basic System Block Diagram ----------- gg
14 First Optical Configuration- ---------- gj
15 Second Optical Configuration ---------- g2
16 Modulator Frequency Linearity- --------- 93
17 Spatial Intensity Recording of Modulated
and Zero-Order Diffracted Beams- --------94
18 Map of NPS ------ - - - 95
19 Observed Beat Frequency as a Function
of Range -------------------- 95
1 ;>
20 Map of Local Area- ---------------- 97
8
21 Photograph of Beam Path used for 305 Yard
Range Measurements ---------------93
22 Local Oscillator Contaminating Signal 4
Minutes after Laser Energization --------98
23 Local Oscillator Contaminating Signal 5
Minutes after Laser Energization --------99
24 Local Oscillator Contaminating Signal 7
Minutes after Laser Energization --------99
25 Local Oscillator Contaminating Signal 8
Minutes after Laser Energization -------- 100
26 Zero Range Offset- ------ ._.. 100
27 Heterodyne Detected Signal Prior to Signal
Processing ------------------- 101
28 Processed Signal of a Stationary Target
at 305 Yards ------------------ 101
29 Processed Signal of a Stationary Target
at 2400 Yards- ----------------- 102
30 Processed Signal of a Stationary Target
at 2400 Yards- ---------------- -102
31 Target Velocity Generation Apparatus ------ 103
32 Target Retro-reflector ------------- 103
33 Simultaneous Range and Velocity information
from a Moving Target at 305 Yards- ------- 104
34 Information of Fig. 33 with Different
Velocity -------------------- 104
35 Information of Fig. 33 with Different
Velocity ----- -- _.. tqs
36 Information of Fig. 33 with Different
Velocity ---------- _____ tqs
37 Photograph of the Optical Components
Utilized in the First Optical Configuration
(Fig. 14)- - - - 106
38 Photograph of the Optical Components
Utilized in the Second Optical Configuration
(Fig. 15)- ------------------- 106
ACKNOWLEDGEMENTS
I wish to express sincere appreciation to the Naval
Electronics Laboratory Center San Diego, Code 2500, par-
ticularly to Dr. Greg Mooradian and Rudy Krautwald. Without
their advice and equipment support, this project would not
have been possible. I am additionally grateful for the
assistance of Professors John Powers and T. F. Tao of the
Naval Postgraduate School.
I wish to thank Dr. J. Lcngo of the Rockwell Inter-
national Science Center and Dr. Peter Wang of Aerojet
General Corporation for the loan of state of the art
detectors which were used or evaluated in this project.
Lastly I would like to thank my co-worker and academic
colleague Lieutenant Thomas H. Chance, with whom this
joint project was conducted.
10
I. INTRODUCTION
One of the primary advantages of a laser communications
or radar system is the enormous economically achievable
antenna gain and hence excellent spatial resolution.
Another advantage is the extreme receiver sensitivity if
shot noise limited operation is achieved with a heterodyne
detection system [Refs. 1, 2, 3, and 4]. Several communica-
tions and radar systems have been built that approach within
an order of magnitude the theoretical quantum or shot noise
limit [Ref . 5] .
The combinations of the well known 8-13 micron atmospher-
ic window; high available output powers', high conversion
efficiencies ; and detectors with high quantum efficiencies
indicate that a carbon dioxide wavelength system may be the
best candidate for an optical radar [Ref. 5] .
Coherent (heterodyne) detection is more advantageous at
longer optical wavelengths due to reduced quantum noise,
less stringent optical alignment requirements, less atmo-
spheric coherence degradation effects, and a larger
resulting field of view for a fixed receiver aperture.
Thus a longer wavelength system is capable of a greater
sensitivity for a fixed field of view [Ref. C] .
In addition to the greater achievable sensitivity of
heterodyne detection, it also offers considerably reduced
background radiation interference and spatial signal
11
discrimination. The latter two combine to form a certain
degree of immunity to jamming from hostile sources.
Heterodyne detection also preserves the phase information
contained within the transmitted signal. This allows the
application of powerful signal processing techniques to
extract maximum information from the returned signal.
A frequency modulated continuous-wave radar system has
several advantages over the more conventional pulsed type
radars. It has no minimum range; the range resolution is
potentially greater than for pulsed systems; and it is
less susceptible to narrow band jamming. The radar sensi-
tivity is a function of the average output power, and for
a continuous wave system this is the full laser power. In
a pulsed system the average power is the laser power reduced
by the duty cycle. The inherent coherent property of a
continuous wave system allows for simultaneous measurement
of both velocity and range information. Since the doppler
information is proportional to the transmitted frequency,
the extremely high frequency of an optical system should
allow velocity measurements to a fraction of a meter per
second .
The excellent angular resolution of the laser coupled
with the excellent range resolution and high average powers
of the continuous wave radar in conjunction with the ease
and accuracy of velocity measurements indicate that a
heterodyne detection, FM-CW laser radar may be a potentially
useful system capable of development with current technology,
12
II. GENERAL THEORY OF OPTICAL DETECTION
Photo detectors fall into two classes: photon or
quantum detectors and thermal detectors. The latter type
are extremely narrow bandwidth low-pass devices and hence
are inappropriate for communications or radar detector
applications. Photon detectors can further be broken down
into four basic types of detectors: photoemissive , photo-
conductor, photovoltaic, and photoelectromagnetic . Since
photoemissive devices do not extent to wavelengths much
greater than 1 u [Ref s . 3, 4, 8, and 9], and since the
cumbersome magnetic field requirements and limited band-
widths of the photoelectromagnetic have limited its practi-
cal application, only photoconductor and photovoltaic
devices will be discussed.
A. DETECTION MECHANISM IN SEMICONDUCTORS
For a photon to be absorbed by a semiconductor, it is
necessary that the photon energy be greater than the
material energy gap, E . This places a long wavelength
limit, X , on the photogeneration process given by
. _ hv
Ac E~ '
8
The short wavelength limit is determined by carrier absorp-
tion at the semiconductor surface through surface trapping
and absorption effects [Refs. 4, 9, and 10]. The detection
13
mechanism is a creation of excess charge carriers caused
by the absorbed photon.
1 . Photo conductors
This is the simplest type of detector. It consists
of a single crystal slab of semiconductor material or of a
thin film of semiconductor material deposited on a substrate.
The thin film may be either single crystal or polycrystal-
line depending upon the material used. The detector is
biased with an external potential as shown in Figure 12.
The incident radiation changes the carrier concentration and
hence detector conductivity, and the resulting change in
detector current develops a potential signal across the load
resistor. The signal current can be expressed by the
following [Ref . 4] .
1 + (tOTj
m
where t is the carrier lifetime, and T is the carrier transit
time. From this it can be seen that the bandwidth of a pho-
toconductor detector is carrier lifetime limited.
Extrinsic infrared photoconductors rely upon optical
excitation of energy sites within the host crystal band gap.
These energy sites are caused by impurities such as copper,
mercury, cadmium or zinc. The prime disadvantage of this
type of detector is the low temperature operating require-
ments [Refs. 3 and 9].
14
Intrinsic photoconductors utilize band-to-band
excitation for the detection mechanism. Recent developments
have been achieved with such mixed crystals as Hg, CD Te
' &l-x x
and Pb1_xSnxTe [Refs. 11, 12, 13, and 14]. The operating
wavelength of these detectors can be controlled by varying
the molar content x. The significance of these detector
materials is a higher permissible operating temperature.
2 . Photovoltaic or Photodiode Detectors
A photovoltaic detector consists of a p-n junction
formed within an intrinsic semiconductor. Incident photons
create an electron hole pair within the crystal bulk. If
this carrier pair is generated within the depletion region
or within a diffusion length of the depletion region, the
internal electronic field of the diode will separate the
charge carriers and a potential will be developed across
the diode terminals. The photovoltaic detector's frequency
response is a function of the carrier diffusion time,
drift time within the depletion region, and junction
resistance -capacitance.
However, measurement by Peyton and others [Ref. 2]
on 2.5 urn HgCdTe photovoltaic detectors showed that the
frequency response was limited by the RC constant of the
p-n junction. Burke and Koehler have performed measurements
on HgCdTe photodiodes at an elevated temperature of 170°K.
A bandwidth of 200 MHz was achieved at 10.6 urn with 1.7
volt reverse bias [Ref. 15] . The technical importance of
these measurements lies in satisfying the cooling requirement
15
with thermoelectric coolers. In general, photovoltaic
detectors exhibit the highest detectivity when they are
operated in the short-circuit or zero bias mode. For mod-
erate or wide band operation, however, a reverse bias is
usually applied to increase the bandwidth by decreasing
junction capacitance.
B. SENSITIVITY COMPARISON OF OPTICAL HETERODYNE AND
ENVELOPE (DIRECT) DETECTION
1 . Envelope Detection Sensitivity
Photodetectors convert the absorbed optical radiation
into electrical output signals. They are square law devices
that respond to the intensity of light averaged over a few
optical cycles. This is due to the limited speed of
response of the carrier transport and relaxation processes
within the detector. These responses do not have sufficient-
ly short time constants to respond to the optical field
variations. The expression for the conversion of the
incoming optical power into a direct current is :
nqP^
This is a fairly important relationship in that is
can be used to calculate the quantum efficiency of a detector
if the optical power and signal current are known. The
minimum signal that the detector is capable of detecting
is a function of the different noise sources within the
system. The familiar expression for shot noise due to an
average DC durrcnt is N~, = 2qIBR. vhere t is the average
DC current .
16
The thermal noise contributed by the detector is
NS2 = 4KTB, and the noise caused by the effective temperature
of the following amplifier is 4KT ffB. The power signal-
to-noise ratio thus becomes:
O 2rl
S/N = — ^ . (3)
2qB {jH (PS+PB) + ID} RL * 4KB (T+Tef£)
Now if the thermal noise dominates the shot noise (usual
case for a well designed and state of the art detector) ,
then the signal-to-noise ratio reduces to:
nqPo ?
HrT^ RL
S/N = AVP, (t+t ,.) (4)
' ^ e±±J
Equations (2) and (3) are for a photovoltaic
detector as generation-recombination noise is not included.
Also 1001 intensity modulation is assumed. Assume a 50 OHM
detector operating into a matched load. The following
amplifier has a 3 db noise figure. The noise equivalent
power is the value of optical signal power required to
produce a signal- to-noise ratio of 1. Therefore:
NEP _ 2hv (K (T"W}l/2 t (5)
/B nq R
If the quantum efficiency is assumed to be 1, and
the following amplifier has a noise figure of 3 db , then the
17
amplifier has an effective noise temperature of 290°K.
Using these values (5) yields:
— = 2.36 x 10"12 Watts/(Hz)1/2
/B
2 . Heterodyne Detection Sensitivity
As with conventional radio or radar receivers,
optical heterodyne detection involves the mixing of the
return signal with an optical local oscillator to produce
a current at some intermediate frequency; however, it is
done for different reasons and produces different system
results. The mean value of this intermediate current can
be shown to be [Ref . 1] :
i 2 = rH£h ? P P (6~)
XIF lhvJ L LO S l0)
The transducer gain is defined as the IF output
power divided by the available signal power. Using (6)
and the definition of transducer gain, it can be shown
[Ref. 2] that:
nq^2
2[GD (1 ♦ CDRS) + ^FRSCD-
r _ IF VlTvJ rL0 (1,
bT - -p s y~ • *-/J
Thus it can be seen that the transducer gain is directly
proportional to the local oscillator power. If (6) is
substituted into (.3) then the value of S/N becomes:
18
7 /-nib 2p p p
** lhvJ *L0 S 1
S/N = . (8)
2«B C CPS+PB+PLO> + V RL + 4KB <T+Teff>
If P.q is made large enough to dominate the
denominator then (8) becomes:
S/N - Ev§ ' ■ (9)
Using the previous definition of NEP, it is trivial to
show that:
NE P _ hv
B n
(10)
This is the well known "shot noise limited operation"
of a heterodyne detection system.
If the quantum efficiency is assumed to be 1 as was
the example of envelope detection then:
^— = 1.88 x 10"20 Watts/Hz .
Therefore, for a bandwidth of one hertz the heterodyne system
has a theoretical sensitivity eight orders of magnitude
greater than a comparable envelope detection system.
Several workers have reported achieving heterodyne systems
that approach within an order of magnitude of shot noise
limited operation [Refs. 2, 5. and 6].
19
B. REQUIREMENTS FOR OPTICAL HETERODYNE DETECTION
For the relationship of (6) to represent the physical
situation the signal and local oscillator wavefronts must
maintain the same phase relationship over the entire sensitive
area of the detector. This requirement can only be satisfied
by the following conditions [Ref . 16] :
1. The two beams must have the same mode structure.
The dominant TEM mode is preferred.
oo r
2. The two beams must be spatially coincident, and to
provide maximum signal- to-ncise ratio, their diameters must
be equal.
3. The two beam pointing vectors must be coincident.
This implies angular alignment of the beams must be main-
tained. A quantitative discussion of this alignment will
be presented in a later section.
4. The wavefront must have the same curvature. Both
must be plane waves or, if curved, both must have the same
radius of curvature.
5. The beams must be identically polarized so their
electic vectors wil] be coincident.
Although these requirements appear to impose extremely
stringent requirements upon the system, they can be satis-
fied at the longer infrared wavelengths as indicated by the
experimental results cited in the discussion of heterodyne
detection sensitivity.
20
III. GENERAL THEORY OF FM-CW RADAR
A. PRINCIPLES OF OPERATION
A radar detects the presence of targets by transmitting
electromagnetic energy and absorbing the returned energy.
A conventional pulsed type radar transmits a short
pulse and measures the elapsed time to the returned energy.
The pulse duration and time between pulses are the control-
ling parameters of range resolution and the maximum unambig-
uous range. In a CW radar the measurable information is the
target's velocity determined by the doppler relationship:
2v
At 10.6 micrometers wavelength a velocity of 1 m/sec corre-
sponds to a doppler shift of 188.7 kHz. Since filters of a
few kHz are readily available doppler measurements of
velocities to a fraction of a meter per second are readily
achievable .
1 . Determination of Range
A simple CW radar cannot measure range since it
has no method of correlating the returned signal to the
instant of its transmitted time. This inability is related
to the extremely narrow bandwidth of the transmitted wave-
form. Some sort of timing mark must be applied to a CW
carrier to allow this correlation. The more distinctive
the marker, the more precise will be the range measurement
21
and the broader will be the transmitted spectrum. This
interaction follows from the properties of the Fourier
transform. One of the easiest methods to produce this
timing mark is to linearly frequency modulate the CW
carrier with either a sawtooth or triangular waveform.
For either case the returned signal heterodyne
spectrum will consist of a envelope. The spectral
width will be a function of target range [Refs. 17, 18,
and 19] . The closer the range the narrower will be the
spectrum and hence, the greater the range resolution. This
is usually the desirable condition.
The relationship between the transmitted, received,
and heterodyned frequencies is illustrated in Figure 1.
The heterodyne or beat frequency is determined by:
£b = § At . (12)
Since the frequency deviation is linear, ppr- can
be replaced by 2f Af, and At can be replaced by the signal
transit time 2R/C. Substituting and rearranging results
in the range equation
(13)
4f Af
m
The range resolution will be determined by the
combination of the IF filter bandwidth and the returned
signal heterodyne spectrum.
22
The range resolution will be determined by the IF
filter bandwidth if it is smaller than the target spectral
bandwidth. If not, the target spectral bandwidth will
control the range resolution.
2 . Determination of Velocity
The frequency relationships of Figure 1 and the
preceding discussion assumed a stationary target. If the
target has relative motion with respect to the radar, a
doppler frequency shift will be superimposed on the beat
frequency and an erroneous range will result. Figure 2
shows the frequency relationships of Figure 1 modified by
doppler information.
On one portion of the frequency-modulation cycle,
the beat frequency will be either increased or decreased
by the doppler frequency depending upon the sign of the
relative velocity of the target. On the other portion of
the cycle, the opposite direction frequency shift will be
observed.
As an example, consider a closing target. During
the rising portion of the transmitter spectrum, the doppler
frequency will subtract from the beat frequency and during
the falling portion of the spectrum, the doppler frequency
will add to the doppler frequency.
fb (up) = fr - fd_ (a)
(14)
fb (down) - f '+ f. (b)
23
By summing the two frequencies the range information
can be determined, and by taking the difference the velocity
information can be determined. Since the two frequencies
involved occur during different time increments a memory
device must be utilized. High-speed counters and digital
arithmetic units should be capable of performing the required
functions .
If precise velocity measurements were desired, a
dual mode feature could be utilized as shown in Figure 3.
During the CW velocity mode, the heterodyne spectrum
broadening caused by the waveforms of Figures 1 and 2 would
be absent, and the velocity resolution would be limited only
by the filter bandwidth of the measuring device. However,
this velocitv precision would not be reciuired for most
radar applications.
B. COMPARISON OF FM-CW RADAR WITH PULSED SYSTEMS
The FM-CW radar does not have the minimum range
restrictions of a pulsed radar. It has the added advantage
of an average to peak power of unity which is highly desir-
able when using a transmitter which is peak-power limited.
For equal transmitted bandwidths , pulsed and FM-CW radars
have comparable range resolutions [Ref. 20]; however, the
pulsed type radar must have a receiver bandwidth comparable
to the transmitted bandwidth. The FM-CW radar can have a
receiver bandwidth which is a small fraction of the trans-
mitted bandwidth. This implies an increased permissible
receiver signal-to-noise performance and fewer permissible
24
stages of amplification due to the gain-bandwidth
product.
To change range resolution in the FM-CW radar,
one merely has to change the transmitted frequency deviation,
To produce a comparable change in a pulsed system, one
generally would have to change the pulse width, pulse
repetition frequency, and the receiver bandwidth.
For equal average transmitted powers, target
illumination times, receiver noise figures, antenna gains,
integration efficiencies, and optimized bandwidths , pulse
and FM-CW radars have comparable maximum range capabilities
[Refs. 18 and 20] .
25
IV. OPTICAL MODULATION
A. GENERAL CONSIDERATIONS
Since the phase velocity of an electromagnetic wave is
a function of the index of refraction of the medium through
which it passes, any material whose index of refraction can
be varied can be used to modulate an optical beam.
If the index of refraction is varied by means of an
electric field by the Pockels or Kerr effect, then it is
known as an electrooptic modulator. By various physical
configurations this type of modulator can be used for
intensity, phase, polarization and deflection modulation
[Refs. 21 and 22]. If the index of refraction is varied
by a magnetic field, it is known as a magnetooptic modulator
Here the prime mechanism is a rotation of the wavefront
polarization and hence, it is not suitable for intensity
or frequency modulation. If the index of refraction is
varied by a mechanical force (pressure) , it is known as
an acousto-optic modulator. The mechanism is the diffrac-
tion of an optical wave by a traveling acoustic wave. The
diffracted wave is not only deflection modulated, but it
is also simultaneously frequency modulated.
B. ACOUSTO-OPTIC MODULATION
An explanation of the interaction of light and sound
can be obtained through the dual particle-wave nature of
light and sound energy [Ref . 23] .
26
Using this approach a light beam with a propagation
vector K and a frequency to can be considered a stream of
photons with momentum tiK. and energy "hoj. In a like manner
the sound beam is modeled as photons with momentum hK and
r s
energy "hw . Since the momentum and energy of the system
must be conserved, the propagating vector of the diffracted
wave must be:
K = K. + K and the frequency must be:
o L s
Figure 4 shows the required momentum conservation and
the resultant direction of propagation of the diffracted
beam.
Figure 5 is a schematic diagram of a basic acousto-
optic modulator. Note that if the optical beam is
orthogonal to the acoustic beam, a zero-order beam
(unmodulated) , first order diffracted beams (modulated
by the acoustic frequency) , and higher order beams (not
shown) will result.
If the optic and acoustic waves are offset from the
orthogonal condition by the bragg angle as shown in
Figure 6, a zero order and higher order diffracted beams
in a single direction only will be observed. For this
case, the energy of the single first order diffracted
beam will be twice that of a single order diffracted beam
of the basic modulator. The frequency modulation ;\rill be
27
the same. Diffracted beams of higher order than the first
are usually neglected due to the extremely low energy content
For a more detailed explanation concerning bandwidths , opti-
cal demands, power requirements and general parameters, see
Refs. 4, 24, 25, 26 and 27.
V. GENERAL SYSTEM ASPECTS OF A 10.6 y LASER RADAR
On a macroscopic scale a radar consists basically of
a transmitter, a receiver and a propagation medium through
which the electromagnetic energy must pass. For the purpose
of this discussion the laser will be viewed solely as a
radar transmitter. Only those parameters that pertain to
system performance will be considered. The receiver can be
considered to consist of an optical antenna, detector and
associated signal processing. Since the majority of the
important detector characteristics have already been covered,
this section will discuss the antenna or optical considerations
LAbcn KAJJ/A.R iRAiNOi'lx 1 xcis.
Since the acousto-optic modulator is polarization
sensitive, the laser transmitter should be polarized and
have a polarization corresponding to maximum efficiency
of the modulator.
Previous discussions indicated that the transmitter and
LO beams should have the same spatial coherence. For maximum
signal-to-noise ratio and ease of alignment, this coherence
should be restricted to the dominant TEM mode of propaga-
tion. Additional restrictions are minimal for systems
deriving the transmitted and local oscillator beams from
a single laser source.
29
B. ATMOSPHERIC PROPAGATION
The medium through which an optical beam travels has
a significant effect on the system performance. Absorption
and scattering by the atmospheric constituants and sus-
pended aerosols must be considered for a homogeneous medium.
In the more usual case of a turbulent medium, the additional
system perturbations of the beam shape, dimensions and
electromagnetic properties must be considered.
1 . Absorption
The fraction of an optical beam intensity passing
through a transmission medium is proportional to the distance
traveled [Refs. 3 and 8]. Stated mathematically this
becomes :
HT = "KAdL •
Solving for I results in:
I(x) = I e"KAL . (15)
o
A number of different atmospheric constituants
attenuate an infrared beam as it passes through the atmo-
sphere. The primary effect can be attribuyed to water
vapor (H?0) , carbon dioxide (C02) and ozone (0_) (high
altitudes only). Considerably lesser effect may be
observed from methane (CM.) , nitrous oxide (N^O) , and
carbon monoxide (CO) if the path is long. The amount of
water vapor in the path varies over a wide range, and while
30
the carbon dioxide is mixed more nearly uniformly, it may
still vary appreciably in various air masses. Thus a
detailed knowledge of the meteorological conditions is
necessary to perform an exact calculation of the infrared
transmission. Even though such detailed information is
seldom, if ever, available it is usually possible to make
gross predictions using such meteorological parameters as
temperature, pressure and relative humidity. Early workers
performed many field measurements over typical paths and
under a variety of weather conditions. Results of these
measurements in graphical and tabular form are available
[Refs. 8 and 9], and are of value to the system engineer who
must estimate the system performance under field conditions.
2. Scattering
Scattering has the same intensity as a function of
path length relationship as absorption. The effect of
scattering on infrared transmission can, therefore, be
represented by (15) with the absorption coefficient K
replaced by the scattering coefficient K . The total
transmittance of the path can be represented by:
t = e"aL (16)
where a = K. + K„ and is known as the extinction coefficient
- Two major models for scattering exist. If the
scattering particle is considerably smaller than the optical
wavelength, it is known as Rayleigh scattering. Here the
31
4
scattering coefficient is inversely proportional to A . As
such, shorter wavelengths are scattered much more than
longer wavelengths, and for all practical purposes Rayleigh
scattering can be neglected at 10.6 micrometers. The second
model occurs if the size of the scattering particle is
comparable to, or larger than, the optical wavelength.
Mie scattering can be described by the following
empirical relationship [Ref. 4]
v _ 3.91 r A ,-.585V1/3 ,17.
Ks " ~y— [755] (17)
where V is the visual range in kilometers, the wavelength
is in microns, and the path length is in kilometers. If
the scattering centers are spherical, the relationship:
Ks = Tmyr2 (18)
can be used where y is the scattering area ratio and is a
measure of the efficiency with which a center scatters the
incident energy. n is the number of scattering centers per
cubic centimeter and y is the radius of the scattering center
Figure 7 shows the relationship between y and the ratio of
the scattering center radius to wavelength (r/A) . Notice
that the center is the most efficient scatterer when the
center radius size is approximately equal to the wavelength.
Measurements of the droplets in fogs show that their radii
32
range from 0.5 to 80 microns. The peak of their size
distribution usually occurs between 5 and 15 microns.
Therefore, fog particles are efficient scatterers at the
10.6 micron wavelength [Ref. 9].
3. Beam Distortion
In the presence of a turbulent propagation medium,
the beam may be distorted in several different ways. These
beam distortions may appreciably affect the system perfor-
mance. This turbulence which is a manifestation of thermal
and pressure inhomogeneities causes a change in the index
of refraction of the medium. The changes in the index of
refraction modify the propagation constant and Poynting
vector of the optical beam and this modification can be
summarized as follows [Ref. 28] :
a. Beam steering - angular deviation from the
line of sight path.
b. Image dancing - variations in the angle of
arrival of the beam wavefront.
c. Beam spreading - small angle spreading which
increases the beam divergence.
d. Beam scintillation - small-scale destructive
interference within the beam cross section.
e. Spatial coherence degradation - losses in phase
coherence across the beam phase front.
If the turbulent medium is modeled as consisting
of discrete blobs, each homogeneous but with a different
index of refraction from adjacent blobs, then the relative
33
size of the optical beam and the turbulent blobs will
determine which of the above distortions will predominate.
If the blob size is I and the beam size is dR , then for
dg/£ << 1 the major effect will be beam steering and image
dancing. For dR, >> 1, the major effect will be beam
spreading, beam scintillation, and spatial coherence
degradation.
Numeric models of a turbulent medium have been
developed by Tatarski [Ref. 29]. In Tatarski's model, the
degree of atmospheric turbulence and its relationship to
the optical properties of the atmosphere can be characterized
by a structure constant for refractive index fluctuations,
C (r) . The value of the structure constant varies with
altitude and time of day. Typical values for daytime
conditions near the earth are:
Weak turbulence - C (r) = 8 x 10"9 m"1'3
n
- 8 -1/3
Intermediate turbulence - C (r) = 4 x 10 m '
n
- 7 - 1/3
Strong turbulence - C (r) = 5 x 10 m
If the smallest inhomogeniety blob is I , the
largest is L , and the path length is L, then the fluctua-
tions in phase between points of the wavefront separated
by the distance p, as postulated by Tatarski, can be
represented by:
1.46(^-)2 P5/3 / C 2(r)dz for I < p < (AL)1/2
■y a o n o
% (P) - L (19)
* ?ir ? 5/3 ? 1 /?
2.91(^fr P ' „/ C Z(r)dz for L > p > (XL) 7
A on o
34
The first condition leads to beam spreading, beam
scintillation, and spatial coherence degradation. The second
condition leads to beam steering and image dancing. Figure 8
shows the lateral phase coherence length as a function of
path length for intermediate turbulence. Figure 9 shows the
phase front angle of arrival deviation as a function of beam
diameter for intermediate turbulence.
Fried has derived an expression for the degradation
in the signal-to-noise ratio for an optical heterodyne
receiver in terms of the relative sizes of the receiver
aperture, dR, and the phase coherence dimension, r [Ref. 30]
An examination of his derivation reveals that little improve-
ment in the signal-to-noise ratio will result by increasing
the receiver aperture beyond r . The quantity r is related
r J o l ' o
to the transmission wavelength, zenith angle, 6, and
receiver altitude, H , by the relationship:
o
05 [X]6/5 cos3/5(0)
T(2/3)
r f ? / \ Ho 1
l^L> J 3200.L
3/5
(20)
where r(x,y) is the incomplete Gamma function [Ref. 31]. The
coherence dimension can also be related to Tatarski's atmo-
spheric structure constant by [Ref. 31] .
rQ = 1.2 x 10"8 U)6/5 (L)~3/5 (Cn(r))"6/5 (21)
From the X ' dependence of r , it can be seen that
the coherence dimension is about 30 times greater for 10.6 u
than for . 63 u .
35
The X ' dependence of r has been experimentally
verified and reported by Gilmartin and Holtz. Their
measurements encompassed wavelengths from the visible through
10 microns under conditions of medium and severe atmospheric
degradation of resolution [Ref . 32] .
In addition, they have shown that focused beam size
and focused beam wander can be predicted from relatively
simple visual resolution measurements. Additional theoret-
ically predicted and experimentally verified methods of
determining the degree of coherence of a laser beam passing
through a turbulent medium has been reported by Grant and
Ageno [Ref. 35] . This work presents the relationship
between the mean square angular deflection of a laser beam
and the corresponding degree of coherence of the wave front.
C. RECEIVER OPTICS
An optical heterodyne receiver can be separated into
two separate components: an optical antenna and an optical
receiver. Since the important aspect of receiver sensitivity
has been discussed in a previous section, this section will
consider the directional characteristics and spatial require-
ments of an optical antenna.
1 . Unfocused Heterodyne Detection
If the local oscillator and signal beams are both
colimated (plane wavefront) and spatially misaligned as
shown in Figure 10, the following mathematical model of
heterodyne signal spatial degradation can be developed
[Refs . 3 and 4] .
36
Assume the detector to be an ideal square law
device. If the signal and local oscillator fields
respectively are:
Es(t) = As cos (ojgt + $s - -£— )
EL(t) = AL cos (w t + $L)
then the instantaneous detector input signal will be equal
to
S(t) = [Ag cos Ogt + $g - -—-) + AL cos (w t + $L)] (22)
x
The intermediate or heterodyne signal component is
determined by performing the squaring operation and applying
trigonometric operations. If this is done, the detector
heterodyne current will be the time average and spatial
integral over the detector surface.
w„x
iTt3 « // cos (wTC + *c - $. + -^-) dA (23)
IF AREA IF S L V
Performing the integration yields
sin (w~d/2v )
ilc tt Ac At cos (wTr + $c + $t ) — r — j/-?,, -^ (24)
IF S L v IF S 1/ (o)cd/2v J '
If the signal degradation due to misalignment angle
$ is to be kept less than 10%, then the term (w^d/Zv )
must be less than .8 radian. By making the substitution
37
C < X
v = — t-Z — r it can be shown that i() ~ it must be maintained,
x sin ty 4d
The first null in the heterodyne signal will occur for
2TrCd
uqd X
4 — - 3-14 w
x 2C
sm ty
or sin ty = -r . For small 4> , ty z -r .
If a 5 mil x 5 mil detector is assumed, then
. 10.6x10 m 0, .,,. ,.
ip = j— = 83 milliradian
1.27 x 10"4m
Of course, the obvious disadvantage of this simple receiver
is the extremely small intercepted signal energy and hence
poor receiver signal-to-noise ratio.
2 . Simultaneously Focused Beams with a Single Aperture
Corcoran has performed an analysis of heterodyne
detection with focused signal and local oscillator beams by
a single aperture [Ref. 34]. His analysis assumes that the
focusing element is in the Fraunhofer region or far field
of the light sources. As in the previous analysis, this
assumes that the light incident upon the aperture is effec-
tively a plane wave. Figure 11 is a representation of the
detection process. His results show that
[(sin 6 - sin ei)OdA/X)]
i (t) « cos a)TT,t sin ci — r~ ■ 5 -tt* o^ (26)
IFV ■* IF irS' (.sin G? - sin 6-J
38
when 8. =0, the first zero of the heterodyne current occurs
ird.
when sin(99) — =-^ = tt. For small values of 9- this reduces
to 8 = J- .
2 dA
Focusing with a single aperture, therefore,
decreases the receiver field of view when the receiving
aperture is larger than the detector. Corcoran' s analysis
of Fraunhofer region unfocused detection agrees with
Refs. 3 and 4.
Additional theoretical predictions relating the
receiver aperture, wavelength and angular field of view
have been presented by Siegman [Ref. 35]. By several
methods he has shown that the product of the effective
receiver aperture, AR , and the solid angle field of view,
ftR, are approximately constant and related by
ARnR = X2 . (27)
n .Re., 2
\ v ^
By making the substitution -j— = — n it can be
ttR
seen that for a circular aperture of diameter d.
6 - % • . <28>
This is in good agreement with Corcoran's results.
3 . Signal Beam Only Focused
Read and Turner have presented theoretical calcula-
tions and experimental verification of an optical heterodyne
39
technique which greatly reduces the stringent angular
requirements [Ref . 36] . In this method the signal beam is
focused with a diffraction-limited lens or mirror. A
colimated local oscillator beam is superimposed on the
resulting Airy pattern and optical heterodyning results.
The Airy disk and local oscillator beam can interact effi-
ciently because both wavefronts will be plane [Ref. 23] .
The greatest heterodyne efficiency and signal-to-noise
ratio will be achieved if only the central disk of the
Airy pattern is used. This is due to phase reversals of the
successive diffraction rings.
The electric field strength due to the signal beam
in the focal plane is
J Car)
Es(t) = as cos cv ♦ *s) -far (29)
in which
a = 2iTa/Xf
a = lens radius
f = lens focal length
r = radius from disk center
J, (x) = Bessel function of order one
R = disk radius .
At a distance I from the axis, the two wavefronts
are out of phase by — —- .
The elemental signal is, therefore, proportional to
lT(f) Z
dijp a AgA. cos (ojTpt + $„ - $. ) ccs (— — ) . (30)
40
The total signal is obtained by integrating over the
area of the Airy disk.
This integral is not expressive in elementary
functions for finite disk sizes, but if the assumption is
made that the signal beam intensity is uniform over the
disk area, then the integral results in
A A
iIF(t) " HP" R Jl C^T^ • C31)
Now if the substitution R = 1.22 ^f/dA is made,
the voltage signal-to-noise voltage ratio can be expressed
as
K J, (2.44TT(f> f/d )
(S/N)v = ± ^ ^- (32)
dA
The first zero will occur for cf> = . 5-t— . If the optics
has a speed of f/(no) of 10, then the angular field of view
will be 50 mrad. This reduced angular requirement is not
without restrictions, however, since the center of the
focused spot must not deviate from the center of the
aperture in the focal plane by more than a fraction of its
diameter .
41
VI. DEVELOPMENTAL HETERODYNE DETECTION FM-CW RADAR
A. SYSTEM DESCRIPTION
Two optical configurations were constructed and tested
for the developmental system. Figures 37 and 38 are photo-
graphs of the physical optics utilized. Several high-quality
photovoltaic detectors were available and each was tried in
the two-system configurations.
1. Basic System Block Diagram
The over-all system block diagram is shown in
Figure 13. Equipment model numbers and component parameters
are given in Appendix A. The system optics (two separate
configurations) are shown in more specific detail in a
subsequent section.
2 . Major Component Description
a. Laser
The laser used for the developmental system
was a Honeywell model 3000 three-watt continuous wave CO-
laser. The laser output was vertically polarized, and
output power was continuously controllable over a range
of .2 - 3 watts. A piezoelectric transducer (PZT) was
attached to the rear cavity mirror for control of the
selected radiation line although this feature was not used
for the described system. The laser was a sealed cavity
type and was water-cooled. The water flow requirement was
about .25 gallons per minute.
42
b. Modulator
The 10.6 micron AO modulation system consisted
of three main components: a water-cooled Germanium AO
modulator; a pair of Germanium focusing lenses, each with
a 5" focal length; and a combination voltage controlled
oscillator (VCO) and RF power amplifier. This latter
combination will be collectively referred to as the modu-
lator driver. The modulator driver was capable of supply-
ing six watts of CW power to the modulator. The bandwidth
of the modulator system was measured and found to be in
excess of 20 MHz. The output frequency linearity as a
function of modulator driven DC control voltage input was
measured and it was found that non-linearities existed
above 45 MHz and below 35 MHz. The effective bandwidth of
the modulator for this system was concluded to be 10 MHz.
c. Detectors
Three detectors were used during the system
tests. All three detectors were of the PbSnTe photovoltaic
type. One detector was a Raytheon IR-101 PN photovoltaic
detector packaged within a glass dewar. The other two
detectors were Rockwell International PIN photovoltaic
detectors packaged within stainless steel dewars . Perti-
nent detector parameters are given in Appendix A.
3 . Theory of Operation
Refer to Figure 13 for the following description
of system operation. The bias supply V, determine the
center frequency of the modulated optical beam. Since the
43
linear bandwidth of the modulation systems extends from
35-45 MHz, the value of V, was set to -7.5 volts corre-
sponding to a center frequency of 40 MHz. The sweep
voltage was obtained from a function generator and passed
through a high-pass filter. It was adjusted for an ampli-
tude to produce a frequency sweep from 35-45 MHz out of
the modulator driver. Since the laser output beam is fre-
quency swept in a triangular fashion, the detected laser
return will also be frequency swept in a triangular fashion
from 35-45 MHz. This detected signal is amplified by a
combination of wideband amplifiers resulting in 53 db of
amplification prior to insertion of a balanced mixer. The
output of the VCO is available at the DC input to the modula-
tor driver and was extracted by a high-pass filter. This
filter passes the 35-45 MHz swept signal while blocking the
much lower frequency sweep voltage. The 35-45 MHz swept
signal is mixed in a double balanced mixer with a 30 MHz
signal and translated to 65-75 MHz. This signal is then
passed through a 60 db wideband amplifier to raise the
signal to the proper level to drive the "L" port of a
double balanced mixer. The 65-75 MHz filter eliminates all
undesirable mixer products from mixer number 1. If the
detected signal and the "L" port signals of mixer 2 were
sweeping from 35-45 MHz and 65-75 MHz, respectively, in
synchronism (zero range and zero doppler) , then a single
output frequency of 30 MHz would be observed. As the range
is increased, the 30 MHz signal splits into two distinct
44
frequencies, one above and one corresponding below 30 MHz.
The frequency shift from 30 MHz is proportional to the
range. Doppler information manifests itself in a shift
of the signal pair centroid from 30 MHz. The mathematical
relationship describing these phenomena was presented in
Chapter III, A, sections 1 and 2.
4. Optical Configurations
Two system optical configurations were tried. The
first configuration was initially tried at the Naval
Electronics Laboratory Center (NELC) in May 1974 during
the author's industrial experience tour. Refer to Figure 14
for a description of this configuration. The designation
x/y refers to a beam splitter and means x per cent of the
beam energy is reflected by the beamsplitter and y per cent
is transmitted by the splitter. Ninety per cent of the
laser beam is reflected to become the radar output beam.
The 5" focal length lens prior to the AO modulator focuses
the beam to a small waist size to increase the modulator
bandwidth. The 5" focal length lens after the AO modula-
tor colimates the two emerging beams and restores the output
beam divergence. The blocking aperture passes the modulated
beam and blocks the zero-order beam. The 50/50 splitter
following the blocking aperture passes 50% of the output
beam and also reflects 501 of the return beam down to the
detector. The 5/95 splitter in the output path passes 95%
of the output signal and returned signal but reflects
essentially 100% of the visible He-Ne beam. Fine
45
adjustment of this splitter superimposes the C0_ and He-Ne
beams, and hence the radar output beam can be aimed according
to the visual beam. The output mirror is adjustable in
both vertical and azimuth and facilitates the steering pro-
cess. The mirror in the He-Ne path has a small hole near
the center through which the visible beam passes. By aiming
the telescope at the front surface of this mirror, the
telescope reticles can be precisely aligned with the
illuminated area. The local oscillator beam is the beam
transmitted through the 90/10 splitter. It is folded three
times with the front surfaced mirrors and passed through a
limiting aperture. This aperture controls the amount of
local oscillation power incident upon the detector. The
5/95 splitter just prior to the 8" focal length lens passes
951 of the returned energy and 5% of the local oscillator
energy. It is adjusted to superimpose the signal and local
oscillator beams. The 8" focal length lens focuses both
the local oscillator and signal beams to an Airy disk com-
parable to the detector size. The absorber absorbs that
portion of the transmit beam reflected by the 50/50 splitter.
This power is appreciable and oculd become a health hazard
if it were not blocked.
The second optical configuration is shown in
Figure 15. This configuration was constructed to overcome
two distinct disadvantages of the first system. The 50/50
splitter attenuates 3 db of the output beam and also 3 db
on the receive path, hence elimination of this splitter
46
should immediately cause a 6 db increase in system
performance.
An increase in the receiver aperture would also
cause an increase in the system performance; hence a simple
Newtonian antenna system was chosen. The path through the
modulation system is essentially unchanged. The zero order
blocking aperture was replaced by a folding mirror and the
zero order beam was used as the local oscillator beam.
Beam elevators had to be used to raise the local oscillator
and signal beams to a height required by the larger diameter
of the receiver optics. The small front obstruction mirror
was double surfaced. The output beam was reflected from
the front surface, and the returned signal was reflected
from the rear surface. This was an elliptical mirror
measuring 1-1/2" by 2-1/4". The primary reflecting mirror
was a 6" diameter mirror with a 60" focal length. Notice
that the local oscillator beam was unfocused. The visual
alignment mechamism was the same as in the previous
configuration.
B. EXPERIMENTAL PROCEDURES
1. Modulator Measurements
Parameter measurements of the basic system components
consisted of modulator frequency deviation and power output
of the modulated beam as a function of VCO driver voltage.
The deflection efficiency and over-all transmission effi-
ciency was measured at the center of the effective
4 7
bandwidth (40 MHz) . Figure 16 shows the modulation linearity
as a function of VCO driving voltage. Figure 17 shows the
modulated and zero order beams after the colimating lens.
This figure is the result of recording the spatial intensity
patterns of the beams on temperature-sensitive paper. The
Bragg angle for this moudlator at 10.6 microns was approxi-
mately 2.21 degrees. The modulator was mounted on a stand
with x, y, z and Bragg angle micrometer adjustments, and
each dimension was adjusted for maximum power output of the
modulated beam. The zero order beam was blocked with an
aperture. The power input to the modulator, and the modu-
lated beam and zero order diffraction beam powers were
measured at 40 MHz. If the deflection efficiency (DE) is
defined as:
DE
Power in Modulated Be
am
Total Power Output
and the transmission efficiency (TE) is defined as
TF = P°wer Output
Power Input
then the resultant measured efficiencies were a deflection
efficiency of 53% and a transmission efficiency of 55%.
2 . Alignment Procedures
The alignment procedures for the two optical
configurations were appreciably different; therefore, each
procedure will be discussed separately.
48
a. First Optical Configuration
Refer to Figure 14 for this discussion. Prior
to insertion of the AO modulator, the 90/10 splitter was
adjusted to produce a 90° beam reflection which was parallel
with the optical table top. After the AO modulator was
inserted, it was adjusted for maximum power in the modulated
beam. The colimation lens was adjusted to produce a constant
separation of the modulation and zero order beams over the
length of the optical table (distance of about three feet) .
The blocking aperture was then adjusted to allow unobstructed
passage of the modulated beam and total blockage of the zero
order beam. The 50/50 and 5/95 splitters and steering mirror
were positioned such that the output beam struck each as
closely as possible to the center. The He-Ne laser was
adjusted for coincidence of the He-Ne and CO? beams at a
distance of approximately 40 feet from the steering mirror.
For CO? beam powers of approximately .25 watts or greater
the beam was sensed with temperature-sensitive chart paper
(see Figure 17) . For CO beam powers of a few milliwatts
to approximately .25 watts, the beam was sensed with
temperature-sensitive liquid crystal paper. The liquid
crystal paper was able to detect approximately one milliwatt
at the focal plane of the 8" focal length lens. The 5/95
splitter in the receive path was placed to allow maximum
transmission of the received signal while simultaneously
reflecting the local oscillator beam. These two beams were
adjusted to be as coaxial as possible at the exit surface
49
of the 5/95 splitter. The focusing lens was positioned
such that both beams were concentric with the lens center.
For precise alignment between the signal and local oscilla-
tor beams a folding mirror was placed approximately 20 feet
from the folding mirror, and it was adjusted to precisely
fold the visible beam back upon itself. The -50/50 splitter
and the 5/95 splitter in the receive path were then adjusted
to provide coincidence of the signal and local oscillator
beams at the focal point of the detector focus lens. As
a final precision adjustment a pinhole aperture (approxi-
mately .01 inch diameter) was placed at the focal point of
the lens and the two beams were adjusted to pass through
the aperture. Every time this procedure was followed,
there was sufficient beam overlap to produce heterodyne
operation. Once heterodyne operation was achieved, fine
adjustment of the beam splitters to produce a maximum
signal-to-noise ratio was easily achieved. The detector
was placed at the lens focal point by inserting a "whisper"
fan just prior to the detector. The detector was then
raster scanned until the low frequency (approximately 100 Hz)
"chopped" signal was observed on an oscilloscope. The
detector was then positioned to produce a maximum chopped
signal .
b. Second Optical Configuration
Refer to Figure 15 for this discussion. The
alignment procedure through the modulator was the some as
for the first configuration. The transmit beam elevator
50
was adjusted to place the transmit beam in the center of
the front face of the elliptical obstruction mirror of the
Newtonian antenna. The He-Ne and C02 beams were adjusted to
be coaxial as in the first configuration. A two-inch
diameter gold surface retro-reflector was placed on the top
of Ingersoll Hall. With the 5/95 splitter removed, the
visible return from the retro-reflector could be seen on
the dewar window. By chopping the returned signal, the
detector could relatively easily be placed at the antenna
focal spot. Once the chopped signal was located, the 5/95
splitter was set in place. The detector would then have
to be moved horizontally to compensate for the beam offset
caused by the beam splitter. This was always accomplished
with relative ease. A small folding mirror was placed
after the modulator and its height was adjusted to fold the
zero order beam while passing the modulated beam. The zero-
order beam was folded parallel to the transmit beam and
adjusted in height and direction to strike the center of the
5/95 splitter. The 5/95 splitter was then adjusted to pro-
duce a maximum chopped local oscillator signal at the
detector. Since the local oscillator beam was unfocused,
this task was relatively simple.
C. EXPERIMENTAL RESULTS
The aforementioned systems were constructed and aligned
in Spanagel 704. This was a seventh-story location with
open-window access to both land and sea targets. Initial
51
tests were conducted over path lengths of several meters
to 70 meters on the roof of Spanagel Hall with the aid of
several folding mirrors. The majority of the tests were
conducted several times, and in all cases the results were
highly reproducible. Initial feasibility tests of the first
optical configuration, less the processing electronics, were
conducted during the author's experience tour at NELC
(Code 2500) during May 1974. Subsequent testing was over
the period of July to December 1974.
1 . Zero Range Error
Two distinct and related effects will be discussed
in this section. During the initial radar feasibility tests
conducted with the optic configuration of Figure 14, a
40 MHz heterodyne signal was observed when the detector
was illuminated by the local oscillator alone. This is an
extremely troublesome effect for a CW radar xvith a stationary
target since a signal is present at the detector whether a
target return is present or not. It was postulated that
this effect was caused by a portion of the modulated beam
being reflected by the exit surface of the AO modulator and
entering the local oscillator beam by one of two mechanisms.
These mechanisms could be: reflections of the AO reflected
beam from the front face of the laser, or internal laser
cavity amplifications of the AO reflected beam which would
then appear superimposed on the local oscillator beam. The
following remedy verified the reflection source and eliminated
its effect. A quarter wave plate was inserted between the
52
90/10 splitter and the AO modulator. A wire grid polarizer
was inserted in the local oscillator path and adjusted to
the polarization plane of the local oscillator. The effect
of the quarter wave plate was to shift the polarization
of the reflected beam orthogonal to the incident beam. The
wire grid polarizer was then able to pass the true local
oscillator beam while blocking the reflected beam. Although
this remedy reduced an average heterodyne signal-to-noise
ratio of 40 db to 2-3 db and verified the source of the
initial reflections, it contributed no further evidence as
to the mechanism by which the reflections entered the local
oscillator beam.
When the system was moved to NPS and reconstructed,
no detectable local oscillator contaminating signal could
be detected for the first two weeks of laser operation.
During the interim period between initial tests at NELC
and the resumption of testing at NPS, the laser cavity had
been recharged to restore the laser power output to the
proper level. After a laser use period of about two weeks,
the local oscillator contaminating signal reappeared. Its
characteristics were somewhat altered from those originally
observed at NELC. The contaminating signal strength and
frequency response were a function of laser operating time.
In general, when the laser was first energized, the contami-
nating signal would be absent and would appear after about
five minutes of laser operation.
53
De-energizing the laser for a period of 10-15 minutes
before re-energizing resulted in fairly reproducible results.
Figures 22, 23, 24, and 25 show the amplitude and frequency
response of the contamination signal as a function of laser
energized time. Although these tests were by no means
conclusive, it was strongly suspected that the local
oscillator entry mechanism was via laser cavity amplifica-
tions since the laser cavity gain curve and frequency
response are a function of laser cavity temperature and
pressure«
The zero range offset was caused by the finite time
delay for the acoustic wave to propagate from the PZT
launch location to the optical beam waist location. This
delay varies as to the physical location of the optical
beam within the AO modulator optical input aperture.
However, once the optical system is adjusted, the delay
remains fixed, and it can easily be accounted for when
computing range and velocity information. Figure 26 shows
the system presentation with no signal return. The zero
range offset is visible via the previously discussed local
oscillator contaminating signal.
2 . Range Measurement
Initial range measurements of a stationary target
were performed on the roof of Spanagel Hall with folding
mirrors. The range to each target location was accurately
measured, and the resulting beat frequency was recorded.
54
During this measurement phase the zero offset range was not
clearly visible, and initial confusion concerning the
range measurements resulted. Since the short range specu-
lar reflector targets yielded a large signal return, the
Raytheon IR-101 detector was used in conjunction with the
optical configuration of Figure 14. No photographs of the
system presentation were taken during this measurement
phase; however, Table 1 and Figure 19 are the results of
these measurements. Figure 19 gives the value of the zero
offset range. Using the relationships of Chapter III,
Section A.l, the range information was easily computed by
Cf
R = 4F~5T"
m
where f = frequency shift observed minus the zero range
offset frequency.
The second phase of the stationary range measure-
ments was conducted at a range of approximately 305 yards.
The target was a 2" diameter, gold-surfaced retro-reflector.
The target was placed on the roof of Ingersoll Hall.
Figures 18 and 21 show the propagation path for this measure
ment phase. The optical configuration of Figure 15 in
conjunction with the Rockwell Internation number 5-128-4
detector (on loan from NELC) was used. Figure 26 clearly
shows the amount of zero range offset to be about 37 KHz.
Figure 27 shows the unprocessed detector output, and
Figure 28 shows the processed presentation. Note that
55
Figure 28 also shows the zero range offset at a much reduced
amplitude. Using the system parameters given below, the *
range was computed to be 295 yards.
Transmit Frequency Deviation 35-45 MHz
Modulation Frequency 1.75 KHz
Target Range 305 yards
Observed Range Frequency 100 KHz.
The retro-reflector was next moved to the edge of
El Estero Lake for range measurements. This range was
considerably beyond 300 yards. Although the exact range
was unknown, a visual estimation of the range was 1000-1200
yards. The retro-reflector was easily acquired, and a
S/N ratio of 35 db resulted. The corresponding measured
range frequency was 260 KHz including the zero range offset.
This computed to a range of 1055 yards. Since the received
signal power is a R relationship, the S/N ratio should
have decreased approximately 21 db . Allowing a several
additional db decrease for atmospheric transmission effects,
the results were well within expectations.
The longest range attempted was from Spanagel Hall
to the U.S. Coast Guard Pier. This range was approximately
2400 yards to the end of the pier. Target acquisition was
somewhat more difficult than for the El Estero Lake measure-
ments, and this was attributed to misalignment of the
visible and CC> transmit beams and the narrow receiver field
of view. The measured data were a S/N ratio of 20 db
and a range frequency of 540 KHz including the zero range
56
offset. This computed to a range of 2370 yards which was
in excellent agreement with the known value. The range
difference from 305 yards should have caused a S/N reduction
of approximately 36 db . If several db were allowed for
atmospheric losses, the results again were well within
expectations. Figures 29 and 30 show the spectrum analyzer
display of the signal return from the Coast Guard pier.
Figure 20 shows the local area map and the signal paths
used for the El Estero Lake and Coast Guard pier range
measurements .
3. Velocity Measurements
A velocity measurement experiment was conducted to
demonstrate that the target's relative velocity could be
measured, and that this measurement could be accomplished
simultaneously with the range measurement. To accomplish
this measurement, a model train was set up on the roof of
Ingersoll Hall. The track was oval with one of the straight
segments parallel to, and illuminated by, the C0? radar
beam. An electric timer in conjunction with a meter stick
was used for elapsed time velocity measurements. Both up
and down doppler informations were obtained by reversing
direction of the train. Figures 31 and 32 show the experi-
mental set-up used for these measurements. These figures
also show the target retro-reflector that was used for the
majority of the target informations. Due to the mechanical
nature of the apparatus, the velocity measurements were
57
rather imprecise. Due to elevation differences between the
transmitter and target locations, a correction factor of
cos 5.5° must be applied to compute the horizontal velocity
of the train. Figures 33, 34, 35 and 36 and Table 2 are
the results of this experiment. The figures clearly indi-
cate that simultaneous velocity and range measurements are
possible. The direction of velocity was immediately apparent,
and the computed magnitude was well within the experimental
error of the apparatus.
Although this experiment demonstrated the range and
velocity determination capabilities of the system as expected,
an unexpected phenomenon was observed with the optical
configuration of Figure 15. In addition to the expected
doppler shift of the two range frequencies, the stationary
target range frequencies were also present. They were,
however, at a reduced amplitude. The experiment was con-
ducted several times under varying conditions and essentially
the same results were obtained. Figures 31, 34, 35, and
36 show the simultaneous presence of both the stationary
target and moving target signals. The target was clearly
in motion while these photographs were taken; also, the
signal was definitely from the moving retro-reflector and
not from the train track or any other adjacent stationary
target. The experiment was essentially repeated at a later
date with the optical configuration of Figure 14. The
range and velocity results were as expected; however, the
58
stationary target frequencies were no longer present. No
cause for this unexpected phenomenon has been postulated.
An additional puzzle is why it should appear with one opti-
cal configuration and not with the other.
4 . Receiver Field of View
The receiver field of view measurements were con-
ducted at 305 yards using the retro-reflector. The
retro-reflector was positioned to maximize the returned
signal. The retro-reflector was then moved a known
distance and the resulting S/N ratio was recorded. Two
receiver fields of view were calculated: one at the point
where the S/N ratio decreased by 3 db and one where the
S/N ratio was reduced to unity. The results are given
below :
Optical Configuration of Figure 14
S/N reduced by 3 db S/N reduced to unity
Not Recorded 7.4m radian
Optical Configuration of Figure 15
S/N reduced by 3 db S/N reduced to unity
179 u radian 6.8 m radian
An additional rough measure of the receiver field
of view was conducted at the 1055 yard range. The arc-
length which resulted in a S/N ratio of unity was approxi-
mately one yard. This resulted in a field of view of
.95 m radian. A comparable decrease of 35 db S/N ratio
59
at 305 yards results in a field of view of .65 m radian.
Although all of these measurements are indicative of the
system performance, the most useful are most likely the
3 db figures.
The analysis of Chapter V, Section C.2 most closely
represents the physical system utilized. The 3 db field of
view of the optical configuration of Figure 15 is in
reasonable agreement with these results. A closer agree-
ment can be envisioned when the effect of the relatively
large size of the retro-reflector compared to an ideal point
source is considered. Also the retro-reflector was not
within the Fraunhofer or far field region of the optics.
By observing the chopped retro-return signal while
the retro-reflector position was moved, it appeared as
though the detector size was the primary field of view
limiting factor. As the target is moved in the object
plane, the Airy disk moves proportionately in the focal
plane. For a small detector, a small movement in the
object plane will cause the focused signal to depart from
the active detector area and detection ceases. The rela-
tionship describing this situation is
x
f
where x is the detector diameter and f is the optics focal
length. For both optical configurations, the detector
60
diameter was approximately seven mils. For the optical
configuration of Figure 15, the field of view predicted
by this relationship is 116 u radian, and for the optical
configuration of Figure 14, the field of view is predicted
as 875 u radian. Both of these predicted results remain
considerably smaller than the observed results. Again,
however, this prediction is predicated upon the target
being in the optics far field. It also assumes the signal
energy to be confined within the Airy disk. Due to the
considerably increased sensitivity of heterodyne detection,
the system should easily be capable of detecting energy
well out into the diffraction ring area. This may be the
cause of the larger than predicted field of view.
5 . Transmitter Divergence
In a laser radar system the transmitted beam
divergence is the counterpart of the microwave radar's
transmitter antenna gain. It must be fairly accurately
known before any meaningful system performance calculations
can be made. If the transmitted and received powers and
target and receiver apertures are known, the beam divergence
can be easily computed. The divergence calculations were
made with the optical configuration of Figure 15. The
retro-reflector was at a range of 305 yards.
The linear arc length of a beam of a given diver-
gence at a range R is
d = R9 .
61
The area covered by the beam at this range is
A /•(!-> 2 ,.R9 -. 2
The portion o£ the returned energy is the ratio of the area
of the retro-reflector to the area of the beam.
Dt 2
Pt * (2^}
Returned energy = ~ (33)
u Cj-0
If the retro reflector preserves the transmitted divergence
(a reasonable assumption) , then the received portion of
retransmitted energy will be
D 9
77 ^2~}
.Re. 2
The value of the received power will be
(34)
D. , D _
Pt * ( *) ^ 7T (/)2
P = — ^ . (35)
r ,R9.2 ,R6.2
tt {—) tt (— )
Solving for 9 results in
P. D ^ D 2
e = [p1 t 4 r ] (36)
r R
62
Using the below data, a transmitter divergence of 1.71
milliradians was calculated.
P = .65 watts
P = .75 milliwatts
r
D = 6 inches
r
D = 2 inches
R = 305 yards
The transmitted power was measured with the higher
power analog power meter while the received power was
measured with the lower power digital power meter. Accurate
results were predicated upon the absolute calibration of
each of these devices. A calibration error of either device
would, of course, be a source of error.
6. Detector Evaluation
Four detectors were used or evaluated during this
project. The first detector used was a Raytheon IR-101
PbSnTe photovoltaic detector. It was extremely useful for
alignment procedures since it exhibited a high quantum
efficiency and was capable of detecting powers up to 200 mw.
It was packaged in a glass dewar, however, and as such was
extremely vulnerable to external RFI . This proved to be
a considerable problem. Also the frequency response of the
detector was below the required value of 35-45 MHz. To
achieve a reasonable system S./N level, the modulator had
to be operated at a frequency below the desired linear
region of 35-45 MHz.
G3
The second detector used was a Rockwell International
5-128-4 PbSnTe photovoltaic detector. Its frequency response
was well beyond the required value, and although the quantum
efficiency appeared to be lower than several of the other
detectors, its system performance exceeded all others. The
received S/N ratio as a function of detector bias was
recorded for the optical configuration of Figure 14. As
can be seen by Table 3, a reverse bias beyond .2 volts
results in no further received S/N ratio.
The next detector evaluated was a Rockwell Inter-
national 7-229A. This detector exhibited a quantum
efficiency approximately twice that of the other Rockwell
detector, but the frequency response was again a problem.
A reverse bias of .075-. 6 volts was applied, but this was
not sufficient to raise the frequency response to the
required level. With the system frequencies lowered to
27-36 MHz, the best attainable S/N was 5 db below the
Rockwell 5-128-4 detector.
The last detector evaluated was an Aerojet General
PbSnTe, hetero junction, photovoltaic detector. It was a
high- impedance detector designed primarily for passive
system applications. Although the quantum efficiency
appeared reasonably high, the frequency response was much
too low to achieve heterodyne detection at the 35-45 MHz
region. A reverse bias of .075-. 5 volts was applied;
however, no heterodyne detection was observed.
64
7 . Receiver Sensitivity
The measure of the receiver sensitivity of a
radio- frequency radar is usually termed the minimum
discernible signal or MDS. For optical systems, however,
this parameter is more commonly expressed as the noise
equivalent power or NEP. It is defined as the signal
power necessary to produce a signal-to-noise ratio of unity.
This can be easily computed if the received power, effective
noise bandwidth, and measured system S/N ratio are known.
The actual received power incident upon the detector could
not be directly measured, but an estimation was obtained
from the measured receive power prior to the 5/95 splitter
and a knowledge of the optical parameters.
The Airy disk diameter for a far field target can
be computed by
d = 2.44 X (f/no)ef£ (37)
where the effective (f/no) for a Newtonian antenna was
computed from [Ref. 9]
1/2
(f/no)eff = 5
;> D - CD0BS/Dp)L|
(38)
For this relationship f is the focal length of the primary
mirror, D is the diameter of the primary mirror and D^p-
is the diameter of the obstructing mirror. The diameter
of the detector used was known to be approximately 8 mils.
6
r
The detector, therefore, was estimated to be 75% of the
Airy disk diameter. From the energy density profile of the
Airy disk [Ref . 4] , it was estimated that approximately 801
of the energy within the Airy disk was incident upon the
detector. Of the total incident energy approximately 841
of the energy is contained within the Airy disk [Ref. 9] .
The dewar window transmittance was estimated to be approxi-
mately .8. The far field of a lens system can be computed
from [Ref. 9]
D2
x = jk (39)
where D is the lens aperture diameter. The far field of
the optical configuration of Figure 15 was computed to be
1001 yards. With the retro-reflector at 305 yards, the
effective object range was 610 yards. It was assumed that
the difference between this range and the antenna's far
field would have little effect upon the computed Airy disk
size .
Using the above relationships and the experimental
conditions listed below, the system NEP for the optical
configuration of Figure 15 was calculated to be
NEP _ S 3.8x10 Watts „ , ,n-15UT ., Iu fArt,
~1T- HT7mTr = £ T " 7.6x10 Watts/Hz (40)
B (S/N)B (10G)(5xl04Rz)
Total Spectrum S/N 5 5 db
Range Frequency S/N 60 db
Transmitted Power - .65 w
66
Returned Signal Power .71 mw
Chopped Signal Output 5.5 mv
Chopped L.O. Output 1.4 mv
The returned signal power for the optical configura-
tion of Figure 14 was too low to be measured with the avail-
able power measuring equipments. Using the transmitted power,
target distance, beam divergence, and effective receiver
aperture, the signal power was computed to be 7.53 x 10
watts from the following relationship:
p = _J i 1 . (41)
r ,R6-.4 2
The factor — -r— is the effective receiver aperture. it is
the area of an ellipse with major and minor axes a and b,
respectively. The elliptical area is a result of the beam-
splitter's subtending a 45° angle to the received wavefront.
The factor .45 accounts for the two 5/95 and the 50/50
splitters within the receive path. Using the data listed
below, the system NEP was calculated to be
NEP 7.53xlO"6Watts - 01 m-l?.,, *< /u
— g— = 7 j = 3.01x10 Watts/Hz
* (5x10°) (5xl04Hz)
Range Frequency S/N 67 db
Transmitted Power .23 w
67
Returned Signal Power 7.53x10 w
(Calculated)
Local Oscillator Power 3 mw
From equation (9) it can be seen that for shot
noise limited heterodyne detection, the theoretical limit
of NEP is
^~- = 1.88xlO"20Watts/Hz .
The previously measured value of n by the Rockwell Interna-
tional Science Center was .32 [Ref. 37]; therefore, the
theoretical limit of NEP for the detector used was
NEP = 5.86xlO"20Watts/Hz
Therefore, it can be seen that the best attainable system
performance resulted in approximately three orders of
magnitude below the theoretical limit. The local oscillator
power of 3 mw for the optical configuration of Figure 14
should have been sufficient to produce shot noise limited
operation. Shot noise limited operation can be easily
verified by observing the background receiver noise as the
local oscillator power is increased. No such increase
was observed for either optical configuration. The
theoretical power required for shot noise limited operation
can be computed from [Ref. 2].
MCD . 2K (T + T' ) G- ,
NEP = £v {1 + ^m IFJ D (_hv,n (42)
B n l q ^ LO
68
where T is the physical detector temperature, T'Tp is
the amplifier effective noise temperature, and G~ is the
detector conductance.
For the system investigated, G~ = -r^r , n = .32,
T = 77°K and T'TC ~ 870°K (a generous estimate) which
m Lb
corresponds to an amplifier noise figure of 6. db .
For shot noise limited operation, the first term
of (42) must predominate the equation \\rhich requires
— « 2K (T + T' ) Gn (— )2 5
n *• m IF' D Kr\Q P
W 'L0
or
PTn >> 2K (T + T'J Gn (— ) (43)
LO m IF D ^nq
Using the specified system parameters given above, the re-
quired value of PT0 must be PJ0 >> 42 mw. Thus it can be
seen that a local oscillator power of 3 mw should induce
enough shot noise power to dominate other noise sources
With a S/N ratio of 60 db registered from the
retro-reflector at 305 yards, a diffuse reflector was
placed immediately in front of the reflector. No signal
return could be observed from the diffuse reflector. The
diffuse reflector consisted of a three-inch by six-inch
piece of sand-blasted aluminum. If the diffuse reflector
were a perfect Lambertian surface, the returned signal
irradiance (II) can be computed from
69
WA
H = — \ (44)
where WA is the diffuse reflector power returned and is
expressed by the relationship
PtAD
WAD = -7W72 <45>
77 Cj-)
where W is the radiant intensity and is the power density
returned per unit area of the reflector.
From these relationships it can be seen that
P.A^A
IT Co— J TTK
where A_ is the area of the diffuse reflector, and A is
the effective receiver aperture.
Comparing this result with equation (35) , it can be
seen that a reduction of 53.8 db could be expected from a
perfect Lambertian reflector of the size used. If the
reflection coefficient of the sandblasted diffuse target
is considered to be .79 [Ref. 9], then an additional reduc-
tion of 1 db would result. If the reflection coefficient
were as low as .1 due to oxidation, then the additional
decrease could be as much as 10 db . The actual reflection
coefficient encountered was expected to lie somewhere in
this mentioned range. An additional degradation of the
70
signal level was a masking effect caused by the strong local
oscillator contaminating signal. This signal could easily
have masked the small expected return from the diffuse
reflector.
71
VII. CONCLUSIONS AND RECOMMENDATIONS
A. CONCLUSIONS
The receiver sensitivities achieved were quite
disappointing. If the theoretical limit had been achieved,
an approximate sensitivity increase of 30 db would have
been realized with the optical configuration of Figure 14.
Also, if the optical configuration of Figure 15 had
realized the theoretical S/N improvement of approximately
15 db over that of Figure 14, its theoretical performance
should have been about 112 db S/N. With this degree of
system sensitivity, a three-inch by six-inch diffuse target
should be observable at a range of approximately 2500 yards.
It is doubtful that this degree of performance would be
acceptable for a search, tracking or threat warning system
for a Naval environment. It was suspected that a portion
of the less than expected receiver sensitivity problem was
caused by noise introduced by the signal processing elec-
tronics. Better filtering and lower noise amplifiers within
the processing area should improve the obtainable system
S/N. The system sensitivity is an inverse function of the
receiver effective noise bandwidth. If this bandwidth were
reduced considerably, the over-all system sensitivity could
be appreciably increased. The processed signal bandwidth
was also larger than desirable. This could have been
caused by frequency non-linearity of the modulator driver or
72
instabilities of the laser. Before the receiver noise
bandwidth can be reduced, the processed signal bandwidth
must be reduced accordingly.
From the range and velocity measurements, it was
clearly evident that the developmental system was capable
of simultaneous range and velocity measurements. The velocity
resolution measurements were quite impressive. The develop-
mental system would have been capable of detecting several
targets simultaneously providing the ranges or velocities
were appreciably different. The extremely narrow receiver
field of view would limit the system detection to a single
or a few targets.
B. AREAS OF POSSIBLE IMPROVEMENT AND FURTHER STUDY
From the receiver 3 db field of view and transmitter
divergence measurements, it was obvious that the trans-
mitted energy was inefficiently utilized. A better approach
would have been to use separate receive and transmit optics
with the transmitter divergence more nearly corresponding
to the receiver's effective field of view. A beam expander
within the transmitter optics could easily perform this
function. Tiie local oscillator contaminating signal pre-
vented close range measurements; therefore, its generation
mechanism should be investigated and suppressed. Although
this effect was suppressed with the combination of a
quarter-wave plate and a wire-grid polarizer, the transmitter
power was also reduced by more than 3 db. By eliminating
73
this effect through its generation mechanism, a larger
transmitted power can be achieved. Once this generation
mechanism is clearly understood, it is conceivable that it
could be optimized and utilized for a variety of system
applications. As an example, it could possibly be utilized
for a short-range data link type communication system.
The full signal processing advantage of a coherent detec-
tion process would be available while the receiver would be
greatly simplified since a local oscillator laser would
not be necessary. Since sufficient local oscillator power
would not be available to produce shot noise limited
operation, the system sensitivity would be considerably
reduced. The degradations of atmospheric amplitude scintil-
lations could easily be overcome by FM or PM modulation
methods. Also, the doppler shift from a rapidly moving
receiver or transmitter would not be a problem since the
modulated and reference signals would be doppler shifted by
the same amount. Also, it is not beyond reason to visualize
a radar system similar to the developmental system in which
a low power AO modulator is utilized to reflect power back
into the cavity of a high-power laser for re-amplification.
Such a system could possibly achieve transmitted powers
several orders of magnitude above the powers achieved with
the developmental system.
A study of alternate methods of signal processing or
modulation could be conducted to determine an optimum
technique.
74
Possible digital coding modulation schemes could be
used that would not require a high degree of frequency
linearity of the modulator.
An automatic range and velocity display system could
easily be devised using a digital counter and logic circuits,
Lastly, an automatic tracking system would most likely
be required within the optics to accommodate the extremely
small receiver field of view and stringent heterodyne detec-
tion requirements of the received wavefront.
C. POSSIBLE SYSTEM APPLICATIONS
One of the primary applications of an FM-CW radar
would be a precision tracking system capable of near the
horizon operation. If the .system sensitivity were adequate,
this aspect should be easily achievable due to the extremely
narrow transmitted beam and receiver field of view. The
excellent velocity resolution could easily allow a threat
velocity search to be conducted and identified. Also due
to the excellent velocity resolution, it is conceivable
that target vibrations could be sensed and categorized for
target identification.
Another application could be an airborne clear air
turbulence indicator. It could provide the turbulence
range and velocity information. A correlation of the
signal intensity with the measured range could possibly
yield information concerning the magnitude of index of
75
refraction change. This should yield information regarding
the degree of turbulence expected.
If the system sensitivity could not be increased to
effectively perform the above applications, an optical
return augmentation device might be used for cooperative
targets. Such an application might be for aircraft landing
systems in which it is highly desirable to precisely measure
the landing aircraft's velocity, range and descent angle.
76
TABLE 1
OBSERVED FREQUENCY AS A FUNCTION OF RANGE
Triangular Modulation
Af = 10 MHz
.fm = 10 KHz
OBS FREQ = Observed Frequency
CORR FREQ = Corrected Frequency
COMP FREQ = Computed Frequency
RANGE
(FEET)
OBS FREQ
(KHz)
CORR FREQ
(KHz)
COMP FREQ
(KHz)
43
345
15
17.5
67
360
30
27.2
91
567
37
37.0
115
375
45
46.8
139
387.5
57.5
56.3
163
395
65
66.3
187
405
75
76.0
211
415
85
85.8
Average Frequency Error: 1.43 KHz
Average Range Error: 3.5 Feet
77
TABLE 2
DOPPLER FREQUENCY MEASUREMENTS
Figure 33
1
v
Expected Frequency Shift
Observed Frequency Shift
Figure 34
1
v
Expected Frequency Shift
Observed Frequency Shift
Figure 35
1
v
Expected Frequency Shift
Observed Frequency Shift
Figure 36
1
v
Expected Frequency Shift
Observed Frequency Shift
Down Doppler Shift
~ 2.7 sec/m
~ .37 m/sec
69.5 KHz
8 0 KHz
Down Doppler Shift
~ 1.8 sec/m
~ .56 m/sec
104 KHz
110 KHz
Up Doppler Shift
~ 1.6 sec/m
~ .63 m/sec
117 KHz
120 KHz
Up Doppler Shift
~ 1.7 sec/m
~ .59 m/sec
110 KHz
115 KHz
78
TABLE 3
S/N AS A FUNCTION OF DETECTOR BIAS
Optical Configuration of Figure 14
Rockwell International Detector 5-128-4
S/N Reverse Bias
60 db .075 v
6 2 db .1 v
65 db .15 v
67 db .2 v
67 db .25 v
67 db .3 v
67 db .35 v
67 db .4 v
67 db .5 v
PLQ = 3 mw
Range = 305 Yards
79
TIME
Figure 1 .
Linear frequency modulation, stationary target,
(a) Frequency variation of signals.
(b) Observed beat frequency of a single stationary target
(a)
TIME
Li
5h
U
m
^>
h cy
< w
Figure 2.
near frequency modulation, moving target,
(a) Frequency variation of signals.
Cb) Observed beat frequency of a single moving target
(b)
TIME
The above figures indicate the relationship between the
transmitted and received signals of a linear FM homodyne
radar. The resulting beat frequency contains the target's
range and velocitv information.
SO
VELOCITY
MODE
TIME
Figure 3.
Dual mode operation of the system.
The CW mode could be used for precision velocity measurement
or possible target identification. The FM mode could be used
to measure the range information.
Figure 4.
Incident and resulting wave numbers of the optical and
acoustic energies.
81
LASER
OPTIC INPUT BEAM
FREQUENCY
MODULATED
ACOUSTIC
ABSORBER
ZERO-ORDER
DIFFRACTED BEAM
^ORDF.R
DIFFRACTEI
BEAM
CTED
ACOUSTIC
TRANSDUCER
MODULATOR
DRIVER
INFORMATION SIGNAL
Figure 5.
Basic acousto-optic modulator.
82
INCIDENT LASuR
BEAM
MODULATOR
DRIVER
ACOUSTIC
ABSORBER
ZERO-ORDER
.DIFFRACTED
(UNMODULATED)
-1 ORDER
D3
-ACOUSTIC TRANSDUCER
INFORMATION SIGNAL
FIGURE 6
>ragg angle acousto-optic modulator.
83
>-
o
■H
4->
03
o3
0)
03
bO
•H
4J
■P
rt
u
CO
12 3 4 5 6 7
Ratio of scattering center radius to wavelength, r/X
Figure 7.
Scattering area ratio after spherical water drops (Mie
scattering) .
84
10" 10" 10" 10'
Turbulence path length, L, meters
Figure 8.
Lateral phase coherence length for intermediate turbulence
85
V)
C
cd
•H
a
Sh
o
u
6
v
r-i
C
nj
c
o
•H
+->
cd
■H
>
cd
o
•H
+->
cd
•H
>
ID
tJ
T3
cd
cd
-(->
to
Beam diameter, dR, in meters
Figure 9.
Standard deviation in beam arrival angle due to intermediate
atmospheric turbulence.
86
LOCAL OSCILLATOR
WAVE FRONT
SIGNAL
VE FRONT
PHOTODIODE
SURFACE
Figure 10.
Spatial misalignment of local oscillator and signal beams
87
N
<
T3
03
DO
•H
£
(D
u
o
o
• H
u
o
•p
o
o
c
o
!-.
0
■P
<U
88
OUTPUT
SIGNAL
-O
INCIDENT
PHOTON
ELECTRICAL CONTACT
BULK CRYSTAL
Figure 12.
Simple Photo-conductor Circuit
89
2
O
0
i— i
Z
E-i
H
i
X <
i— i
<D
/ *J
S co
X
< ^
CD U
\
2 t-H
\,i
K Oh [
E- O
2 rt
^ W
!X M
H>h
U hJ
w <
&< 2
CD <
w
>co
W t-H
U H
uj a,
« o
6
ctf
f-.
U)
rt
•H
Q
to
U
i— 1
O
i— l
<d
CO
5-;
3
6
bO
0
•H
■P
Ph
tn
X
CO
u
•H
(/i
ctf
m
'a a,
CL,
<-
2
a,
o <£
m
vO
o
<
90
>
OS
pq
CM CO
o <
w
ts
J3
c^
H
o
rV
Pi
CO
w
OS
2:
Dh
h- 1
w
<
S
-J
3
o
•H
+->
C
o-
u
o
4->
(.0
S-
•H
PL,
9]
/-
oi
pj
(NI co
o <
U -J
.<*>
<
>
IT
I I
OS
o
w
w
<
w
W
s
N.j„_
ail
<
w
/K~
H
1
1
!S
w
CO
w
z
•r~i
J.
~K-^
c
O
• H
+-)
Cd
bO
•H
m
c
o
u
u
•H
-p
a,
o
Pi
o
u
<D
CO
92
9.5
9.0
u
■M
rH
O
>
P*
c
■H
S-c
>
•H
|h
O
8.5
8.0
/ . o
7.0
6.5
6.0
©
©
© L
©
©
©
©
©
©
©
©
©
©
©
©
©
©
©
J L
I I I I I I
35 37 39 41 43 45
Heterodyne frequency, MHz
Figure 16.
Modulator frequency linearity.
47 49
93
1 -
i
|
.
-
>&£.,-
"
, ;■,,, C?
-
...
<,:,-.
Figure 17.
Spatial intensity recording of modulated and zero-order
diffracted beams. Recording was taken after the modulator
colimating lens.
94
RETRO -
REFLECTOR
HALLIGAN
HALL
BULLARD
HALL
RADAR
BEAM
INGERSOLL HALL
ROOT HALL
305
Yards
SPANAGEL HALL
Figure 18
Map of NFS
95
0
a
u
o
o
"o
o
o
o
r^j
o
00
yo
«*
rsj
«*
«3
to
to
to
to
Z|(>{ 'Xouotibojj '4i-0{[ poAJOsqo
or.
RETRO-REFLECTOR AT END OF
U.S. COAST GUARD PIER
BREAKWATER
EL ESTERO
LAKE
TRANSMIT
BEAM
SPANAGEL HALL
Figure 20.
Map of Local Area.
97
Figure 21.
Photograph of beam path used for 305 yard range
measurements .
Figure 22.
Local oscillator contaminating signal 4 minutes after laser
was re-energized: Center frequency 40 MHz, Spectrum analyzer
dispersion 2 MHz/Div, Right side of figure is low frequency.
98
Figure 23.
Local oscillator contaminating signal 5 minutes after laser
was re-energized. All measurement parameters are as in
Figure 22.
Figure 24.
Local oscillator contaminating signal 7 minutes after laser
was re-energized. All measurement parameters are as in
Figure 22.
99
Figure 25.
Local oscillator contaminating signal 2 minutes after laser
was re-energized. All measurement parameters are as in
Figure 22.
Figure 26.
Zero range offset. The range offset is visible via the local
oscillator contaminating signal: Center frequency 30 MHz,
Spectrum analyzer dispersion 50 kHz/Div.
100
Figure 27.
Heterodyne detected signal prior to electronic processing
Center frequency 40 MHz, Spectrum analyzer dispersion
2 MHz/Div.
Figure 28.
Processed signal of a stationary target at 305 yards. 40 db
of attenuation is inserted prior to the spectrum analyzer.
Note the visible zero range offset. Center frequency 30 Mhz,
Spectrum Analyzer dispersion 50 kHz/Div.
101
Figure 29.
Processed signal from a stationary target at 2400 yards.
No attenuation inserted. Note that the zero range offset is
clearly visible: Center frequency 30 MHz, Spectrum Analyzer
dispersion 200 kHz/Div.
Figure 30.
Processed signal from a stationary target at 2400 yards,
All measurement information is the same as in Figure 29
102
Figure 29.
Processed signal from a stationary target at 2400 yards.
No attenuation inserted. Note that the zero range offset is
clearly visible: Center frequency 30 MHz, Spectrum Analyzer
dispersion 200 kHz/Div.
Figure 30.
Processed signal from a stationary target at 2400 yards
All measurement information is the same as in Figure 29
102
>Vj,UfH>MMW",!t;.^ ■:..;': ■
Figure 31.
Target velocity generation apparatus.
Figure 32.
Target retro-reflector,
103
Figure 33.
Simultaneous range and velocity information from a moving
target at 305 yards: Center frequency 30 MHz, Spectrum
analyzer dispersion 50 kHz/Div.
Figure 34.
Simultaneous range and velocity information from a moving
target at 305 yards. Note the presence of a stationary
target return. All measurement data are the same as for
Figure 33.
104
Figure 35.
Simultaneous range and velocity information from a moving
target at 305 yards: Measurement data are the same as
for Figure 33.
Figure 36.
Simultaneous range and velocity information from a moving
target at 305 yards: Measurement data are the same as for
Figure 33.
105
Figure 37.
Photograph of optical components utilized in the first
optical configuration (Figure 14) .
Figure 38.
Photograph of the optical components utilized in the second
optical configuration (Figure 15).
106
APPENDIX A
EQUIPMENT LIST
Unidek Optical Table, Quarter inch tapped holes drilled on
a one inch grid spacing.
CO- Laser, Honeywell Model 3000.
Modulator-Driver, Isomet Model DE-IR-10/S.
Acousto-Optic Modulator, Isomet Model DE-IR-10.
Spectrum Analyzer, Tektronix, Inc., Model 491.
Oscilloscope, Tektronix, Inc., Model 546.
Detector Bias and Pre-amplifier , Mfg. by NELC Code 2500.
(23 db Amplification with a 3 db Noise Figure)
Signal Amplifier (2-100 MHz), Miteq Model AV-1A-3078-3 .
Wideband Amplifier. Two Cascaded Hewlett Packard Model 461A.
Function Generator, Wavetek Model 134.
Signal Generator, Hewlett Packard Model 606A.
Analog Optical Power Meter, Coherent Radiation Model 201.
Digital Optical Power Meter, Jordon Model PM-550.
Balanced Modulators (Mixers - .2-500 MHz), Relcom Model Ml.
DC Power Supplies, Hewlett Packard Model 6216A.
Beamsplitters, Laser Optics 2 Inch Diameter Germanium
Splitters with Anti-reflective Coating on the Exit Surface.
Detector Focus Lens, Laser Optics 2 Inch Diameter Germanium
Lens with Anti- reflective Coatings.
Beam Elevators, Spectra-Physics Model 340.
Band Pass Filter (BPF) , 65-80 MHz Butterworth Band Pass
Filter Designed and Constructed by the Author.
High Pass Filter (HPF) , Simple Parallel LC filter with a Center
Frequency of 4 0 MHz. Constructed by the Author.
107
Infrared Detector, Raytheon Model IR-101, Sensitive area
approximately 10 Mils diameter.
Infrared Detector, Rockwell International Number 5-128-4,
Sensitive area approximately 7 Mils diameter.
Infrared Detector, Rockwell International Number 7-229A,
Sensitive area approximately 7 Mils diameter.
Infrared Detector, Aerojet General Corp. Sample number
unknown. Sensitive area approximately 5 Mils ■ diameter .
Encapsulated Liquid Crystal Paper, Edmund Scientific Co.
Stock Number 500224.
108
BIBLIOGRAPHY
1. Oliver, B. M. , "Signal to Noise Ratios in Photoelectric
Mixing," IRE Proceeding, v. 49, p. 1969-1961,
December 1961 .
2. Peyton, B. J., and others, "High-Sensitivity Receiver for
Infrared Laser Communications," IEEE Journal of
Quantum Electronics, v. QE-8, p. 252-263, February
1972.
3. Ross, M. , Laser Receivers, Wiley, 1966.
4. Pratt, W. K. , Laser Communications Systems, Wiley, 1969.
5. Peyton, B., and others, "Coherent Infrared Receivers for
Laser Communications and Radar," Internal paper of
AIL, a division of Cutter-Hammer Melville, Long Island,
New York, 1973.
6. Arams, F. R. , Sard, E. W. , and Peyton, B. J., "5.2 -
Infrared 10.6 - Micron Heterodyne Detection with
Gigahertz IF Capacili ty ," IEEE Journal of Quantum
Electronics, v. QE-3, p. 484-492, November 196T.~
7. Chance, T. H. , FM-CW Laser Radar at 10.6 Microns, Electrical
Engineer Thesis , Naval Postgraduate School, Monterey,
California, 1974.
8. Jamieson, J. A. and others, Infrared Physics and Engineering ,
McGraw-Hill, 1963.
9. Hudson, R. D., Infrared System Engineering, Wiley, 1969.
10. Soze, S. M. , Physics of Semiconductor Devices, Wiley,
1969.
11. Melchior, II., Fisher, M. B., and Arams, F. R. ,
"Photodetectors for Optical Communications Systems,"
Proceedings of the IEEE, v. 58, p. 1466-1485,
October 1970.
12. Holmquist, K. E., Thin- Film Pb„ nSn0 i Se Photoconductive
Infrared Detectors: Photoconductivity Measurements ,
Masters Thesis, Naval Postgraduate School, Monterey,
California, 1972.
13. Santa Barbara Research Center Contract DAAK02- 73-C- 0147 ,
High Density Lead Tin Tclluridc Detector Array, by
F.J. Renda , February 19 74.
109
14. Raytheon Report SM-313, Second Interim Report Development
of Lead-Tin Telluride Detectors , by Debye, P. P . ,
Kennedy , C . A. , and Linden, K. J. , January 1972.
15. Burke, C, and Koehler, T., "10.6 ym Detection at 170°K,"
paper presented at Sixth DOD Laser Conference,
Colorado Springs, Colorado, 26-28 March 1974.
16. DeLangue , 0. E., "Optical Heterodyne Detection,"
IEEE Spectrum, v. 5, p. 77-85, October 1968.
17. Skolnik, M. I., Radar Handbook, McGraw-Hill, 1970.
18. Hoisington, D. B., "Multiple-Target CW-FM Radar,"
paper presented at 6th Asilomar Conference on
Circuits and Systems, Pacific Grove, California,
November 1972.
19. Hymans , A. J., and Lait, J., "Analysis of a Frequency-
Modulated Continuous-Wave Ranging System,"
Proceedings IEE (British), Paper No. 3264E, p. 365-
372, July 1970.
20. Kay, L., "A Comparison between Pulse and Frequency-
Modulation Echo-Ranging Systems." Journal Brit. IRE,
p. 105-113, February 1959'.
21. Kaminow, I. P., and Turner, E. H., "Electrooptic Light
Modulators," Proceedings of the IEEE, v. 54,
■ p. 1374-1389, October 1966. "
22. Chen, F. S., "Modulators for Optical Communications,"
Proceedings of the IEEE, v. 58, October 1970.
23. Born, M. , and Wolf, E., Principles of Optics, 4th ed.,
Pergamon Press, 1970.
24. Gordon, E. I., "A Review of Acoustooptic Deflection
and Modulation Device," Proceedings of the IEEE,
v. 54, p. 1391-1401, October 1966.
25. Adler, R. , "Interaction between Light and Sound,"
IEEE Spectrum, v. 4, p. 42-54, May 1967.
26. Dixon, R. W. , and Gordon, E. I., "Acoustic Light
Modulators Using Optical Heterodyne Mixing," The
Bell System Technical Journal , v. 46, p. 367-389,
February 1967.
27. Yariv, A., Introduction to Optical Electronics, Holt,
Rinehart" and Winston ,"~1 971 .
110
28. Davis, J. I., "Considerations of Atmospheric Turbulence
in Laser System Design," Applied Optics , v. 5,
p. 139-147, January 1966.
29. Tatarski , V. I., Wave Propagation in a Turbulent Medium,
McGraw-Hill, 1961.
30. Fried, D. L., "Optical Resolution through a Randomly
Inhomogeneous Medium for Very Long and Very Short
Exposures," Journal of Optical Society of America,
v. 56, p. 13'/2-I3'/8, October 19bb. '
31. Fried, D. L., "Optical Heterodyne Detection of an
Atmospherically Distorted Signal Wavefront,"
Proceedings of the IEEE, v. 55, p. 57-67, January
TWT.
32. Gilmartin, T. J., and Holtz, J. Z., "Focused Beam and
Atmospheric Coherence Measurements at 10.6 urn and
0.63 urn," Applied Optics , v. 13, p. 1906-1912,
August 1974.
33. ITT Federal Laboratories, Measurement of the Degree of
Coherence for Laser Light Traversing Fogs, by
J. M. Grant and H. Y. Ageno , 9 October 1964.
34. Corcoran, V. J., "Directional Characteristics in Optical
Heterodyne Detection Processes," Journal of Applied
Physics, v. 36, p. 1819-1825, June 1965.
35. Siegman, A. E., "The Antenna Properties of Optical
Heterodyne Receivers," Proceedings of the IEEE,
v. 54, p. 1350-1356, October 1966.
36. Read, W. S., and Turner, R. G., "Tracking Heterodyne
Detection," Applied Optics , v. 4, p. 1570-1573,
December, 1965.
37. Rockwell International Science Center Report SC560-IM/DD,
10.6 Micron Photodiodes, by A. S. Joseph, p. 10,
27 September 1973.
Ill
INITIAL DISTRIBUTION LIST
No. Copies
1. Defense Documentation Center 2
Cameron Station
Alexandria, Virginia 22314
2. Library, Code 0212 2
Naval Postgraduate School
Monterey, California 93940
3. Department Chairman, Code 52 2
Department of Electrical Engineering
Naval Postgraduate School
Monterey, California 93940
4. Assoc. Professor T. N. Tao, Code 52 Tv 1
Department of Electrical Engineering
Naval Postgraduate School
Monterey, California 93940
5. Asst. Professor J. P. Powers, Code 52 Po 1
Department of Electrical Engineering
Naval Postgraduate School
Monterey, California 93940
6. Assoc. Professor G. L. Sackman, Code 52 Sa 1
Department of Electrical Engineering
Naval Postgraduate School
Monterey, California 93940
7. Naval Electronics Laboratory Center 2
Code 2500
271 Catalina Boulevard
San Diego, California 92152
8. LT Thomas H. Chance 1
U.S. Naval Destroyer School, Class 6 Jan 75
Newport, Rhode Island 02840
9. LT Maurice F. Fraunfelder 1
1302 Putnam Avenue
Janesville, Wisconsin 53545
112
/
F7857 P Wm^m
jc.l Fr*"nfe1der 157375
Ah
Thesis
F78E7
c.l
157375
Fraunfelder
A heterodyne detection
FM-CVJ laser radar using
10.6 ^source.
thesF7857
A heterodyne detection FM-CW laser radar
3 2768 OOO 99879 3
DUDLEY KNOX LIBRARY